[BIC-announce] [TODAY at 1 pm] Reminder: Dr Gaël Varoquaux @ BIC - Dec 14

[Zografos Caramanos, Mr]

Subject: Reminder: Dr Gaël Varoquaux @ BIC - Dec 14

Dear All:

It is my great pleasure to announce the visit of Dr. Gaël Varoquaux (INRIA, http://gael-varoquaux.info) at the BIC, Dec 14.

Gaël is a tenured computer-science researcher at Neurospin/INRIA (France). His interests are in statistical learning tools for functional neuroimaging data with application to cognitive mapping of the brain as well as the study of brain pathologies. In addition, he is heavily invested in software development for data science, as project-lead for scikit-learn, one of the reference machine-learning toolboxes, and on joblib, Mayavi, and nilearn. Varoquaux has contributed key methods to learn functional brain atlases and connectome structure from task-based and rest fMRI, and methods for statistical mapping and decoding of functional brain imaging. He holds a PhD in quantum physics and is a graduate from Ecole Normale Superieure, Paris.

Dr. Varoquaux will deliver a BIC Lecture that day (Dec 14 @ 1pm, MNI de Grandpré - see below): "Methods for Resting-State Connectome Biomarkers" and will be available for further discussion. If you would like to meet with Gaël in person, feel free to contact him directly ( gael.varoquaux at inria.fr<mailto:gael.varoquaux at inria.fr>)

Cheers,

Sylvain.

[cid:09D08E94-892E-416F-BD4D-7088F0A18B58]

Resting-state fMRI is a promising source of functional biomarkers as, unlike typical fMRI paradigms, it can be applied to all subject and patient populations. I will discuss our efforts on understanding the different modeling steps in an inter-subject connectome classification pipeline e.g., to predict subject phenotypes. Namely, the questions are: How to define nodes, or
functional brain regions? How to measure functional connectivity in a subject? How to compare it across subjects? How to build predictive models? I will discuss theoretical and experimental validation of each step. In particular I will review linear decompositions (such as ICA and dictionary learning) and clustering to choose nodes, and various inverse covariance estimators to estimate graphs.
Validating these choices is challenging, as they are based on assumptions on the data. Based on our understanding of the various steps, we have built a full pipeline that predicts Autism from rest-fMRI on unseen scanning site in the ABIDE dataset. To our knowledge, this is the first prediction of a clinically-relevant diagnosis status that carries over in inhomogeneous acquisitions settings. This full-blown experiment, on 871 subjects, also highlights what the important choices are in a population-level connectome analysis.

Sylvain Baillet, PhD

Professor, Neurology, Neurosurgery & Biomedical Engineering
Acting Director, McConnell Brain Imaging Centre
MNI Killam and FRQS Senior Scholar
Montreal Neurological Institute
McGill University
http://mcgill.ca/bic

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bic.mni.mcgill.ca/pipermail/bic-announce/attachments/20151214/26ef2331/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: varoquaux.jpg
Type: image/jpeg
Size: 20201 bytes
Desc: varoquaux.jpg
URL: <http://www.bic.mni.mcgill.ca/pipermail/bic-announce/attachments/20151214/26ef2331/attachment-0001.jpg>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: ATT00001.txt
URL: <http://www.bic.mni.mcgill.ca/pipermail/bic-announce/attachments/20151214/26ef2331/attachment-0001.txt>