NAMIC Wiki - User contributions [en]

2014 Summer Project Week:Cortical Dysplasia Identification

2014-06-27T14:11:00Z

Murta: /* Project Description */

==Key Investigators==
* Luiz Murta
* Emylin Sousa
* Tina Kapur
* Ron Kikinis

==Project Description==
<div style="margin: 20px;">
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Objective</h3>
* Find, construct and validate tools to investigate the presence and location of the epileptogenic focus through analysis of: cortical thickness, texture patterns.
* Investigates image registration as computational tool to compare pre and post MRI in epilepsy surgery.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Approach, Plan</h3>
* Structure and tissue segmentation of cortex.
* Cortical thickness evaluation, grey matter texture analysis.
* Pre and post surgical MRI registration, localization of resection.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Progress</h3>
* Gray matter segmentation using expectation-maximization (EMS) algorithm has shown accuracy and robustness in cortical dysplasia data.
* Started understanding and adapting existing code (ARCTIC) to estimate thickness. (with the help of Clement Vachet and Marcel Prastawa)
* Scripted module for registration between pre- and post-surgical volumes and quadratic difference volume computation to highlight resection. (Almost done, thanks for help from Jim Miller and Steve Pieper)
</div>
</div>

2014 Summer Project Week:Cortical Dysplasia Identification

2014-06-27T14:01:21Z

Murta: /* Key Investigators */

==Key Investigators==
* Luiz Murta
* Emylin Sousa
* Tina Kapur
* Ron Kikinis

==Project Description==
<div style="margin: 20px;">
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Objective</h3>
* Find, construct and validate tools to investigate the presence and location of the epileptogenic focus through analysis of: cortical thickness, texture patterns.
* Investigates image registration as computational tool to compare pre and post MRI in epilepsy surgery.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Approach, Plan</h3>
* Structure and tissue segmentation of cortex.
* Cortical thickness evaluation, grey matter texture analysis.
* Pre and post surgical MRI registration, localization of resection.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Progress</h3>
* Gray matter segmentation using expectation-maximization (EMS) algorithm has shown accuracy and robustness in cortical dysplasia data.
* Started understanding and adapting existing code (ARCTIC) to estimate thickness. (with the help of Clement Vachet and Marcel Prastawa)
* Scripted module for registration between pre- and post-surgical volumes and quadratic difference volume computation to highlight resection. (Almost done, thanks for the help from Jim Miller and Steve Pieper)
</div>
</div>

2014 Summer Project Week:Cortical Dysplasia Identification

2014-06-27T14:00:59Z

Murta: /* Project Description */

==Key Investigators==
* Luiz Murta
* Emylin Souza
* Tina Kapur
* Ron Kikinis

==Project Description==
<div style="margin: 20px;">
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Objective</h3>
* Find, construct and validate tools to investigate the presence and location of the epileptogenic focus through analysis of: cortical thickness, texture patterns.
* Investigates image registration as computational tool to compare pre and post MRI in epilepsy surgery.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Approach, Plan</h3>
* Structure and tissue segmentation of cortex.
* Cortical thickness evaluation, grey matter texture analysis.
* Pre and post surgical MRI registration, localization of resection.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Progress</h3>
* Gray matter segmentation using expectation-maximization (EMS) algorithm has shown accuracy and robustness in cortical dysplasia data.
* Started understanding and adapting existing code (ARCTIC) to estimate thickness. (with the help of Clement Vachet and Marcel Prastawa)
* Scripted module for registration between pre- and post-surgical volumes and quadratic difference volume computation to highlight resection. (Almost done, thanks for the help from Jim Miller and Steve Pieper)
</div>
</div>

2014 Summer Project Week:Cortical Dysplasia Identification

2014-06-27T13:30:40Z

Murta: /* Project Description */

==Key Investigators==
* Luiz Murta
* Emylin Souza
* Tina Kapur
* Ron Kikinis

==Project Description==
<div style="margin: 20px;">
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Objective</h3>
* Find, construct and validate tools to investigate the presence and location of the epileptogenic focus through analysis of: cortical thickness, texture patterns.
* Investigates image registration as computational tool to compare pre and post MRI in epilepsy surgery.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Approach, Plan</h3>
* Structure and tissue segmentation of cortex.
* Cortical thickness evaluation, grey matter texture analysis.
* Pre and post surgical MRI registration, localization of resection.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Progress</h3>
* Gray matter segmentation using expectation-maximization (EMS) algorithm has shown accuracy and robustness in cortical dysplasia data.
* Started understanding and adapting existing code (ARCTIC) to estimate thickness. (with the help of Clement Vachet and Marcel Prastawa)
* Scripted module for registration between pre- and post-surgical volumes and quadratic difference volume computation to highlight resection. (almost done, thanks to help from Jim Miller and Steve Piepen)
</div>
</div>

2014 Summer Project Week:Cortical Dysplasia Identification

2014-06-27T06:33:37Z

Murta: /* Project Description */

==Key Investigators==
* Luiz Murta
* Emylin Souza
* Tina Kapur
* Ron Kikinis

==Project Description==
<div style="margin: 20px;">
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Objective</h3>
* Find, construct and validate tools to investigate the presence and location of the epileptogenic focus through analysis of: cortical thickness, texture patterns.
* Investigates image registration as computational tool to compare pre and post MRI in epilepsy surgery.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Approach, Plan</h3>
* Structure and tissue segmentation of cortex.
* Cortical thickness evaluation, grey matter texture analysis.
* Pre and post surgical MRI registration, localization of resection.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Progress</h3>
* Gray matter segmentation using expectation-maximization (EMS) algorithm has shown accuracy and robustness in cortical dysplasia data.
* Started understanding and adapting existing code (ARCTIC) to estimate thickness. (with the help of Clement Vachet and Marcel Prastawa and Marcel Prastawa)
* Scripted module for registration between pre- and post-surgical volumes and quadratic difference volume computation to highlight resection. (almost done, thanks to help from Jim Miller and Steve Piepen)
</div>
</div>

2014 Summer Project Week:Cortical Dysplasia Identification

2014-06-27T06:23:19Z

Murta: /* Project Description */

==Key Investigators==
* Luiz Murta
* Emylin Souza
* Tina Kapur
* Ron Kikinis

==Project Description==
<div style="margin: 20px;">
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Objective</h3>
* Find, construct and validate tools to investigate the presence and location of the epileptogenic focus through analysis of: cortical thickness, texture patterns.
* Investigates image registration as computational tool to compare pre and post MRI in epilepsy surgery.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Approach, Plan</h3>
* Structure and tissue segmentation of cortex.
* Cortical thickness evaluation, grey matter texture analysis.
* Pre and post surgical MRI registration, localization of resection.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Progress</h3>
* Gray matter segmentation using expectation-maximization (EMS) algorithm has shown accuracy and robustness in cortical dysplasia data.
* Started understanding and adapting existing code (ARCTIC) to estimate thickness. (with the of Clement Vachet and Marcel Prastawa and Marcel Prastawa)
* Scripted module for registration between pre- and post-surgical volumes and quadratic difference volume computation to highlight resection. (almost done, thanks to help from Jim Miller and Steve Piepen)
</div>
</div>

2014 Summer Project Week:Cortical Dysplasia Identification

2014-06-27T06:05:01Z

Murta: /* Project Description */

==Key Investigators==
* Luiz Murta
* Emylin Souza
* Tina Kapur
* Ron Kikinis

==Project Description==
<div style="margin: 20px;">
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Objective</h3>
* Find, construct and validate tools to investigate the presence and location of the epileptogenic focus through analysis of: cortical thickness, texture patterns.
* Investigates image registration as computational tool to compare pre and post MRI in epilepsy surgery.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Approach, Plan</h3>
* Structure and tissue segmentation of cortex.
* Cortical thickness evaluation, grey matter texture analysis.
* Pre and post surgical MRI registration, localization of resection.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Progress</h3>
* Gray matter segmentation using expectation-maximization (EMS) algorithm has shown accuracy and robustness in cortical dysplasia data.
* Started understanding and adapting existing code (ARCTIC) to estimate thickness.
* Scripted module for registration between pre- and post-surgical volumes and quadratic difference volume computation to highlight resection. (almost done, thanks to help from Jim Miller and Steve Piepen)
</div>
</div>

2014 Summer Project Week:Cortical Dysplasia Identification

2014-06-27T05:43:23Z

Murta: /* Project Description */

==Key Investigators==
* Luiz Murta
* Emylin Souza
* Tina Kapur
* Ron Kikinis

==Project Description==
<div style="margin: 20px;">
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Objective</h3>
* Find, construct and validate tools to investigate the presence and location of the epileptogenic focus through analysis of: cortical thickness, texture patterns.
* Investigates image registration as computational tool to compare pre and post MRI in epilepsy surgery.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Approach, Plan</h3>
* Structure and tissue segmentation of cortex.
* Cortical thickness evaluation, grey matter texture analysis.
* Pre and post surgical MRI registration, localization of resection.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Progress</h3>
* Gray matter segmentation using expectation-maximization (EMS) algorithm has shown accuracy and robustness in cortical dysplasia data.
* Scripted module for registration between pre- and post-surgical volumes and quadratic difference volume computation to highlight resection. (almost done, thanks to help from Jim Miller and Steve Piepen)
</div>
</div>

2014 Summer Project Week:Cortical Dysplasia Identification

2014-06-27T05:42:31Z

Murta: /* Project Description */

==Key Investigators==
* Luiz Murta
* Emylin Souza
* Tina Kapur
* Ron Kikinis

==Project Description==
<div style="margin: 20px;">
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Objective</h3>
* Find, construct and validate tools to investigate the presence and location of the epileptogenic focus through analysis of: cortical thickness, texture patterns.
* Investigates image registration as computational tool to compare pre and post MRI in epilepsy surgery.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Approach, Plan</h3>
* Structure and tissue segmentation of cortex.
* Cortical thickness evaluation, grey matter texture analysis.
* Pre and post surgical MRI registration, localization of resection.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Progress</h3>
* Gray matter segmentation using expectation-maximization (EMS) algorithm shown acuracy and robustness in cortical dysplasia data.
* Scripted module for registration between pre- and post-surgical volumes and quadratic difference volume computation to highlight resection. (almost done, thanks to help from Jim Miller and Steve Piepen)
</div>
</div>

2014 Summer Project Week:Cortical Dysplasia Identification

2014-06-23T14:24:59Z

Murta: Created page with ' ==Key Investigators== * Luiz Murta * Emylin Souza * Tina Kapur * Ron Kikinis ==Project Description== <div style="margin: 20px;"> <div style="width: 27%; float: left; padding-…'

2014 Summer Project Week

2014-06-23T14:16:51Z

Murta: /* Image-Guided Therapy */

__NOTOC__

[[image:PW-MIT2014.png|300px]]

Dates: June 23-27, 2014.

Location: MIT, Cambridge, MA.

