My research group is looking to fill a post-doc position on learning biomarkers from functional connectivity.
The challenge is to use resting-state fMRI at the level of a population to understand how intrinsic functional connectivity captures pathologies and other cognitive phenotypes. Rest fMRI is a promising tool for large-scale population analysis of brain function as it is easy to acquire and accumulate. Scans for thousands of subjects have already been shared, and more is to come. However, the signature of cognitions in this modality are weak. Extracting biomarkers is a challenging data processing and machine learning problem. This challenge is the expertise of my research group. Medical applications cover a wider range of brain pathologies, for which diagnosis is challenging, such as autism or Alzheimer’s disease.
This project is a collaboration with the Child Mind Institute, experts on psychiatric disorders and resting-state fMRI, as well as coordinators of the major data sharing initiatives for rest fRMI data (eg ABIDE).
Objectives of the project
The project hinges on processing of very large rest fMRI databases. Important novelties of the project are:
- Building predictive models that can discriminate multiple pathologies in large inhomogeneous datasets.
- Using and improving advanced connectomics and brain-parcellation techniques in fMRI.
Expected results include the discovery of neurophenotypes for several brain pathologies, as well as intrinsic brain structures, such as functional parcellations or connectomes, that carry signatures of cognition.
The analysis framework is based on algorithmic tools developed in Python (crucially, leveraging scikit-learn for predictive modeling).
We are looking for a post-doctoral fellow to hire in spring. The ideal candidate would have some, but not all, of the following expertise and interests:
- Experience in advanced processing of fMRI
- General knowledge of brain structure and function
- Good communication skills to write high-impact neuroscience publications
- Good computing skills, in particular with Python. Cluster computing experience is desired.
A great research environment
The work environment is dynamic and exiting, using state-of-the-art machine learning to answer challenging functional neuroimaging question.
The post-doc will be employed by INRIA, the lead computing research institute in France. We are a team of computer scientists specialized in image processing and statistical data analysis, integrated in one of the top French brain research centers, NeuroSpin, south of Paris. We work mostly in Python. The team includes core contributors to the scikit-learn project, for machine learning in Python, and the nilearn project, for statistical learning in NeuroImaging.
In addition, the post-doc will interact closely with researchers from the Child Mind Institute, with deep expertise in brain pathologies and in the details of the fMRI acquisitions. Finally, he or she will have access to advanced storage and grid computing facilities at INRIA.
Contact information: gael dotnospam varoquaux atnotspam inria dotnospam fr