I am senior research scientist at Inria, and team leader at the Paris Brain Institute (ICM) located at the Pitié-Salpêtrière hospital in Paris. I am fellow of the Paris AI Research Institute (PR[AI]RIE). Since 2022, I am co-founder and CEO of Qairnel, a spin-off company dedicated to making precision neurology available to everyone.

My research lies at the intersection of statistical learning and differential geometry with applications in neuroscience and clinical trial designs. We develop statistical methods for the analysis of structured data such as images or meshes. In particular, we invented the technique named "disease course mapping" for the analysis of longitudinal data, e.g. repeated data of the same subjects over time. This technique allowed the construction of personalised models of disease progression for several neurodegenerative diseases: Alzheimer, Parkinson or Huntington. These models are used to spot the optimal moment to test new therapies in each patient.

I received several awards including the MICCAI Young Investigator Award in 2008, the second Gilles Kahn Award for best dissertation in computer science in 2010. In 2015, I was awarded an ERC Starting Grant from the European Research Council. I was the first laureate of a Sanofi iDEA award outside the USA in 2019. In 2020, I received the Inria - Académie des Sciences young researcher award.



  • Editor of 6 archived conference proceedings (4 in the Springer LNCS series)
  • Author of more than 150 peer-reviewed publications, among which
    • methodological journal articles in the fields of neuroimaging (Neuroimage), image analysis (Medical Image Analysis, IEEE Trans. Medical Imaging, SIAM J. Imaging Sciences), computer vision (International J. of Computer Vision), medical journal articles in the fields of neurology such as JAMA Neurology,
    • peer-reviewed conference papers essentially in the field of medical image analysis (Medical Image Computing and Computer-Assisted Intervention (MICCAI), Information Processing in Medical Imaging (IPMI), International Symposium on Biomedical Imaging (ISBI)) and more occasionally in the fields of statistical learning and computer vision (Neural Information Processing System (NeurRIPS), Computer Vision and Pattern Recognition (CVPR))



Deformetrica: learn from shapes is a software project which provides statistical tools to analyze shape data using space deformations. Input data may be images, curve- or surface- meshes in 2D or 3D without the need for point correspondence. It can be used to build dynamical model of shape changes from longitudinal shape data sets. Brief introduction in this Medium post.

For more details, visit www.deformetrica.org


Leaspy: learn spatiotemporal patterns in python is a software project which allows the construction of personalized models of progression from longitudinal data sets, e.g. multiple subjects observed at multiple non-corresponding time-points. Data may be sets of biomarkers, or data distributed at the nodes of a fixed mesh. Brief introduction in this Medium post.

Fork me on gitlab!