==Agenda==

{|border="1"
|-style="background:#b0d5e6;color:#02186f"
!style="width:10%" |Time
!style="width:18%" |Monday, June 23
!style="width:18%" |Tuesday, June 24
!style="width:18%" |Wednesday, June 25
!style="width:18%" |Thursday, June 26
!style="width:18%" |Friday, June 27
|-
|
|bgcolor="#dbdbdb"|'''Project Presentations'''
|bgcolor="#6494ec"|'''NA-MIC Update Day'''
|
|bgcolor="#88aaae"|'''IGT Day'''
|bgcolor="#faedb6"|'''Reporting Day'''
|-
|bgcolor="#ffffdd"|'''8:30am'''
|
|bgcolor="#ffffaa"|Breakfast
|bgcolor="#ffffaa"|Breakfast
|bgcolor="#ffffaa"|Breakfast
|bgcolor="#ffffaa"|Breakfast
|-
|bgcolor="#ffffdd"|'''9am-12pm'''
|
|'''10-11:30am''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: DICOM|DICOM]] (Steve Pieper)
[[MIT_Project_Week_Rooms|Grier Room (Left)]]
|
'''9:00-10:30am''' [[2014_Tutorial_Contest|Tutorial Contest Presentations (Sonia Pujol)]] <br>
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
<br>----------------------------------------<br>
'''11am-12noon''' Breakout Session: [[2014_Project_Week_Breakout_Session: Slicer for users| Slicer for users]] (Ron Kikinis)
<br>
|'''10am-12pm: <font color="#4020ff">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: IGT Neuro|Image-Guided Therapy - Neurosurgery]] (Alexandra Golby, Tina Kapur) <br>
[[MIT_Project_Week_Rooms#Star|Star]]

<br>----------------------------------------<br>
'''12pm''' [[Events:TutorialContestJune2014|Tutorial Contest Winner Announcement]]
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
|'''10am-12pm:''' [[#Projects|Project Progress Updates]]
|-
|bgcolor="#ffffdd"|'''12pm-1pm'''
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch boxes; Adjourn by 1:30pm
|-
|bgcolor="#ffffdd"|'''1pm-5:30pm'''
|'''1-1:05pm: <font color="#503020">Ron Kikinis: Welcome</font>'''
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
<br>----------------------------------------<br>
'''1:05-3:30pm:''' [[#Projects|Project Introductions]] (all Project Leads)
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
<br>----------------------------------------<br>
'''3:30-4:30pm''' [[2014 Summer Project Week Breakout Session:SlicerExtensions|Slicer4 Extensions]] (Jean-Christophe Fillion-Robin) <br>
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Room (Left)]]
|'''1-3pm:''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: QIICR|QIICR]] (Andrey Fedorov)
[[MIT_Project_Week_Rooms#Kiva|Kiva]]
|'''1-2:30pm:''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: Contours|Contours]] (Adam Rankin, Csaba Pinter)
[[MIT_Project_Week_Rooms#Kiva|Kiva]]
|'''1-3pm:''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: IGT Prostate|Image-Guided Therapy - Prostate Interventions]] (Clare Tempany, Noby Hata)
[[MIT_Project_Week_Rooms#Star|Star]]
<br>----------------------------------------<br>
|
|-
|bgcolor="#ffffdd"|'''5:30pm'''
|bgcolor="#f0e68b"|Adjourn for the day
|bgcolor="#f0e68b"|Adjourn for the day
|bgcolor="#f0e68b"|Adjourn for the day
|bgcolor="#f0e68b"|Adjourn for the day
|
|}

== '''Background''' ==

Project Week is a hands on activity -- programming using the open source [[NA-MIC-Kit|NA-MIC Kit]], algorithm design, and clinical application -- that has become one of the major events in the NA-MIC, NCIGT, and NAC calendars. It is held in the summer at MIT, typically the last week of June, and a shorter version is held in Salt Lake City in the winter, typically the second week of January.

Active preparation begins 6-8 weeks prior to the meeting, when a kick-off teleconference is hosted by the NA-MIC Engineering, Dissemination, and Leadership teams, the primary hosts of this event. Invitations to this call are sent to all NA-MIC members, past attendees of the event, as well as any parties who have expressed an interest in working with NA-MIC. The main goal of the kick-off call is to get an idea of which groups/projects will be active at the upcoming event, and to ensure that there is sufficient NA-MIC coverage for all. Subsequent teleconferences allow the hosts to finalize the project teams, consolidate any common components, and identify topics that should be discussed in breakout sessions. In the final days leading upto the meeting, all project teams are asked to fill in a template page on this wiki that describes the objectives and plan of their projects.

The event itself starts off with a short presentation by each project team, driven using their previously created description, and allows all participants to be acquainted with others who are doing similar work. In the rest of the week, about half the time is spent in breakout discussions on topics of common interest of subsets of the attendees, and the other half is spent in project teams, doing hands-on programming, algorithm design, or clinical application of NA-MIC kit tools. The hands-on activities are done in 10-20 small teams of size 3-5, each with a mix of experts in NA-MIC kit software, algorithms, and clinical. To facilitate this work, a large room is setup with several tables, with internet and power access, and each team gathers on a table with their individual laptops, connects to the internet to download their software and data, and is able to work on their projects. On the last day of the event, a closing presentation session is held in which each project team presents a summary of what they accomplished during the week.

A summary of all past NA-MIC Project Events is available [[Project_Events#Past|here]].

Please make sure that you are on the [http://public.kitware.com/mailman/listinfo/na-mic-project-week na-mic-project-week mailing list]

=Projects=
* [[2014_Project_Week_Template | Template for project pages]]

==Image-Guided Therapy==

* [[2014_Summer_Project_Week:SlicerIGT|SlicerIGT extension: testing, tutorials, website]] (Tamas Ungi, Nobuhiko Hata)
* [[Gestural Point of Care Interface for IGT]] (Saskia, Franklin, Steve, Tobias)
*[[2014_Summer_Project_Week:MR-Ultrasound_Registration_for_Prostate_Interventions | MR-Ultrasound Registration for Prostate Interventions]] (Chenxi Zhang, Andriy Fedorov, Andras Lasso)
*[[2014_Summer_Project_Week:Surface_approximation_from_contour_points | Surface approximation from contour points]] (Chenxi Zhang, Csaba Pinter, Andrey Fedorov)
* [[2014_Summer_Project_Week:Intelligent_Steering | Steered image registration using intelligent interfaces for minimal user interaction]] (Marcel Prastawa, Jim Miller, Steve Pieper)
* [[2014_Summer_Project_Week:Image To Mesh Conversion for Brain MRI | Image To Mesh Conversion for Brain MRI]] (Fotis Drakopoulos, Yixun Liu, Andrey Fedorov, Ron Kikinis, Nikos Chrisochoides)
* [[2014_Summer_Project_Week:An ITK implementation of Physics-Based Non-Rigid Registration method for Brain Shift | An ITK implementation of Physics-Based Non-Rigid Registration method for Brain Shift]] (Fotis Drakopoulos, Yixun Liu, Andriy Kot, Andrey Fedorov, Olivier Clatz, Ron Kikinis, Nikos Chrisochoides)
* [[2014_Summer_Project_Week:Robot_Control_With_OpenIGTLink | Robot Control With OpenIGTLink]] ( Gregory Fischer(WPI), Nirav Patel(WPI), Nobuhiko Hata (BWH) )
* [[2014_Summer_Project_Week:Open_source_electromagnetic_trackers_usingOpenIGTLink| Open-source electromagnetic trackers using OpenIGTLink]] (Peter Traneus Anderson, Tina Kapur, Sonia Pujol)
*[[2014_Summer_Project_Week:Intraoperative_Registration_of_preoperative_CT_and_C-arm_CT_of_the_lung | Intraoperative Registration of preoperative CT and C-arm CT of the lung]] (Katharina Breininger, Jay Jagadeesan)
*[[2014_Summer_Project_Week:Image guided neuroendoscope | Making realistic clinical story board for image guided skull base endoscopic surgery]] (Keryn Palmer, Nobuhiko Hata)
*[[2014_Summer_Project_Week:Cortical_Dysplasia_Identification | Tools for Dysplasia Identification in Epilepsy]] (Luiz Murta; Emylin Souza; Tina Kapur; Ron Kikinis)

==Huntington's Disease==
*[[2014_Summer_Project_Week:FiberTractDispersion| Fiber Tract Dispersion and UKF Tractography]] (Peter Savadjiev, Yogesh Rathi, Hans Johnson, C-F Westin)

==Head and Neck Cancer==
*[[2014_Summer_Project_Week:Interactive_DIR| Interactive DIR]] (Greg Sharp, Ivan Kolesov, Allen Tannenbaum)
*[[2014_Summer_Project_Week:DIR_validation_tools| DIR validation tools]] (Greg Sharp, Ivan Kolesov, Allen Tannenbaum)
*[[2014_Summer_Project_Week:Upload_HN_data| Upload H&N data]] (Greg Sharp, Paolo Zaffino)
*[[2014_Summer_Project_Week:DIR_stop_and_restart| DIR stop and restart]] (Paolo Zaffino, Greg Sharp)
*[[2014_Summer_Project_Week:InteractiveRegistration| Interactive Registration]] (Ivan Kolesov, Greg Sharp, Allen Tannenbaum)
*[[2014_Summer_Project_Week:External Beam Planning| External Beam Planning]] (Kevin Wang, Greg Sharp, Maxime Desplanques)
*[[2014_Summer_Project_Week:Proton_pencil_beam| Proton pencil beam dose calculation]] (Maxime Desplanques, Kevin Wang, Greg Sharp)

==TBI==
*[[2014_Summer_Project_Week:TBI_Segmentation| Interactive segmentation for traumatic brain injury ]] (Bo Wang, Marcel Prastawa, Andrei Irimia, John D. Van Horn, Guido Gerig)

==Stroke==
*[[2014_Summer_Project_Week:Stroke-ImagingGenetics | Stroke Imaging Genetics]] (Adrian Dalca, Ramesh Sridharan, Polina Golland)
*[[2014_Summer_Project_Week:Stroke-SuperResolution | Stroke Super Resolution]] (Adrian Dalca, Ramesh Sridharan, Polina Golland)

==Chronic Obstructive Pulmonary Disease, Lung, Chest ==

*[[2014_Summer_Project_Week: Pectoralis muscle segmentation| Pectoralis muscle segmentation]] (Rola Harmouche, James Ross, Raul San Jose)
*[[2014_Summer_Project_Week:Image_Registration_with_Sliding_Motion_Constraints | Image Registration with Sliding Motion Constraints]] (Alexander Derksen, Kanglin Chen, Gregory Sharp)
*[[2014_Summer_Project_Week:Multiscale_Non_Local_Means_filter_(NLM)_for_chest_CT_images | Multiscale Non Local Means filter (NLM) for chest CT images]] (Pietro Nardelli, Raul San Jose)

==[http://qiicr.org QIICR]==
* [[2014_Summer_Project_Week: RWV mapping support|Real world value mapping support]] (Andrey, Ethan, Andras, Steve, Jim)
* [[2014_Summer_Project_Week: CLI Derived DICOM Data| Proper formatting of DICOM Derived Data from CLI]] (Steve, Andrey, Jim, {Michael and David remotely})
* [[2014_Summer_Project_Week: DICOM SEG conversion to support archival of QIN Grand challenges results|DICOM SEG conversion to support archival of QIN Grand challenges results]] (Jayashree, Andrey, Steve, {David remotely})

