Models and Inference for Neuroimaging Data
Models and Inference for Neuroimaging Data

MIND is a joint Inria and CEA (NeuroSpin center) project-team whose overall scientific objective is to develop statistical approaches from a theoretical and computational point of view, and open software tools to study the functioning and structure of the brain from both a cognitive and clinical point of view. The field of neuroscience is currently facing many inferential and computational challenges. To do so, the team relies on several brain imaging modalities, such as functional and diffusion magnetic resonance imaging (MRI) at very high magnetic field to aim at high spatial resolution, and electrophysiology techniques, such as electro- and magneto-encephalography, which measure brain activity in real time. In order to extract the most relevant information from these data, the team pays particular attention to the resolution of inverse problems in 3 or even 4 dimensions, linear or non-linear, which when informed by biophysical models allow non-invasive access to quantitative parameters of interest in the brain. Furthermore, beyond answering the questions "where, when, and how can neural activity be identified with certainty?" MIND is interested in the analysis of the causes inducing the activity observed in a specific brain area. Answering these questions with the help of computer programs requires the development of advanced methods based on causal inference, logic, knowledge base representation and high-dimensional statistics.

Centre(s) inria
Inria Saclay Centre
In partnership with
Centre CEA-Saclay


Team leader

Marie Enee

Team assistant