NAMIC Wiki - User contributions [en]

Project Week 25/CNN for PseudoCT Generation from T1T2 MR

2017-06-30T12:11:27Z

Sscaramuzzino: /* Project Description */

__NOTOC__

Back to [[Project_Week_25#Projects|Projects List]]

==Key Investigators==

*[http://www.imagenglab.com/pileggi_29.html Giampaolo Pileggi] (Magna Graecia University, Italy/German Cancer Research Center (DKFZ), Germany)
*[http://www.imagenglab.com/newsite/paolo_zaffino/ Paolo Zaffino] (Magna Graecia University, Italy)
*[http://www.imagenglab.com/newsite/salvatore_scaramuzzino/ Salvatore Scaramuzzino] (Magna Graecia University/ASL Vercelli, Italy)
*[http://www.imagenglab.com/newsite/mf_spadea/ Maria Francesca Spadea] (Magna Graecia University, Italy)
*[http://isgwww.cs.uni-magdeburg.de/cas/team.php Gino Gulamhussene] (University of Magdeburg, Germany)
*[http://isgwww.cs.uni-magdeburg.de/isg/meyer.html Anneke Meyer] (University of Magdeburg, Germany)

==Project Description==
This tool allows the user to generate an HU map (Pseudo-CT) from T1/T2 input MRI. The tool is still in the early development stages with good early results in terms of Mean Absolute Error and Bias.

{| class="wikitable"
! style="text-align: left; width:27%" | Objective
! style="text-align: left; width:27%" | Approach and Plan
! style="text-align: left; width:27%" | Progress and Next Steps
|- style="vertical-align:top;"
|
The main object is to understand which topology of CNN is the most suited for the Pseudo-CT generation task.
|
Test of different structures with training on low-res images in order to speed-up computational time.
Analysis of the different output with MAE and Bias metrics
|
The most suited CNN to be used for this task consisted of an edited U-Net that does not make use of the final Dense Layer, useful for segmentation but not for regression purposes.
Also unpooling layer was used in the decoding part, with zero-filling instead of repetitions for the missing values.
Next planned steps will be the fine tuning of the CNN in order to lower the MAE and BIAS, the translation to TensorFlow (now the code uses Theano as a backend) and the data parallelization.
Interactions with Annette, Gino and Gabriele were important in order to understand how to improve the CNN topology.
|}

==Illustrations==
[[File:MRI2CT.png]]

==Background and References==

Project Week 25/Human-Computer Interaction under sterile conditions

2017-06-27T08:44:01Z

Sscaramuzzino: contributor added to sterile conditions project

__NOTOC__

Back to [[Project_Week_25#Projects|Projects List]]

==Key Investigators==
*[http://isgwww.cs.uni-magdeburg.de/cas/ Christian Hansen] (University of Magdeburg, Germany)
*[http://isgwww.cs.uni-magdeburg.de/cas/team.php Julian Hettig] (University of Magdeburg, Germany)
*[http://isgwww.cs.uni-magdeburg.de/cas/team.php Benjamin Hatscher] (University of Magdeburg, Germany)
*[http://www.researchgate.net/profile/David_Black11 David Black] (University of Bremen; Fraunhofer Institute for Medical Image Computing MEVIS, Bremen, Germany)
*[http://www.dkfz.de/en/mic/team/people/Marco_Nolden.html Marco Nolden] (German Cancer Research Center (DKFZ), Germany)
*[http://juanruizalzola.com/about-juan/ Juan Ruiz Alzola] (University of Las Palmas de Gran Canaria, Spain)
*[http://www.imagenglab.com/newsite/salvatore_scaramuzzino/ Salvatore Scaramuzzino] ("Magna Graecia" University - ASL Vercelli, Italy)