==Feature Extraction==
*[[2014_Summer_Project_Week:Tumor_DCE-MRI_Segmentation | Breast Tumor Segmentation]] (Vivek Narayan, Jay Jagadeesan)
*[[2014_Summer_Project_Week:Tumor_Heterogeneity_Analysis | Breast Tumor Heterogeneity Analysis]] (Vivek Narayan, Jay Jagadeesan)
*Quantitative image feature extraction in Non-Small Cell Lung Cancer (Hugo Aerts)
*[[2014_Summer_Project_Week:Invariant_Feature_Extraction_Slicer | Invariant Feature Methods in Slicer]] (Matthew Toews, Nicole Aucoin, Sandy Wells)

==Brain==
*[[2014_Summer_Project_Week:Slicer_Murin_Shape_Analysis | Shape Analysis for the developing murine skull]] (Murat Maga, Ryan Young, Seattle Chidren's Hospital).
*[[2014_Summer_Project_Week:Slicer_LDDMM_Shape_Analysis | Slicer Interface to LDDMM shape anlaysis]] (Saurabh Jain, JHU; Steve Pieper, Isomics; Josh Cates, SCI, Utah; Hans Johnson, Iowa; Martin Styner, UNC)
*[[2014_Summer_Project_Week:Atlas Selection | Atlas Selection]] (Kanglin Chen, Gregory Sharp)
*[[2014_Summer_Project_Week:CAD_Toolbox_for_Neurological_Disorders | CAD Toolbox for Neurological Disorders]] (Sidong Liu, Siqi Liu, Fan Zhang, Yang Song, Weidong Cai, Sonia Pujol, Ron Kikinis)
*[[2014_Summer_Project_Week:Longitudinal_patient_specific_DTI_analysis | Longitudinal patient-specific DTI analysis using Slicer for neonatal asphyxia]] (Anuja Sharma, SCI, Utah; Francois Budin, UNC; Martin Styner, UNC; Guido Gerig, SCI, Utah)
*[[2014_Summer_Project_Week:mipiX | Rapid Visualization of Large Image Collections]] (Adrian, Ramesh, Polina)

==Slicer4 Extensions==

*[[2014_Summer_Project_Week:Multidim Data| Multidimensional Data]] (Kevin Wang, Andras Lasso)
*[[2014_Summer_Project_Week:DICOM-SRO import| DICOM-SRO import]] (Kevin Wang)
*[[2014_Summer_Project_Week:PLM_engineering| Plastimatch extension re-engineering]] (Greg Sharp, Paolo Zaffino, Andras, Csaba, Kevin)
*[[2014_Summer_Project_Week:DRAMMS_Slicer| Integrating DRAMMS deformable registration into Slicer]] (Yangming Ou, Steve Pieper, Andriy Fedorov, Tina Kapur, Christos Davatzikos, Ron Kikinis, Randy Gollub, Jayashree Kalpathy-Cramer)

==Infrastructure==
*Slicer 4.4 Release (JC, Steve, Nicole)
* [[2014_Summer_Project_Week: Chronicle| Chronicle]] (Steve)
* [[2014_Summer_Project_Week: Factory and Testing Process Post NA-MIC| Post NA-MIC Factory and Testing]] (Steve, Jc, Ron)
* [[2014_Summer_Project_Week: Volume Registration|Volume Registration]] (Steve, Greg, Marcel, Jim)
* [[2014_Summer_Project_Week:Markups | Markups]] (Nicole Aucoin)
*[[2014_Summer_Project_Week:Pluggable Label Statistics |Pluggable Label Statistics]] (Andrey , Ethan, Steve, Brad, Jim)
*[[2014_Summer_Project_Week:Subject_hierarchy_integration | Subject hierarchy integration]] (Csaba, Steve, Jc, Andras)
*[[2014_Summer_Project_Week:Contours | Contours]] (Adam Rankin, Csaba, Andras, Steve, Jc)
*[[2014_Summer_Project_Week:Parameter Node Serialization | Parameter Node Serialization]] (Kevin Wang, Andras, Steve, Jim, Csaba)
*[[2014_Summer_Project_Week:Self-tests for non-linear transforms | Self-tests for non-linear transforms]] (Xining Du)

== '''Logistics''' ==

*'''Dates:''' June 23-27, 2014.
*'''Location:''' [[MIT_Project_Week_Rooms| Stata Center / RLE MIT]].
*'''REGISTRATION:''' https://www.regonline.com/namic2014summerprojectweek. Please note that as you proceed to the checkout portion of the registration process, RegOnline will offer you a chance to opt into a free trial of ACTIVEAdvantage -- click on "No thanks" in order to finish your Project Week registration.
*'''Registration Fee:''' $300.
*'''Hotel:''' Similar to previous years, no rooms have been blocked in a particular hotel.
*'''Room sharing''': If interested, add your name to the list: [[2014_Summer_Project_Week/RoomSharing|here]]

== '''Registrants''' ==

Do not add your name to this list - it is maintained by the organizers based on your paid registration. ([https://www.regonline.com/namic2014summerprojectweek Please click here to register.])

#Hugo Aerts, Dana Farber/Harvard, hugo_aerts@dfci.harvard.edu
#Nassim Alikacem, Brigham & Women's Hospital, Nassim.Alikacem@gmail.com
#Peter Anderson, retired, traneus@verizon.net
#Nicole Aucoin, Brigham & Women's Hospital, nicole@bwh.harvard.edu
#Eva Breininger, Brigham & Women's Hospital, ebreininger@partners.org
#Francois Budin, NIRAL-UNC, fbudin@unc.edu
#Saskia Camps, SPL, saskiacamps@gmail.com
#Lucia Cevidanes, University of Michigan, luciacev@umich.edu
#Laurent Chauvin, SPL, lchauvin@bwh.harvard.edu
#Kanglin Chen, Fraunhofer MEVIS, kanglin.chen@mevis.fraunhofer.de
#Adrian Dalca, MIT CSAIL, adalca@mit.edu
#Alexander Derksen, Fraunhofer MEVIS, alexander.derksen@mevis.fraunhofer.de
#Maxime Desplanques, MGH/Politecnico di Milano, maxime.desplanques@cnao.it
#Fotis Drakopoulos, Old Dominion University, fdrakopo@gmail.com
#Sneha Durgapal, Brigham & Women's Hospital, durgapalsneha@gmail.com
#Andriy Fedorov, BWH, fedorov@bwh.harvard.edu
#Jean-Christophe Fillion-Robin, Kitware, jchris.fillionr@kitware.com
#James Fishbaugh, SCI Institute/University of Utah, jfishbaugh@gmail.com
#Jessica Forbes, University of Iowa, jessica-forbes@uiowa.edu
#Polina Golland, MIT CSAIL, polina@csail.mit.edu
#Jeffrey Grethe, University of CA San Diego, jgrethe@ncmir.ucsd.edu
#Rola Harmouche, Brigham & Women's Hospital, rolaharmouche@gmail.com
#Nobuhiko Hata, Brigham & Women's Hospital, hata@bwh.harvard.edu
#Saurabh Jain, Johns Hopkins University, saurabh@cis.jhu.edu
#Hans Johnson, University of Iowa, hans-johnson@uiowa.edu
#Jayashree Kalpathy-Cramer, MGH, kalpathy@nmr.mgh.harvard.edu
#Tina Kapur, BWH/Harvard Medical School, tkapur@bwh.harvard.edu
#Ron Kikinis, HMS, kikinis@bwh.harvard.edu
#Regina Kim, University of Iowa, eunyoung-kim@uiowa.edu
#Franklin King, Queen's University, franklin.king@queensu.ca
#Tassilo Klein, SPL/BWH, TJKlein@bwh.harvard.edu
#Farukh Kohistani, BWH Radiology, kohistan@bc.edu
#Robin Kouver, BWH/SPL, r.kouver@gmail.com
#Andreas Lasso, PerkLab - Queen's University, lasso@queensu.ca
#Yangming Li, University of Washington, ymli81@uw.edu
#Sidong Liu, SPL/BWH, sliu@bwh.harvard.edu
#Siqi Liu, University of Sydney, sliu4512@uni.sydney.edu.au
#Bradley Lowekamp, National Institutes of Health, blowekamp@mail.nih.gov
#Murat Maga, Seattle Children's Research Institute, maga@uw.edu
#Katie Mastrogiacomo, SPL/BWH, kmast@bwh.harvard.edu
#Alireza Mehrtash, SPL/BWH, mehrtash@bwh.harvard.edu
#Dominik Meier, Brigham & Women's Hospital, meier@bwh.harvard.edu
#Jim Miller, GE Research, millerjv@ge.com
#Luiz Otavio Murta Junor, SPL/BWH, lmurta@partners.org
#Vivek Narayan, NCIGT, narayan.vivek9@gmail.com
#Pietro Nardelli, University College Cork, pietro@bwh.harvard.edu
#Yangming Ou, MGH, yangming.ou@uphs.upenn.edu
#Danielle Pace, MIT CSAIL, dfpace@mit.edu
#Keryn Palmer, Brigham & Women's Hospital, kpalmer5@partners.org
#Nirav Patel, WPI, napatel@wpi.edu
#Tobias Penzkofer, SPL, pt@bwh.harvard.edu
#Steve Pieper, Isomics Inc, pieper@isomics.com
#Csaba Pinter, Queen's University, csaba.pinter@queensu.ca
#Marcel Prastawa, GE Research, marcel.prastawa@ge.com
#Somia Pujol, Harvard Medical School, spujol@bwh.harvard.edu
#Adam Rankin, Queen's University, rankin@queensu.ca
#Aymeric Reshef, Brigham & Women's Hospital, areshef@bwh.harvard.edu
#Rahul Sastry, BWH/SPL, rahul_sastry@hms.harvard.edu
#Peter Savadjiev, Brigham & Women's Hospital, petersv@bwh.harvard.edu
#Gregory Sharp, MGH, gcsharp@mgh.harvard.edu
#Emylin Sousa, BWH/SPL, emylin.sousa@gmail.com
#Ramesh Sridharan, MIT CSAIL, rameshvs@csail.mit.edu
#Matthew Toews, BWH/Harvard Medical School, mt@bwh.harvard.edu
#Ethan Ulrich, University of Iowa, ethan-ulrich@uiowa.edu
#Tamas Ungi, Queen's University, ungi@queensu.ca
#Kevin Wang, Princess Margaret Cancer Centre, kevin.wang@rmp.uhn.ca
#David Welch, University of Iowa, david-welch@uiowa.edu
#William Wells, Brigham & Women's Hospital, sw@bwh.harvard.edu
#Phillip White, BWH/Harvard Medical School, white@bwh.harvard.edu
#Alex Yarmarkovich, ISOMICS Inc., alexy@bwh.harvard.edu
#Ryan Young, Seattle Children's Research Institute, ryan.young@seattlechildrens.org
#Paolo Zaffino, University Magna Graecia of Catanzaro, p.zaffino@unicz.it
#Chenxi Zhang, Brigham & Women's Hospital, chenxizhang@fudan.edu.cn
#Fan Zhang, University of Sydney, fzha8048@uni.sydney.edu.au

2014 Summer Project Week

2014-06-22T17:20:42Z

Murta: /* Other */

__NOTOC__

[[image:PW-MIT2014.png|300px]]

Dates: June 23-27, 2014.

