You are here: indexactivitiestheme1dbscomp

DBS-comp project

The DBS-comp project: Computational Imaging and modeling for studying anatomo-­functional correlation to optimize deep brain stimulation

The objective of this project is to develop an innovative computational medical imaging strategy for optimizing clinical outcome of Deep Brain Stimulation (DBS) for Parkinson’s disease.


Rennes MediCIS

Paris ICM

  • Eric Bardinet
  • Sara Fernandez Vidal
  • Jerome Yelnik
  • Marie-­Laure Welter
  • Carine Karachi

Paris Neurospin

  • Cyril Poupon
  • Achille Teillac
  • Justine Beaujoin

Boston BWH

  • Sonia Pujol


The objective of this project is to develop an innovative computational medical imaging strategy for optimizing clinical outcome of Deep Brain Stimulation (DBS) for Parkinson’s disease. We aim at developing models that outline relationships between structure and function of the basal ganglia and its surroundings based on population and patient specific analysis of multimodal data. Knowing from post operative images where the electrode and the activated contact are, fiber tracts activated by DBS will be correlated to function and clinical data including electro-­physiological signals acquired during surgery and clinical scores. For such purpose, we will propose a new adapted tractography approach from Diffusion Tensor Imaging. Additionally, we will study non-­linear machine learning approaches to express such correlation and validate it on clinical data sets. The resulting models will allow both prediction and understanding of clinical outcome and side effects.

Crossings of the pyramidal pathway with subthalamo-pallidal and pallido-subthalamic fibers (MC-det approach) in HCP subject 100307. Source: In vivo Exploration of the Connectivity between the Subthalamic Nucleus and the Globus Pallidus in the Human Brain Using Multi-Fiber Tractography, Pujol et al.

inserm rennes1 ltsi