==Project Description==
{| class="wikitable"
! style="text-align: left; width:27%" | Objective
! style="text-align: left; width:27%" | Approach and Plan
! style="text-align: left; width:27%" | Progress and Next Steps
|- style="vertical-align:top;"
|
Human-Computer Interaction under Sterile Conditions.
* Review of state of the art (tochless interaction)
* Development of new user interfaces to support surgical interterventions

|

# Long, intensive conceptualization (2 days) of possibilities for gesture interaction and audio feedback for it
## We will need moderation kit (paper, pencil, markers, etc)
# Prototypes of possible auditory/visual feedback based on conceptualization
## With OSC communication protocol, David will make quick, flexible sound methods
## foot/eye interaction (?)
# [[File:OSC Test Input.pd.zip|thumb|Test incoming OSC messages using PureData]]
##Here is a test patch for incoming messages, also makes sound. Need to [https://puredata.info/downloads/pd-extended download PD-Extended] to use this.

|


|}

==Background==
[[File:Screen Shot 2017-06-26 at 12.14.57.png|500px|example HCI with freehand gestures for sound]]


Example of previous related work using standard surgical gloves and OR compatible plastic draping confirmed to be compatible with sterile requirements at BWH AMIGO:
]
<embedvideo service="youtube">https://www.youtube.com/watch?v=zSqO2pUEodw</embedvideo>

== References==

[http://isgwww.cs.uni-magdeburg.de/cas/pub/2017_IJCARS_MewesHensenWackerHansen.pdf Mewes et al. (2017) Touchless Interaction with Software in Interventional Radiology and Surgery: A Systematic Literature Review]

[http://isgwww.cs.uni-magdeburg.de/cas/pub/2017_Hettig_JCARS.pdf Hettig et al. (2017) Comparison of Gesture and Conventional Interaction Techniques for Interventional Neuroradiology]

2015 Summer Project Week:BigDataFeatures

2015-06-23T12:40:03Z

Sscaramuzzino: /* Key Investigators */

__NOTOC__
<gallery>
Image:PW-Summer2015.png|[[2015_Summer_Project_Week#Projects|Projects List]]
</gallery>

==Key Investigators==
* Matthew Toews, École de Technologie Supérieure
* William Wells, BWH, Harvard Medical School
* Raul San Jose Estepar, BWH, Harvard Medical School
* Tina Kapur, BWH, Harvard Medical School
* Utsav Pardasani, Robarts (Observing!)
* Salvatore Scaramuzzino (Interested!)
* Andrey Fedorov, BWH

==3D SIFT-Rank Visualization, SLC 2015, IPMI 2015 ==
[[File:Lung-ct.png|300px|thumb|left|Lung CT Features]]
[[File:Img-feat.png‎|300px|thumb|left|Data Reduction for 20000 lung CT volumes]]
[[File:3D_SIFT_Prostate.png|300px|thumb|left|Prostate US Features]]

==Project Description==
<div style="margin: 20px;">
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Objective</h3>
* This project will investigate the use of 3D SIFT-RANK image features for organizing and deriving information from 3D medical image volumes.
* Technology: invariant feature extraction, descriptor representation.
* Application domains: registration, segmentation, classification.
* Image domains: lung CT, brain MR, prostate and brain ultrasound.
* Clinical use case scenarios: chronic obstructive pulmonary disease, Alzheimer's disease, cancer.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Approach, Plan</h3>
* Discussion and documentation
* Algorithms: fast KNN methods, hashing, robust estimation (RANSAC, Hough transform).
* Mathematical formalisms: probabilistic inference, kernel methods, manifold learning.
</div>
<div style="width: 27%; float: left; padding-right: 3%;">
<h3>Progress</h3>
*
</div>
</div>

==References==
[http://www.na-mic.org/Wiki/index.php/3D_SIFT_VIEW] SIFT View, NAMIC 2015 SLC Project Week <br>
[http://www.matthewtoews.com/publications.html] "A Feature-based Approach to Big Data Analysis of Medical Images", M. Toews, C. Wachinger, R. S. J. Estepar, W.M. Wells III.
Information Processing in Medical Imaging (IPMI), 2015.<br>
[http://www.matthewtoews.com/publications.html] "Keypoint Transfer Segmentation", C. Wachinger, M. Toews, G. Langs, W.M. Wells IIIi, P. Golland. Information Processing in Medical Imaging (IPMI), 2015.