Location: MIT, Cambridge, MA.

==Agenda==

{|border="1"
|-style="background:#b0d5e6;color:#02186f"
!style="width:10%" |Time
!style="width:18%" |Monday, June 23
!style="width:18%" |Tuesday, June 24
!style="width:18%" |Wednesday, June 25
!style="width:18%" |Thursday, June 26
!style="width:18%" |Friday, June 27
|-
|
|bgcolor="#dbdbdb"|'''Project Presentations'''
|bgcolor="#6494ec"|'''NA-MIC Update Day'''
|
|bgcolor="#88aaae"|'''IGT Day'''
|bgcolor="#faedb6"|'''Reporting Day'''
|-
|bgcolor="#ffffdd"|'''8:30am'''
|
|bgcolor="#ffffaa"|Breakfast
|bgcolor="#ffffaa"|Breakfast
|bgcolor="#ffffaa"|Breakfast
|bgcolor="#ffffaa"|Breakfast
|-
|bgcolor="#ffffdd"|'''9am-12pm'''
|
|'''10-11:30pm''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: DICOM|DICOM]] (Steve Pieper)
[[MIT_Project_Week_Rooms|Grier Room (Left)]]
|
|
'''9:00-10:30am''' [[2014_Tutorial_Contest|Tutorial Contest Presentations (Sonia Pujol)]] <br>
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
<br>----------------------------------------<br>
'''10am-12pm: <font color="#4020ff">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: IGT Neuro|Image-Guided Therapy - Neurosurgery]] (Alexandra Golby, Tina Kapur) <br>
[[MIT_Project_Week_Rooms#Star|Star]]
|'''10am-12pm:''' [[#Projects|Project Progress Updates]] <br>
'''12pm''' [[Events:TutorialContestJune2014|Tutorial Contest Winner Announcement]]
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
|-
|bgcolor="#ffffdd"|'''12pm-1pm'''
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch boxes; Adjourn by 1:30pm
|-
|bgcolor="#ffffdd"|'''1pm-5:30pm'''
|'''1-1:05pm: <font color="#503020">Ron Kikinis: Welcome</font>'''
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
<br>----------------------------------------<br>
'''1:05-3:30pm:''' [[#Projects|Project Introductions]] (all Project Leads)
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
<br>----------------------------------------<br>
'''3:30-4:30pm''' [[2014 Summer Project Week Breakout Session:SlicerExtensions|Slicer4 Extensions]] (Jean-Christophe Fillion-Robin) <br>
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Room (Left)]]
|'''1-3pm:''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: QIICR|QIICR]] (Andrey Fedorov)
[[MIT_Project_Week_Rooms#Kiva|Kiva]]
|'''1-2:30pm:''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: Contours|Contours]] (Adam Rankin, Csaba Pinter)
[[MIT_Project_Week_Rooms#Kiva|Kiva]]
|'''1-3pm:''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: IGT Prostate|Image-Guided Therapy - Prostate Interventions]] (Clare Tempany, Noby Hata)
[[MIT_Project_Week_Rooms#Star|Star]]
<br>----------------------------------------<br>
|
|-
|bgcolor="#ffffdd"|'''5:30pm'''
|bgcolor="#f0e68b"|Adjourn for the day
|bgcolor="#f0e68b"|Adjourn for the day
|bgcolor="#f0e68b"|Adjourn for the day
|bgcolor="#f0e68b"|Adjourn for the day
|
|}

== '''Background''' ==

Project Week is a hands on activity -- programming using the open source [[NA-MIC-Kit|NA-MIC Kit]], algorithm design, and clinical application -- that has become one of the major events in the NA-MIC, NCIGT, and NAC calendars. It is held in the summer at MIT, typically the last week of June, and a shorter version is held in Salt Lake City in the winter, typically the second week of January.

Active preparation begins 6-8 weeks prior to the meeting, when a kick-off teleconference is hosted by the NA-MIC Engineering, Dissemination, and Leadership teams, the primary hosts of this event. Invitations to this call are sent to all NA-MIC members, past attendees of the event, as well as any parties who have expressed an interest in working with NA-MIC. The main goal of the kick-off call is to get an idea of which groups/projects will be active at the upcoming event, and to ensure that there is sufficient NA-MIC coverage for all. Subsequent teleconferences allow the hosts to finalize the project teams, consolidate any common components, and identify topics that should be discussed in breakout sessions. In the final days leading upto the meeting, all project teams are asked to fill in a template page on this wiki that describes the objectives and plan of their projects.

The event itself starts off with a short presentation by each project team, driven using their previously created description, and allows all participants to be acquainted with others who are doing similar work. In the rest of the week, about half the time is spent in breakout discussions on topics of common interest of subsets of the attendees, and the other half is spent in project teams, doing hands-on programming, algorithm design, or clinical application of NA-MIC kit tools. The hands-on activities are done in 10-20 small teams of size 3-5, each with a mix of experts in NA-MIC kit software, algorithms, and clinical. To facilitate this work, a large room is setup with several tables, with internet and power access, and each team gathers on a table with their individual laptops, connects to the internet to download their software and data, and is able to work on their projects. On the last day of the event, a closing presentation session is held in which each project team presents a summary of what they accomplished during the week.

A summary of all past NA-MIC Project Events is available [[Project_Events#Past|here]].

Please make sure that you are on the [http://public.kitware.com/mailman/listinfo/na-mic-project-week na-mic-project-week mailing list]

=Projects=
* [[2014_Project_Week_Template | Template for project pages]]

==TBI==
*[[2014_Summer_Project_Week:TBI_Segmentation| Interactive segmentation for traumatic brain injury ]] (Bo Wang, Marcel Prastawa, Andrei Irimia, John D. Van Horn, Guido Gerig)

==Atrial Fibrillation==

==Huntington's Disease==
*[[2014_Summer_Project_Week:FiberTractDispersion| Fiber Tract Dispersion and UKF Tractography]] (Peter Savadjiev, Yogesh Rathi, Hans Johnson, C-F Westin)

==Head and Neck Cancer==
*[[2014_Summer_Project_Week:Interactive_DIR| Interactive DIR]] (Greg Sharp, Ivan Kolesov, Allen Tannenbaum)
*[[2014_Summer_Project_Week:DIR_validation_tools| DIR validation tools]] (Greg Sharp, Ivan Kolesov, Allen Tannenbaum)
*[[2014_Summer_Project_Week:Upload_HN_data| Upload H&N data]] (Greg Sharp, Paolo Zaffino)
*[[2014_Summer_Project_Week:DIR_stop_and_restart| DIR stop and restart]] (Paolo Zaffino, Greg Sharp)

==Slicer4 Extensions==

*[[2014_Summer_Project_Week:Multidim Data| Multidim Data]] (Kevin Wang, Andras, ?)
*[[2014_Summer_Project_Week:DICOM-SRO import| DICOM-SRO import]] (Kevin Wang)
*[[2014_Summer_Project_Week:PLM_engineering| Plastimatch extension re-engineering]] (Greg Sharp, Paolo Zaffino, Andras, Csaba, Kevin)

==Cardiac==

==Stroke==
*[[2014_Summer_Project_Week:Stroke-ImagingGenetics | Stroke Imaging Genetics]] (Adrian Dalca, Ramesh Sridharan, Polina Golland)
*[[2014_Summer_Project_Week:Stroke-SuperResolution | Stroke Super Resolution]] (Adrian Dalca, Ramesh Sridharan, Polina Golland)

==Brain Segmentation==

==Image-Guided Therapy==

* SlicerIGT extension: testing, tutorials, website (Tamas Ungi, Nobuhiko Hata)
* [[Gestural Point of Care Interface for IGT]] (Saskia, Franklin, Steve, Tobias)
*[[2014_Summer_Project_Week:MR-Ultrasound_Registration_for_Prostate_Interventions | MR-Ultrasound Registration for Prostate Interventions]] (Chenxi Zhang, Andriy Fedorov, Andras)
*[[2014_Summer_Project_Week:Surface_approximation_from_contour_points | Surface approximation from contour points]] (Chenxi Zhang, Csaba Pinter, Andrey Fedorov)
* [[2014_Summer_Project_Week:Intelligent_Steering | Steered image registration using intelligent interfaces for minimal user interaction]] (Marcel Prastawa, Jim Miller, Steve Pieper)
* [[2014_Summer_Project_Week:Image To Mesh Conversion for Brain MRI | Image To Mesh Conversion for Brain MRI]] (Fotis Drakopoulos, Yixun Liu, Andrey Fedorov, Ron Kikinis, Nikos Chrisochoides)
* [[2014_Summer_Project_Week:An ITK implementation of Physics-Based Non-Rigid Registration method for Brain Shift | An ITK implementation of Physics-Based Non-Rigid Registration method for Brain Shift]] (Fotis Drakopoulos, Yixun Liu, Andriy Kot, Andrey Fedorov, Olivier Clatz, Ron Kikinis, Nikos Chrisochoides)
* [[2014_Summer_Project_Week:Robot_Control_With_OpenIGTLink | Robot Control With OpenIGTLink]] ( Gregory Fischer(WPI), Nirav Patel(WPI), Nobuhiko Hata (BWH) )
* [[2014_Summer_Project_Week:Open_source_electromagnetic_trackers_usingOpenIGTLink| Open-source electromagnetic trackers using OpenIGTLink]] (Peter Traneus Anderson, Tina Kapur, Sonia Pujol)

==Radiation Therapy==
*[[2014_Summer_Project_Week:External Beam Planning| External Beam Planning]] (Kevin Wang, Greg Sharp, Maxime Desplanques, ?)
*[[2014_Summer_Project_Week:Proton_pencil_beam| Proton pencil beam dose calculation]] (Maxime Desplanques, Kevin Wang, Greg Sharp)

==Chronic Obstructive Pulmonary Disease ==

*[[2014_Summer_Project_Week: Pectoralis muscle segmentation| Pectoralis muscle segmentation]] (Rola Harmouche, James Ross, Raul San Jose)

==[http://qiicr.org QIICR]==
* [[2014_Summer_Project_Week: RWV mapping support|Real world value mapping support]] (Andrey, Ethan, Andras, Steve, Jim, ...)
* [[2014_Summer_Project_Week: CLI Derived DICOM Data| Proper formatting of DICOM Derived Data from CLI]] (Steve, Andrey, Jim, {Michael and David remotely})
* [[2014_Summer_Project_Week: DICOM SEG conversion to support archival of QIN Grand challenges results|DICOM SEG conversion to support archival of QIN Grand challenges results]] (Jayashree, Andrey, Steve, {David remotely})

