[Neuroimaging] New tractography challenge

Maxime Descoteaux Maxime.Descoteaux at USherbrooke.ca
Sat Jan 20 13:24:57 EST 2018


Dear Colleagues,

We are pleased to announce that the IEEE International Symposium on Biomedical Engineering (ISBI) has endorsed the 3-D Validation of Tractography with Experimental MRI (3D-VoTEM) challenge. The goal of the proposed 3D-VoTEM challenge is to assess the anatomical accuracy of common and emerging tractography pipelines and algorithms, resulting in novel insights into both the successes and limitations of these techniques. The findings from this competition will be prepared as a high impact manuscript to Sciences Advances.

Please see our URL for full challenge  details: https://my.vanderbilt.edu/votem/

Additional Challenge Details:
Uniquely, this challenge contains 3 separate datasets/challenges.

(A) The first sub-challenge is based on a 3D anisotropic diffusion phantom, created by Synaptive Medical, which contains complex geometries and anisotropic microstructure intended to mimic linear, crossing, and kissing/touching fiber bundles.

(B) The second-sub challenge features the dataset (provided by the NIH) that has previously been used to provide assessment of both sensitivity and specificity, and anatomical accuracy, of white matter tractography (Thomas et al., PNAS 2014). The dataset is composed of high SNR images of an ex vivo macaque brainat high spatial and angular resolution – a high quality dataset almost devoid of experimental confounds and artifacts that typically arise in in vivo diffusion studies. By using the framework from the original publication, the challenge presents a way to evaluate the anatomical accuracy of several new several new diffusion reconstruction and tracking approaches have been developed.

(C) The third-sub challenge is based on data acquired on the squirrel monkey brain. Both MRI and histological tracer injections were performed on the same specimen, which allows voxel-wise assessment of anatomical accuracy of tractography. In addition, high resolution histology allows spatial analysis of the errors of tractography.

Joining the challenge
To access the data and participate in the challenge, simply join each of the three challenges following the steps outlined on https://my.vanderbilt.edu/votem/registration-and-data-access/ navigate the user through the Synapse challenge interface.

Challenge Timeline
Submission site will be open from January 15-March 4. ISBI 2018 is on April 4.

Contact email
votem-organizers at googlegroups.com<mailto:votem-organizers at googlegroups.com>

Looking forward to your participation.

Best

Max on behalf of all organizers

Maxime Descoteaux, PhD
Full Professor
Research Chair in NeuroInformatics
Sherbrooke Connectivity Imaging Laboratory (SCIL)
Computer Science department
Sherbrooke University
2500 Boul. Université
Sherbrooke, Québec
J1K 2R1, Canada
http://scil.dinf.usherbrooke.ca

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20180120/ef7e15ef/attachment.html>


More information about the Neuroimaging mailing list