As a leader of the team “Signatures of brain disorders” at NeuroSpin, CEA, Université Paris-Saclay, France, I supervise the design of machine learning and statistical models to uncover neural signatures predictive of clinical trajectories in psychiatric disorders. To unlock the access to data required by learning algorithms, I oversee the data management, calculation, and regulation (GDPR) of large-scale national and European initiatives.
Since 2003, he has been designing machine learning models to discover brain imaging signatures of mental disorders. He explored dimension reduction and regularization strategies to overcome the “curse of dimensionality” caused by many neuroimaging measurements. In 2019, he obtained a chair in Artificial Intelligence to develop transfer learning algorithms to bridge the gap between big (heterogeneous) and small (homogeneous) datasets.