==Infrastructure==
*Slicer 4.4 Release (JC, Steve, Nicole)
* [[2014_Summer_Project_Week: Chronicle| Chronicle]] (Steve)
* [[2014_Summer_Project_Week: Volume Registration|Volume Registration]] (Steve, Greg, Marcel, Jim)
* [[2014_Summer_Project_Week:Markups | Markups]] (Nicole Aucoin)
*[[2014_Summer_Project_Week:Pluggable Label Statistics |Pluggable Label Statistics]] (Andrey , Ethan, Steve, Brad, Jim? Dirk?)
*[[2014_Summer_Project_Week:Subject_hierarchy_integration | Subject hierarchy integration]] (Csaba, Steve, Jc, Andras?, ?)
*[[2014_Summer_Project_Week:Contours | Contours]] (Adam Rankin, Csaba, Andras, Steve, Jc, ?)
*[[2014_Summer_Project_Week:Parameter Node Serialization | Parameter Node Serialization]] (Kevin Wang, Andras, Steve, Jim, Csaba, ?)
*[[2014_Summer_Project_Week:Self-tests for non-linear transforms | Self-tests for non-linear transforms]] (Xining Du)

==Feature Extraction==
*[[2014_Summer_Project_Week:Tumor_DCE-MRI_Segmentation | Breast Tumor Segmentation]] (Vivek Narayan, Jay Jagadeesan)
*[[2014_Summer_Project_Week:Tumor_Heterogeneity_Analysis | Breast Tumor Heterogeneity Analysis]] (Vivek Narayan, Jay Jagadeesan)
*Quantitative image feature extraction in Non-Small Cell Lung Cancer (Hugo Aerts)
*[[2014_Summer_Project_Week:Invariant_Feature_Extraction_Slicer | Invariant Feature Methods in Slicer]] (Matthew Toews, Nicole Aucoin, Sandy Wells)

==Other==
*[[2014_Summer_Project_Week:Slicer_Murin_Shape_Analysis | Shape Analysis for the developing murine skull]] (Murat Maga, Ryan Young, Seattle Chidren's Hospital).
*[[2014_Summer_Project_Week:Slicer_LDDMM_Shape_Analysis | Slicer Interface to LDDMM shape anlaysis]] (Saurabh Jain, JHU; Steve Pieper, Isomics; Josh Cates, SCI, Utah; Hans Johnson, Iowa; Martin Styner, UNC)
*[[2014_Summer_Project_Week:Image_Registration_with_Sliding_Motion_Constraints | Image Registration with Sliding Motion Constraints]] (Alexander Derksen, Kanglin Chen, Gregory Sharp)
*[[2014_Summer_Project_Week:Atlas Selection | Atlas Selection]] (Kanglin Chen, Gregory Sharp)
*[[2014_Summer_Project_Week:Multiscale_Non_Local_Means_filter_(NLM)_for_chest_CT_images | Multiscale Non Local Means filter (NLM) for chest CT images]] (Pietro Nardelli, University College Cork (UCC), Ireland)
*[[2014_Summer_Project_Week:Intraoperative_Registration_of_preoperative_CT_and_C-arm_CT_of_the_lung | Intraoperative Registration of preoperative CT and C-arm CT of the lung]] (Katharina Breininger, Jay Jagadeesan)
*[[2014_Summer_Project_Week:CAD_Toolbox_for_Neurological_Disorders | CAD Toolbox for Neurological Disorders]] (Sidong Liu, Siqi Liu, Fan Zhang, Yang Song, Weidong Cai, Sonia Pujol, Ron Kikinis)
*[[2014_Summer_Project_Week:Longitudinal_patient_specific_DTI_analysis | Longitudinal patient-specific DTI analysis using Slicer for neonatal asphyxia]] (Anuja Sharma, SCI, Utah; Francois Budin, UNC; Martin Styner, UNC; Guido Gerig, SCI, Utah)

== '''Logistics''' ==

*'''Dates:''' June 23-27, 2014.
*'''Location:''' [[MIT_Project_Week_Rooms| Stata Center / RLE MIT]].
*'''REGISTRATION:''' https://www.regonline.com/namic2014summerprojectweek. Please note that as you proceed to the checkout portion of the registration process, RegOnline will offer you a chance to opt into a free trial of ACTIVEAdvantage -- click on "No thanks" in order to finish your Project Week registration.
*'''Registration Fee:''' $300.
*'''Hotel:''' Similar to previous years, no rooms have been blocked in a particular hotel.
*'''Room sharing''': If interested, add your name to the list: [[2014_Summer_Project_Week/RoomSharing|here]]

== '''Registrants''' ==

Do not add your name to this list - it is maintained by the organizers based on your paid registration. ([https://www.regonline.com/namic2014summerprojectweek Please click here to register.])

#Hugo Aerts, Dana Farber/Harvard, hugo_aerts@dfci.harvard.edu
#Nassim Alikacem, Brigham & Women's Hospital, Nassim.Alikacem@gmail.com
#Peter Anderson, retired, traneus@verizon.net
#Nicole Aucoin, Brigham & Women's Hospital, nicole@bwh.harvard.edu
#Eva Breininger, Brigham & Women's Hospital, ebreininger@partners.org
#Francois Budin, NIRAL-UNC, fbudin@unc.edu
#Saskia Camps, SPL, saskiacamps@gmail.com
#Lucia Cevidanes, University of Michigan, luciacev@umich.edu
#Laurent Chauvin, SPL, lchauvin@bwh.harvard.edu
#Kanglin Chen, Fraunhofer MEVIS, kanglin.chen@mevis.fraunhofer.de
#Adrian Dalca, MIT CSAIL, adalca@mit.edu
#Alexander Derksen, Fraunhofer MEVIS, alexander.derksen@mevis.fraunhofer.de
#Maxime Desplanques, MGH/Politecnico di Milano, maxime.desplanques@cnao.it
#Fotis Drakopoulos, Old Dominion University, fdrakopo@gmail.com
#Sneha Durgapal, Brigham & Women's Hospital, durgapalsneha@gmail.com
#Andriy Fedorov, BWH, fedorov@bwh.harvard.edu
#Jean-Christophe Fillion-Robin, Kitware, jchris.fillionr@kitware.com
#James Fishbaugh, SCI Institute/University of Utah, jfishbaugh@gmail.com
#Jessica Forbes, University of Iowa, jessica-forbes@uiowa.edu
#Polina Golland, MIT CSAIL, polina@csail.mit.edu
#Jeffrey Grethe, University of CA San Diego, jgrethe@ncmir.ucsd.edu
#Rola Harmouche, Brigham & Women's Hospital, rolaharmouche@gmail.com
#Nobuhiko Hata, Brigham & Women's Hospital, hata@bwh.harvard.edu
#Saurabh Jain, Johns Hopkins University, saurabh@cis.jhu.edu
#Hans Johnson, University of Iowa, hans-johnson@uiowa.edu
#Jayashree Kalpathy-Cramer, MGH, kalpathy@nmr.mgh.harvard.edu
#Tina Kapur, BWH/Harvard Medical School, tkapur@bwh.harvard.edu
#Ron Kikinis, HMS, kikinis@bwh.harvard.edu
#Regina Kim, University of Iowa, eunyoung-kim@uiowa.edu
#Franklin King, Queen's University, franklin.king@queensu.ca
#Tassilo Klein, SPL/BWH, TJKlein@bwh.harvard.edu
#Farukh Kohistani, BWH Radiology, kohistan@bc.edu
#Robin Kouver, BWH/SPL, r.kouver@gmail.com
#Andreas Lasso, PerkLab - Queen's University, lasso@queensu.ca
#Yangming Li, University of Washington, ymli81@uw.edu
#Sidong Liu, SPL/BWH, sliu@bwh.harvard.edu
#Siqi Liu, University of Sydney, sliu4512@uni.sydney.edu.au
#Bradley Lowekamp, National Institutes of Health, blowekamp@mail.nih.gov
#Murat Maga, Seattle Children's Research Institute, maga@uw.edu
#Katie Mastrogiacomo, SPL/BWH, kmast@bwh.harvard.edu
#Alireza Mehrtash, SPL/BWH, mehrtash@bwh.harvard.edu
#Dominik Meier, Brigham & Women's Hospital, meier@bwh.harvard.edu
#Jim Miller, GE Research, millerjv@ge.com
#Luiz Otavio Murta Junor, SPL/BWH, lmurta@partners.org
#Vivek Narayan, NCIGT, narayan.vivek9@gmail.com
#Pietro Nardelli, University College Cork, pietro@bwh.harvard.edu
#Yangming Ou, MGH, yangming.ou@uphs.upenn.edu
#Danielle Pace, MIT CSAIL, dfpace@mit.edu
#Keryn Palmer, Brigham & Women's Hospital, kpalmer5@partners.org
#Nirav Patel, WPI, napatel@wpi.edu
#Tobias Penzkofer, SPL, pt@bwh.harvard.edu
#Steve Pieper, Isomics Inc, pieper@isomics.com
#Csaba Pinter, Queen's University, csaba.pinter@queensu.ca
#Marcel Prastawa, GE Research, marcel.prastawa@ge.com
#Somia Pujol, Harvard Medical School, spujol@bwh.harvard.edu
#Adam Rankin, Queen's University, rankin@queensu.ca
#Aymeric Reshef, Brigham & Women's Hospital, areshef@bwh.harvard.edu
#Rahul Sastry, BWH/SPL, rahul_sastry@hms.harvard.edu
#Peter Savadjiev, Brigham & Women's Hospital, petersv@bwh.harvard.edu
#Gregory Sharp, MGH, gcsharp@mgh.harvard.edu
#Emylin Sousa, BWH/SPL, emylin.sousa@gmail.com
#Ramesh Sridharan, MIT CSAIL, rameshvs@csail.mit.edu
#Matthew Toews, BWH/Harvard Medical School, mt@bwh.harvard.edu
#Ethan Ulrich, University of Iowa, ethan-ulrich@uiowa.edu
#Tamas Ungi, Queen's University, ungi@queensu.ca
#Kevin Wang, Princess Margaret Cancer Centre, kevin.wang@rmp.uhn.ca
#David Welch, University of Iowa, david-welch@uiowa.edu
#William Wells, Brigham & Women's Hospital, sw@bwh.harvard.edu
#Phillip White, BWH/Harvard Medical School, white@bwh.harvard.edu
#Alex Yarmarkovich, ISOMICS Inc., alexy@bwh.harvard.edu
#Ryan Young, Seattle Children's Research Institute, ryan.young@seattlechildrens.org
#Paolo Zaffino, University Magna Graecia of Catanzaro, p.zaffino@unicz.it
#Chenxi Zhang, Brigham & Women's Hospital, chenxizhang@fudan.edu.cn
#Fan Zhang, University of Sydney, fzha8048@uni.sydney.edu.au

2014 Summer Project Week

2014-06-22T17:15:31Z

Murta: /* Other */

__NOTOC__

[[image:PW-MIT2014.png|300px]]

Dates: June 23-27, 2014.

Location: MIT, Cambridge, MA.

