Speaker: Gael Varoquaux
Title: Brain mapping with machine learning: linking brain structures to cognitive function
Time: Oct 20, 2pm
Room: PNI 101
For decades, functional brain imaging has been accumulating images of brain activity in various cognitive tasks. Extracting from this data a clear, unambiguous, link between brain structures and cognitive functions faces the hurdles that activation maps overlap and cognitive tasks are hard to relate.
Considering the problem as relating two sources of high-dimensional data --brain images on the one side and cognition on the other-- machine learning can be a central tool for this endeavor. In an encoding setting, predicting brain activity from the task can explore how rich description of cognition map to the brain. In a decoding setting, it can narrow down on regions that elicit a particular behavior, an inference reversed compared to standard correlation analysis prevailing in brain mapping.
Yet, this research program must tackle challenges in machine learning and cognitive science. Predictive model must control well the error on the brain maps they estimate, which calls for dedicated regularization as in inverse problems. The analysis must encompass and describe a experiments that span the space of possible cognitive processes. I will describe our successes tackling these issues and linking brain structures to cognitive function.
BIO: Gaël Varoquaux is a tenured computer-science researcher at INRIA. His research develops 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 has a PhD in quantum physics and is a graduate from Ecole Normale Superieure, Paris.