NAMIC Wiki - User contributions [en]

File:Eight.jpg

2007-04-12T18:17:51Z

Sw:

File:Ten.jpg

2007-04-12T18:15:21Z

Sw:

File:Nine.jpg

2007-04-12T18:14:59Z

Sw:

File:Seven.jpg

2007-04-12T18:13:42Z

Sw:

File:Five.jpg

2007-04-12T18:13:11Z

Sw:

File:Four.jpg

2007-04-12T18:12:52Z

Sw:

File:Three.jpg

2007-04-12T18:12:26Z

Sw:

File:Two.jpg

2007-04-12T18:12:11Z

Sw:

File:One.jpg

2007-04-12T18:10:45Z

Sw:

EMSegmenter

2007-04-12T18:08:02Z

Sw: /* History */

==History==

Tracing the history of the EMSegmenter...

===1995===
*Image:
*Strengths:
*Weaknesses:
*Publications:

===1999===
*Image:
*Strengths:
*Weaknesses:
*Publications:

The EM segmenter grew out of a collaboration with Shenton's group in
about 1993. The goal was to get good automatic segmentations of white
matter and gray matter from T1 weighted MRI. The biggest difficulty
was the intensity inhomogeneities, or "shading", artifact in the
images. The effect of the artifact was that a single threshold could
not be used to separate white matter and gray matter. At that time,
the MRI scanner used for research at BWH had an annoyingly large
shading artifact.

Various approaches to the problem were tried, some giving good
results, but there were remaining imperfections in the results.
Eventually, we decided to construct an explicit representation of the
intensity artifact, and attempt to recover the artifact and the
segmentation simultaneously.

We chose the Expectation Maximization (EM) algorithm, a statistical
estimation method that is used when some data is considered to be
"missing".

The result was an iterative algoritm that alternates between two
steps.

In the "E" step, the probability of the tissue label at each voxel is
estimated, given the image data and the current estimate of the
intensity artifact.

In the "M" step, the intensity artifact is re-estimated, given the
image data and current estimate of the tissue label probabilities.

The EM segmenter proved to be very robust to shading artifacts,
but in addition, it was also robust to "inter-scan
inhomogeneities". With previous classification approaches to
segmentation, "training" was needed on a per-scan basis, because
of intensity changes from scan to scan.

The EM segmener was the first algorithm that could produce high
quality segmentations of white matter and gray matter from MRI,
with no manual intervention needed on a per case basis. This proved
to be very valuable in a large longitudinal study of MS in the
period 1994 - 1995.

Subsequent developments:

Tina Kapur: added MRF models and Mean Field solver

Kilian Pohl:
* Added the use of anatomical atlases of specific brain
parts, e.g., the hippocampus : started to be a brain
"parcellator"

* Added simultaneous registration of atlas and subject

* developed a hierarchial method for parcellation and
validated it on schizophrenia data

* developed a mean-field level-set post-processor that
is effective for reducing the effects of noise.

---

images:

1) slice of T1 weighted mr (right temporal lobe has bad "shading"
2) threshoding result
3) EM result
4) 3D view of segmented white matter surface from thresholding
5) 3D view of segmented white matter surface from EM
7) longituidinal MS, one subect, segmentaiton result without EM
8) " " " with EM
9) surface coil image
10) surface coil image corrected by EM algorithm

== EM with Priors ==
* [[IMAGE:Ron-ISBI-07.zip| Here]] are the research related slides - I have sometimes a couple of slides on a topic so you can choose.
* The [[Media:EMFeedback3.ppt | Feedback slides]] featuring the transition from Slicer 2 to Slicer 3 were generated by Brad and me.

==sw==

Sw-retreat-notes

2007-02-19T20:17:26Z

Sw:

==Summary of Discussion ==

A general consensus emerged that the regreat was a useful thing to to, and that we should do it again next year.

Some projects were identified as potential areas for collaboration among the P41s, including:

* Sandy Wells, Clare Poynton, Mark Jenkinson: Dewarping and Registering EPI MRI

* A response to an upcoming RFA for BIRN collaborative activity

* some work towards standardizing DTI protocols across sites

* multimodality scanning and analysis of neurosurgery patients
** high quality structural scans
** MEG
** EEG

* obtaining additional modalities / scans of the SPL atlas subject
* obtaining high field ex-vivo scans of a whole brain for Florin's atlas work

* possibly doing optical sensing in the OR (perfusion related)

A discussion ensued about the possibility of "institutionalizing" some
joint activity by creating a fellowship or two to fund collaboration
across the SPL and the Martinos Center. Bruce Rosen and Ron Kikinis
agreed to investigate possible funding for these from hospital or
Partners sources.

P41-2007-Joint-Retreat

2007-02-19T20:08:48Z

Sw: /* 02.30-03.00 Break */

=Joint Retreat =
* What is the objective
**To familiarize the others with the ongoing activties in the respective cores
**To identify opportunities for leverage and augmentation
* Who:
**The MGH Center for Functional Imaging Technologies (CFIT)
**The BWH Neuroimage Analysis Center (NAC)
**Core PI's for the two centers plus not more than 2 people per core
* Where:
**MIT Stata Center, Seminar Room G449 ([[Meeting_Locations:MIT_CSAIL_Star|Patil Conference Room / Kiva]])
* When:
**Monday, Jan. 29, 11am-5pm

=Agenda=
==11.00-12.30 MGH CFIT==
**11.00-11.15 Bruce Rosen, Overview
**11.15-11.40 Bruce Fischl, High resolution anatomical modeling
**11.40-11.55 Larry Wald, High resolution structural and functional MRI
**11.55-12.10 Graham Wiggins, Large-N arrays
**12.10-12.25 David Boas, Optical Imaging
**12.25-12.45 Matti Hamalainen, MEG and mulit-modal inverse problems.
**12.45-01.15 pm Lunch Break

==01.15-02.45 BWH NAC==
**01.15-01.20 Ron Kikinis: [[media:2007-01-29-NAC-Intro.ppt |Overview]]
**01.20-01.35 CF Westin: Diffusion MRI
**01.35-01.50 Sandy Wells: fMRI Informatics
**01.50-02.10 Andrea Mewes: Neonate Imaging
**02.10-02.20 Florin Talos: Computational Neuroanatomy
**02.25-02.35 Michael Halle: Visualization and representation for atlases
**02.35-02.45 Domink Meier: Time Series Analysis PI [[Media:NACRetreat07_TSACore-Meier.ppt | Slides]]
**02.45-03.00 Steve Pieper: Status and plans for Slicer3 [[media:Pieper-eng-retreat2007-01-29.ppt | Slides]]

==02.30-03.00 Break==
*03.00-04.00 General Discussion (Sandy Wells, Bruce Fischl) [[sw-retreat-notes]]
*04.00-05.00 DTI discussion (Bruce Rosen, Ron Kikinis)

=Participants=
#Ron Kikinis, PI, NAC
#Steve Pieper
#Bruce Fischl
#David Boas
#Larry Wald
#Matti Hamalainen
#Bruce Rosen, PI, CFIT
#Carl-Fredrik Westin
#Juliette Selb
#Lilla Zollei
#Jean Augustinack
#Anna Custo
#Thomas Witzel
#Wanmei Ou
#Florin Talos
#Michael Halle
#Andrea Mewes
#Polina Golland
#Sandy Wells
#Graham Wiggins