==Agenda==

{|border="1"
|-style="background:#b0d5e6;color:#02186f"
!style="width:10%" |Time
!style="width:18%" |Monday, June 23
!style="width:18%" |Tuesday, June 24
!style="width:18%" |Wednesday, June 25
!style="width:18%" |Thursday, June 26
!style="width:18%" |Friday, June 27
|-
|
|bgcolor="#dbdbdb"|'''Project Presentations'''
|bgcolor="#6494ec"|'''NA-MIC Update Day'''
|
|bgcolor="#88aaae"|'''IGT Day'''
|bgcolor="#faedb6"|'''Reporting Day'''
|-
|bgcolor="#ffffdd"|'''8:30am'''
|
|bgcolor="#ffffaa"|Breakfast
|bgcolor="#ffffaa"|Breakfast
|bgcolor="#ffffaa"|Breakfast
|bgcolor="#ffffaa"|Breakfast
|-
|bgcolor="#ffffdd"|'''9am-12pm'''
|
|'''10-11:30pm''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: DICOM|DICOM]] (Steve Pieper)
[[MIT_Project_Week_Rooms|Grier Room (Left)]]
|
|
'''9:00-10:30am''' [[2014_Tutorial_Contest|Tutorial Contest Presentations (Sonia Pujol)]] <br>
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
<br>----------------------------------------<br>
'''10am-12pm: <font color="#4020ff">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: IGT Neuro|Image-Guided Therapy - Neurosurgery]] (Alexandra Golby, Tina Kapur) <br>
[[MIT_Project_Week_Rooms#Star|Star]]
|'''10am-12pm:''' [[#Projects|Project Progress Updates]] <br>
'''12pm''' [[Events:TutorialContestJune2014|Tutorial Contest Winner Announcement]]
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
|-
|bgcolor="#ffffdd"|'''12pm-1pm'''
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch
|bgcolor="#ffffaa"|Lunch boxes; Adjourn by 1:30pm
|-
|bgcolor="#ffffdd"|'''1pm-5:30pm'''
|'''1-1:05pm: <font color="#503020">Ron Kikinis: Welcome</font>'''
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
<br>----------------------------------------<br>
'''1:05-3:30pm:''' [[#Projects|Project Introductions]] (all Project Leads)
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Rooms]]
<br>----------------------------------------<br>
'''3:30-4:30pm''' [[2014 Summer Project Week Breakout Session:SlicerExtensions|Slicer4 Extensions]] (Jean-Christophe Fillion-Robin) <br>
[[MIT_Project_Week_Rooms#Grier_34-401_AB|Grier Room (Left)]]
|'''1-3pm:''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: QIICR|QIICR]] (Andrey Fedorov)
[[MIT_Project_Week_Rooms#Kiva|Kiva]]
|'''1-2:30pm:''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: Contours|Contours]] (Adam Rankin, Csaba Pinter)
[[MIT_Project_Week_Rooms#Kiva|Kiva]]
|'''1-3pm:''' <font color="#503020">Breakout Session:'''</font><br>[[2014 Project Week Breakout Session: IGT Prostate|Image-Guided Therapy - Prostate Interventions]] (Clare Tempany, Noby Hata)
[[MIT_Project_Week_Rooms#Star|Star]]
<br>----------------------------------------<br>
|
|-
|bgcolor="#ffffdd"|'''5:30pm'''
|bgcolor="#f0e68b"|Adjourn for the day
|bgcolor="#f0e68b"|Adjourn for the day
|bgcolor="#f0e68b"|Adjourn for the day
|bgcolor="#f0e68b"|Adjourn for the day
|
|}

== '''Background''' ==

Project Week is a hands on activity -- programming using the open source [[NA-MIC-Kit|NA-MIC Kit]], algorithm design, and clinical application -- that has become one of the major events in the NA-MIC, NCIGT, and NAC calendars. It is held in the summer at MIT, typically the last week of June, and a shorter version is held in Salt Lake City in the winter, typically the second week of January.

Active preparation begins 6-8 weeks prior to the meeting, when a kick-off teleconference is hosted by the NA-MIC Engineering, Dissemination, and Leadership teams, the primary hosts of this event. Invitations to this call are sent to all NA-MIC members, past attendees of the event, as well as any parties who have expressed an interest in working with NA-MIC. The main goal of the kick-off call is to get an idea of which groups/projects will be active at the upcoming event, and to ensure that there is sufficient NA-MIC coverage for all. Subsequent teleconferences allow the hosts to finalize the project teams, consolidate any common components, and identify topics that should be discussed in breakout sessions. In the final days leading upto the meeting, all project teams are asked to fill in a template page on this wiki that describes the objectives and plan of their projects.

The event itself starts off with a short presentation by each project team, driven using their previously created description, and allows all participants to be acquainted with others who are doing similar work. In the rest of the week, about half the time is spent in breakout discussions on topics of common interest of subsets of the attendees, and the other half is spent in project teams, doing hands-on programming, algorithm design, or clinical application of NA-MIC kit tools. The hands-on activities are done in 10-20 small teams of size 3-5, each with a mix of experts in NA-MIC kit software, algorithms, and clinical. To facilitate this work, a large room is setup with several tables, with internet and power access, and each team gathers on a table with their individual laptops, connects to the internet to download their software and data, and is able to work on their projects. On the last day of the event, a closing presentation session is held in which each project team presents a summary of what they accomplished during the week.

A summary of all past NA-MIC Project Events is available [[Project_Events#Past|here]].

Please make sure that you are on the [http://public.kitware.com/mailman/listinfo/na-mic-project-week na-mic-project-week mailing list]

=Projects=
* [[2014_Project_Week_Template | Template for project pages]]

==TBI==
*[[2014_Summer_Project_Week:TBI_Segmentation| Interactive segmentation for traumatic brain injury ]] (Bo Wang, Marcel Prastawa, Andrei Irimia, John D. Van Horn, Guido Gerig)

==Atrial Fibrillation==

==Huntington's Disease==
*[[2014_Summer_Project_Week:FiberTractDispersion| Fiber Tract Dispersion and UKF Tractography]] (Peter Savadjiev, Yogesh Rathi, Hans Johnson, C-F Westin)

==Head and Neck Cancer==
*[[2014_Summer_Project_Week:Interactive_DIR| Interactive DIR]] (Greg Sharp, Ivan Kolesov, Allen Tannenbaum)
*[[2014_Summer_Project_Week:DIR_validation_tools| DIR validation tools]] (Greg Sharp, Ivan Kolesov, Allen Tannenbaum)
*[[2014_Summer_Project_Week:Upload_HN_data| Upload H&N data]] (Greg Sharp, Paolo Zaffino)
*[[2014_Summer_Project_Week:DIR_stop_and_restart| DIR stop and restart]] (Paolo Zaffino, Greg Sharp)

==Slicer4 Extensions==

*[[2014_Summer_Project_Week:Multidim Data| Multidim Data]] (Kevin Wang, Andras, ?)
*[[2014_Summer_Project_Week:DICOM-SRO import| DICOM-SRO import]] (Kevin Wang)
*[[2014_Summer_Project_Week:PLM_engineering| Plastimatch extension re-engineering]] (Greg Sharp, Paolo Zaffino, Andras, Csaba, Kevin)

==Cardiac==

==Stroke==
*[[2014_Summer_Project_Week:Stroke-ImagingGenetics | Stroke Imaging Genetics]] (Adrian Dalca, Ramesh Sridharan, Polina Golland)
*[[2014_Summer_Project_Week:Stroke-SuperResolution | Stroke Super Resolution]] (Adrian Dalca, Ramesh Sridharan, Polina Golland)

==Brain Segmentation==

==Image-Guided Therapy==

* SlicerIGT extension: testing, tutorials, website (Tamas Ungi, Nobuhiko Hata)
* [[Gestural Point of Care Interface for IGT]] (Saskia, Franklin, Steve, Tobias)
*[[2014_Summer_Project_Week:MR-Ultrasound_Registration_for_Prostate_Interventions | MR-Ultrasound Registration for Prostate Interventions]] (Chenxi Zhang, Andriy Fedorov, Andras)
*[[2014_Summer_Project_Week:Surface_approximation_from_contour_points | Surface approximation from contour points]] (Chenxi Zhang, Csaba Pinter, Andrey Fedorov)
* [[2014_Summer_Project_Week:Intelligent_Steering | Steered image registration using intelligent interfaces for minimal user interaction]] (Marcel Prastawa, Jim Miller, Steve Pieper)
* [[2014_Summer_Project_Week:Image To Mesh Conversion for Brain MRI | Image To Mesh Conversion for Brain MRI]] (Fotis Drakopoulos, Yixun Liu, Andrey Fedorov, Ron Kikinis, Nikos Chrisochoides)
* [[2014_Summer_Project_Week:An ITK implementation of Physics-Based Non-Rigid Registration method for Brain Shift | An ITK implementation of Physics-Based Non-Rigid Registration method for Brain Shift]] (Fotis Drakopoulos, Yixun Liu, Andriy Kot, Andrey Fedorov, Olivier Clatz, Ron Kikinis, Nikos Chrisochoides)
* [[2014_Summer_Project_Week:Robot_Control_With_OpenIGTLink | Robot Control With OpenIGTLink]] ( Gregory Fischer(WPI), Nirav Patel(WPI), Nobuhiko Hata (BWH) )
* [[2014_Summer_Project_Week:Open_source_electromagnetic_trackers_usingOpenIGTLink| Open-source electromagnetic trackers using OpenIGTLink]] (Peter Traneus Anderson, Tina Kapur, Sonia Pujol)

==Radiation Therapy==
*[[2014_Summer_Project_Week:External Beam Planning| External Beam Planning]] (Kevin Wang, Greg Sharp, Maxime Desplanques, ?)
*[[2014_Summer_Project_Week:Proton_pencil_beam| Proton pencil beam dose calculation]] (Maxime Desplanques, Kevin Wang, Greg Sharp)

==Chronic Obstructive Pulmonary Disease ==

*[[2014_Summer_Project_Week: Pectoralis muscle segmentation| Pectoralis muscle segmentation]] (Rola Harmouche, James Ross, Raul San Jose)

==[http://qiicr.org QIICR]==
* [[2014_Summer_Project_Week: RWV mapping support|Real world value mapping support]] (Andrey, Ethan, Andras, Steve, Jim, ...)
* [[2014_Summer_Project_Week: CLI Derived DICOM Data| Proper formatting of DICOM Derived Data from CLI]] (Steve, Andrey, Jim, {Michael and David remotely})
* [[2014_Summer_Project_Week: DICOM SEG conversion to support archival of QIN Grand challenges results|DICOM SEG conversion to support archival of QIN Grand challenges results]] (Jayashree, Andrey, Steve, {David remotely})

==Infrastructure==
*Slicer 4.4 Release (JC, Steve, Nicole)
* [[2014_Summer_Project_Week: Chronicle| Chronicle]] (Steve)
* [[2014_Summer_Project_Week: Volume Registration|Volume Registration]] (Steve, Greg, Marcel, Jim)
* [[2014_Summer_Project_Week:Markups | Markups]] (Nicole Aucoin)
*[[2014_Summer_Project_Week:Pluggable Label Statistics |Pluggable Label Statistics]] (Andrey , Ethan, Steve, Brad, Jim? Dirk?)
*[[2014_Summer_Project_Week:Subject_hierarchy_integration | Subject hierarchy integration]] (Csaba, Steve, Jc, Andras?, ?)
*[[2014_Summer_Project_Week:Contours | Contours]] (Adam Rankin, Csaba, Andras, Steve, Jc, ?)
*[[2014_Summer_Project_Week:Parameter Node Serialization | Parameter Node Serialization]] (Kevin Wang, Andras, Steve, Jim, Csaba, ?)
*[[2014_Summer_Project_Week:Self-tests for non-linear transforms | Self-tests for non-linear transforms]] (Xining Du)

==Feature Extraction==
*[[2014_Summer_Project_Week:Tumor_DCE-MRI_Segmentation | Breast Tumor Segmentation]] (Vivek Narayan, Jay Jagadeesan)
*[[2014_Summer_Project_Week:Tumor_Heterogeneity_Analysis | Breast Tumor Heterogeneity Analysis]] (Vivek Narayan, Jay Jagadeesan)
*Quantitative image feature extraction in Non-Small Cell Lung Cancer (Hugo Aerts)
*[[2014_Summer_Project_Week:Invariant_Feature_Extraction_Slicer | Invariant Feature Methods in Slicer]] (Matthew Toews, Nicole Aucoin, Sandy Wells)

==Other==
*[[2014_Summer_Project_Week:Slicer_Murin_Shape_Analysis | Shape Analysis for the developing murine skull]] (Murat Maga, Ryan Young, Seattle Chidren's Hospital).
*[[2014_Summer_Project_Week:Slicer_LDDMM_Shape_Analysis | Slicer Interface to LDDMM shape anlaysis]] (Saurabh Jain, JHU; Steve Pieper, Isomics; Josh Cates, SCI, Utah; Hans Johnson, Iowa; Martin Styner, UNC)
*[[2014_Summer_Project_Week:Image_Registration_with_Sliding_Motion_Constraints | Image Registration with Sliding Motion Constraints]] (Alexander Derksen, Kanglin Chen, Gregory Sharp)
*[[2014_Summer_Project_Week:Atlas Selection | Atlas Selection]] (Kanglin Chen, Gregory Sharp)
*[[2014_Summer_Project_Week:Multiscale_Non_Local_Means_filter_(NLM)_for_chest_CT_images | Multiscale Non Local Means filter (NLM) for chest CT images]] (Pietro Nardelli, University College Cork (UCC), Ireland)
*[[2014_Summer_Project_Week:Intraoperative_Registration_of_preoperative_CT_and_C-arm_CT_of_the_lung | Intraoperative Registration of preoperative CT and C-arm CT of the lung]] (Katharina Breininger, Jay Jagadeesan)
*[[2014_Summer_Project_Week:CAD_Toolbox_for_Neurological_Disorders | CAD Toolbox for Neurological Disorders]] (Sidong Liu, Siqi Liu, Fan Zhang, Yang Song, Weidong Cai, Sonia Pujol, Ron Kikinis)
*[[2014_Summer_Project_Week:Longitudinal_patient_specific_DTI_analysis | Longitudinal patient-specific DTI analysis using Slicer for neonatal asphyxia]] (Anuja Sharma, SCI, Utah; Francois Budin, UNC; Martin Styner, UNC; Guido Gerig, SCI, Utah)
*[[2014_Summer_Project_Week:Cortical_Dysplasia_Identification | Tools for Dysplasia Identification in Epilepsy]] (Luiz Murta; Emylin Souza; Tina Kapur; Ron Kikinis)

== '''Logistics''' ==

*'''Dates:''' June 23-27, 2014.
*'''Location:''' [[MIT_Project_Week_Rooms| Stata Center / RLE MIT]].
*'''REGISTRATION:''' https://www.regonline.com/namic2014summerprojectweek. Please note that as you proceed to the checkout portion of the registration process, RegOnline will offer you a chance to opt into a free trial of ACTIVEAdvantage -- click on "No thanks" in order to finish your Project Week registration.
*'''Registration Fee:''' $300.
*'''Hotel:''' Similar to previous years, no rooms have been blocked in a particular hotel.
*'''Room sharing''': If interested, add your name to the list: [[2014_Summer_Project_Week/RoomSharing|here]]

== '''Registrants''' ==

Do not add your name to this list - it is maintained by the organizers based on your paid registration. ([https://www.regonline.com/namic2014summerprojectweek Please click here to register.])

#Hugo Aerts, Dana Farber/Harvard, hugo_aerts@dfci.harvard.edu
#Nassim Alikacem, Brigham & Women's Hospital, Nassim.Alikacem@gmail.com
#Peter Anderson, retired, traneus@verizon.net
#Nicole Aucoin, Brigham & Women's Hospital, nicole@bwh.harvard.edu
#Eva Breininger, Brigham & Women's Hospital, ebreininger@partners.org
#Francois Budin, NIRAL-UNC, fbudin@unc.edu
#Saskia Camps, SPL, saskiacamps@gmail.com
#Lucia Cevidanes, University of Michigan, luciacev@umich.edu
#Laurent Chauvin, SPL, lchauvin@bwh.harvard.edu
#Kanglin Chen, Fraunhofer MEVIS, kanglin.chen@mevis.fraunhofer.de
#Adrian Dalca, MIT CSAIL, adalca@mit.edu
#Alexander Derksen, Fraunhofer MEVIS, alexander.derksen@mevis.fraunhofer.de
#Maxime Desplanques, MGH/Politecnico di Milano, maxime.desplanques@cnao.it
#Fotis Drakopoulos, Old Dominion University, fdrakopo@gmail.com
#Sneha Durgapal, Brigham & Women's Hospital, durgapalsneha@gmail.com
#Andriy Fedorov, BWH, fedorov@bwh.harvard.edu
#Jean-Christophe Fillion-Robin, Kitware, jchris.fillionr@kitware.com
#James Fishbaugh, SCI Institute/University of Utah, jfishbaugh@gmail.com
#Jessica Forbes, University of Iowa, jessica-forbes@uiowa.edu
#Polina Golland, MIT CSAIL, polina@csail.mit.edu
#Jeffrey Grethe, University of CA San Diego, jgrethe@ncmir.ucsd.edu
#Rola Harmouche, Brigham & Women's Hospital, rolaharmouche@gmail.com
#Nobuhiko Hata, Brigham & Women's Hospital, hata@bwh.harvard.edu
#Saurabh Jain, Johns Hopkins University, saurabh@cis.jhu.edu
#Hans Johnson, University of Iowa, hans-johnson@uiowa.edu
#Jayashree Kalpathy-Cramer, MGH, kalpathy@nmr.mgh.harvard.edu
#Tina Kapur, BWH/Harvard Medical School, tkapur@bwh.harvard.edu
#Ron Kikinis, HMS, kikinis@bwh.harvard.edu
#Regina Kim, University of Iowa, eunyoung-kim@uiowa.edu
#Franklin King, Queen's University, franklin.king@queensu.ca
#Tassilo Klein, SPL/BWH, TJKlein@bwh.harvard.edu
#Farukh Kohistani, BWH Radiology, kohistan@bc.edu
#Robin Kouver, BWH/SPL, r.kouver@gmail.com
#Andreas Lasso, PerkLab - Queen's University, lasso@queensu.ca
#Yangming Li, University of Washington, ymli81@uw.edu
#Sidong Liu, SPL/BWH, sliu@bwh.harvard.edu
#Siqi Liu, University of Sydney, sliu4512@uni.sydney.edu.au
#Bradley Lowekamp, National Institutes of Health, blowekamp@mail.nih.gov
#Murat Maga, Seattle Children's Research Institute, maga@uw.edu
#Katie Mastrogiacomo, SPL/BWH, kmast@bwh.harvard.edu
#Alireza Mehrtash, SPL/BWH, mehrtash@bwh.harvard.edu
#Dominik Meier, Brigham & Women's Hospital, meier@bwh.harvard.edu
#Jim Miller, GE Research, millerjv@ge.com
#Luiz Otavio Murta Junor, SPL/BWH, lmurta@partners.org
#Vivek Narayan, NCIGT, narayan.vivek9@gmail.com
#Pietro Nardelli, University College Cork, pietro@bwh.harvard.edu
#Yangming Ou, MGH, yangming.ou@uphs.upenn.edu
#Danielle Pace, MIT CSAIL, dfpace@mit.edu
#Keryn Palmer, Brigham & Women's Hospital, kpalmer5@partners.org
#Nirav Patel, WPI, napatel@wpi.edu
#Tobias Penzkofer, SPL, pt@bwh.harvard.edu
#Steve Pieper, Isomics Inc, pieper@isomics.com
#Csaba Pinter, Queen's University, csaba.pinter@queensu.ca
#Marcel Prastawa, GE Research, marcel.prastawa@ge.com
#Somia Pujol, Harvard Medical School, spujol@bwh.harvard.edu
#Adam Rankin, Queen's University, rankin@queensu.ca
#Aymeric Reshef, Brigham & Women's Hospital, areshef@bwh.harvard.edu
#Rahul Sastry, BWH/SPL, rahul_sastry@hms.harvard.edu
#Peter Savadjiev, Brigham & Women's Hospital, petersv@bwh.harvard.edu
#Gregory Sharp, MGH, gcsharp@mgh.harvard.edu
#Emylin Sousa, BWH/SPL, emylin.sousa@gmail.com
#Ramesh Sridharan, MIT CSAIL, rameshvs@csail.mit.edu
#Matthew Toews, BWH/Harvard Medical School, mt@bwh.harvard.edu
#Ethan Ulrich, University of Iowa, ethan-ulrich@uiowa.edu
#Tamas Ungi, Queen's University, ungi@queensu.ca
#Kevin Wang, Princess Margaret Cancer Centre, kevin.wang@rmp.uhn.ca
#David Welch, University of Iowa, david-welch@uiowa.edu
#William Wells, Brigham & Women's Hospital, sw@bwh.harvard.edu
#Phillip White, BWH/Harvard Medical School, white@bwh.harvard.edu
#Alex Yarmarkovich, ISOMICS Inc., alexy@bwh.harvard.edu
#Ryan Young, Seattle Children's Research Institute, ryan.young@seattlechildrens.org
#Paolo Zaffino, University Magna Graecia of Catanzaro, p.zaffino@unicz.it
#Chenxi Zhang, Brigham & Women's Hospital, chenxizhang@fudan.edu.cn
#Fan Zhang, University of Sydney, fzha8048@uni.sydney.edu.au

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T15:01:58Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in [http://www.cs.waikato.ac.nz/ml/weka/ WEKA] environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring
<br>
<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Once trained, classifier is easily included in this system, and can help surgery decision.
<li>Although these preliminary results are promising, the method still needs to be validated considering TL separatelly and in a larger image database.

<h3>To do list:</h3>

<li>As a nest step, I´ll integrate the code to 3DSlicer.
<li>Try new schems of texture analisys to this application.
<li>Implement few machine learning algorithms, in addition to image processing in C++.
<li>Speed up some time consumming algorithms with GPU programming.
<li>Extend these and other ideas to cotical dysplasia: 2014 !

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T14:48:20Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in [http://www.cs.waikato.ac.nz/ml/weka/ WEKA] environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring
<br>
<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Once trained, classifier is easily included in this system, and can help surgery decision.
<li>Although these preliminary results are promising, the method still needs to be validated considering TL separatelly and in a larger image database.

<h3>To do list:</h3>

<li>As a nest step, I´ll integrate the code to 3DSlicer.
<li>Try new schems of texture analisys to this application.
<li>Implement few learning machine algorithms, in addition to image processing in C++.
<li>Speed up some time consumming algorithms with GPU programming.
<li>Extend these and other ideas to cotical dysplasia: 2014 !

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T14:29:02Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in [http://www.cs.waikato.ac.nz/ml/weka/ WEKA] environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring
<br>
<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Once trained, classifier is easily included in this system, and can help surgery decision.
<li>Although these preliminary results are promising, the method still needs to be validated considering TL separatelly and in a larger image database.

<h3>To do list:</h3>

<li>As a nest step, I´ll integrate the code to 3DSlicer.
<li>Try new schems of texture analisys to this application.
<li>Implement few learning machine algorithms, in addition to image processing in C++.
<li>Extend these and other ideas to cotical dysplasia: 2014 !

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T14:27:37Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in [http://www.cs.waikato.ac.nz/ml/weka/ WEKA] environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring
<br>
<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Once trained, classifier is easily included in this system, and can help surgery decision.
<li>Although these preliminary results are promising, the method still needs to be validated considering TL separatelly and in a larger image database.

<h3>To do list:</h3>

<li>As a nest step, I´ll integrate the code to 3DSlicer.
<li>Try new schems of texture analisys to this application.
<li>Implement few learning machine algorithms, in addition to image processing in C++.
<li>Extend these and other ideas to cotical dysplasia: next year !

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T14:26:04Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in [http://www.cs.waikato.ac.nz/ml/weka/ WEKA] environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring
<br>
<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Once trained, classifier is easily included in this system, and can help surgery decision.
<li>Although these preliminary results are promising, the method still needs to be validated considering TL separatelly and in a larger image database.

<h3>To do list:</h3>

<li>As a nest step, I´ll integrate the code to 3DSlicer.
<li>Try new schems of texture analisys to this application.
<li>Implement few learning machine algorithms, in addition to image processing in C++.
<li>Extend these idea to cotical dysplasia: next year!

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T14:19:52Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in [http://www.cs.waikato.ac.nz/ml/weka/ WEKA] environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring
<br>
<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although these preliminary results are promising, the method still needs to be validated considering TL separatelly and in a larger image database.

<h3>To do list:</h3>

<li>As a nest step, I´ll integrate the code to 3DSlicer.
<li>Try new schems of texture analisys to this application.
<li>Extend these idea to cotical dysplasia: next year!

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T14:06:18Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring
<br>
<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although these preliminary results are promising, the method still needs to be validated considering TL separatelly and in a larger image database.

<h3>To do list:</h3>

<li>As a nest step, I´ll integrate the code to 3DSlicer.
<li>Try new schems of texture analisys to this application.
<li>Extend these idea to cotical dysplasia: next year!

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T14:02:19Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring
<br>
<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although the results are promising, the method should be validated considering TL separatelly and in a larger database.

<h3>To do list:</h3>

<li>As a nest step, I´ll integrate the code to 3DSlicer.
<li>Try new schems of texture analisys to this application.
<li>Extend these idea to cotical dysplasia: next year!

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T10:37:50Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring
<br>
<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although the results are promising, the method should be validated considering TL separatelly and in a larger database.
<li>As a nest step, I´ll integrate the code to 3DSlicer.

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T05:39:29Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring
<br>
<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although the results are promising, the method should be validated considering TL separatelly and in a larger database.

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T05:38:20Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] no blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] blurring

<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, from preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although the results are promising, the method should be validated considering TL separatelly and in a larger database.

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T05:35:37Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] no blurring

<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, the preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although the results are promising, the method should be validated considering TL separatelly and in a larger database.

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T05:34:17Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]] Without blurring
[[File:tlBlurred.png]]
[[File:histBlurred.png]] With blurring

<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, the preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although the results are promising, the method should be validated considering TL separatelly and in a larger database.

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

File:HistNoblur.png

2013-06-21T05:30:45Z

Murta:

File:HistBlurred.png

2013-06-21T05:29:13Z

Murta:

File:TlNoblur.png

2013-06-21T05:28:51Z

Murta:

File:TlBlurred.png

2013-06-21T05:28:26Z

Murta:

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T05:22:41Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.
<br>
[[File:tlNoblur.png]]
[[File:histNoblur.png]]
[[File:tlBlurred.png]]
[[File:histBlurred.png]]

<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, the preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although the results are promising, the method should be validated considering TL separatelly and in a larger database.

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T05:21:00Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.

[[File:TlNoblur.png]]
[[File:HistNoblur.png]]
[[File:TlBlurred.png]]
[[File:HistBlurred.png]]

<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, the preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although the results are promising, the method should be validated considering TL separatelly and in a larger database.

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T05:18:54Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.

[[File:TlNoblur.png]]
[[File:HistNoblur.png]]
[[File:TlBlurred.png]]
[[File:HistBlurred.png]]

<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, the preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although the results are promising, the method should be validated in a larger database.

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T05:09:40Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.

[[File:tlNoblur.png]]
[[File:histNoblur.png]]
[[File:tlBlurred.png]]
[[File:histBlurred.png]]

<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

<h3>Conclusions</h3>

<li>As a result, the preselected 32 texture feature descriptors, 6 were shown to be relevant for classification of TL tissues.
<li>Thus, we observe that using approach of texture analysis of segmented images, can helps the specialist to identify signal change MRIs in the temporal lobes of patients with TLE.
<li>Although the results are promising, the method should be validated in a larger database.

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T05:02:31Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.

<br>
<b>Decision Tree</b>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T05:00:16Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 47%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 50%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 37%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.

<br>
<b>Decision Tree</>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T04:55:47Z

Murta:

__NOTOC__
<gallery>
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* [http://www5.usp.br/en/ USP] - Luiz Murta

<div style="margin: 20px;">
<div style="width: 37%; float: left; padding-right: 3%;">

<h3>Objective</h3>
This project will investigate the presence and location of the epileptogenic focus in temporal lobe by analyzing patterns of texture in magnetic resonance imaging (MRI) after segmentation using anisotropic diffusion filters anomalous and geodesic active contour.

<li>The Temporal Lobe Epilepsy (TLE) with hippocampal sclerosis affects other mesial temporal lobe structures, showing the change in signal intensity of white matter (WM) and gray matter (GM) in Magnetic Resonance Imaging (MRI) in patients with TLE.
<li>Such abnormalities of signal intensity changes in the temporal lobes, or loss of marking between WM and GM are referred as blurring in literature.
<li>TLE is the most common type of refractory epilepsy in adults, and thus strong candidate to surgery.
<br><br>
<b>Purpose</b>
<li>To provide a better quality of life for those with TLE, a surgical procedure is usually proposed for extracting the brain region that is the focus of epilepsy.
<li>This project aims to implment a semi-automated method capable of locating lesions, that is difficult to be detected by visual inspection, is built to help experts identify tissues presenting blurring in temporal lobes.
<li>The proposed software tool was developed in C + +, using three toolkits to support: VTK, ITK and QT.
<li>To provide segmentation of the temporal lobes, we used the Geodesic Active Contour method combined with the anomalous anisotropic diffusion filter.

</div>

<div style="width: 40%; float: left;">

<h3>Progress</h3>
Examples:<br>
[[File:ex_a.png]]<br>
Normal MRI at mesial temporal lobe<br>

[[File:ex_b.png]]<br>
MRI containing blurring phenomena on right side as indicated by the yellow arrow<br>

</div> <br>

<div style="margin: 20px;">
<div style="width: 60%; float: left; padding-right: 3%;">

<h3>Methods</h3>
[[File:param.png]]
[[File:imageUI.png]] <br>
Default adjustable parameters, and original image,<br>
<br>
<b>Segmented TL córtex</b>
[[File:segmB.png]]
[[File:segmR.png]]

</div>

<div style="width: 27%; float: left;">

<h3>Results</h3>
<b>Classifier</b>
<li>Once segmented, the images are analyzed using texture descriptors from the intensity histogram and local pixels dependence measures proposed by Haralick were extracted.
Subsequently, the set of feature vectors from the segmented images is evaluated using three classifiers in WEKA environment.
<li>Images of 32 patients with TLE were used, 15 of those exhibiting blurring phenomenon in TL and 17 do not.
<li>In total, a set of 98 planar images was segmented, and 51 are classified in "with blurring" category and 47 "without blurring" category.
<br>
<li>Of all three classifiers:
<br>-- artificial neural network;
<br>-- nearest neighbour;
<br>-- and decision tree
<li> tested, the one based on "J48" decision tree had the best, and most interesting results.
<li>32 feature have been used to constructo descriptors vectors of segmented images:
<br>-- 24 were extracted from the co-occurrence matrix using Haralick texture descriptors and
<br>-- 8 were intensity statistics obtained from histogram.
<li>When evaluating the performance of the three classifiers selected for the learning process, as a basis of 70 training samples and 28 test.
<li>It was found that, for decision tree, precision and area under the ROC curve were 71.42 % and 69.30 %, respectively.

<br>
<b>Decision Tree</>
<br><br>
D_Hist_MeanIntensity <= 0.453<br>
| D_Hist_stdDev <= 0.138: with_blurring (5.0)<br>
| D_Hist_stdDev > 0.138<br>
| | E_Hist_Kurtosis <= 0.497<br>
| | | E_COOmeanHomogeneity_dist3 <= 0.425<br>
| | | | D_COOmeanHomogeneity_dist2 <= 0.318: without_blurring (9.0/2.0)<br>
| | | | D_COOmeanHomogeneity_dist2 > 0.318: with_blurring (5.0)<br>
| | | E_COOmeanHomogeneity_dist3 > 0.425<br>
| | | | E_COOmeanHomogeneity_dist3 <= 0.895: without_blurring (26.0/1.0)<br>
| | | | E_COOmeanHomogeneity_dist3 > 0.895: with_blurring (3.0/1.0)<br>
| | E_Hist_Kurtosis > 0.497: with_blurring (5.0)<br>
D_Hist_MeanIntensity > 0.453<br>
| E_Hist_Kurtosis <= 0.48: with_blurring (12.0)<br>
| E_Hist_Kurtosis > 0.48<br>
| | E_COOmeanHomogeneity_dist1 <= 0.432: with_blurring (3.0)<br>
| | E_COOmeanHomogeneity_dist1 > 0.432: without_blurring (2.0)<br>
Number of Leaves : 9<br>

</div> <br>

==References==
*Shaker, M. & Soltanian-Zadeh, H., 2008. Voxel-Based Morphometric Study of Brain Regions from Magnetic Resonance Images in Temporal Lobe Epilepsy. Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on, 209-212.

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T04:49:56Z

Murta:

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T04:41:26Z

Murta:

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T04:34:04Z

Murta:

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T04:12:24Z

Murta:

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T04:06:36Z

Murta:

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T04:04:52Z

Murta:

File:SegmR.png

2013-06-21T03:59:55Z

Murta:

File:SegmB.png

2013-06-21T03:59:31Z

Murta:

File:ImageUI.png

2013-06-21T03:58:56Z

Murta:

File:Param.png

2013-06-21T03:58:18Z

Murta:

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T03:57:11Z

Murta:

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T03:33:21Z

Murta:

2013 Summer Project Week:Epilepsy Surgery

2013-06-21T03:32:12Z

Murta: