Presentation

BARBANT – Boston And Rennes Brain image Analysis Team

Overview

BARBANT is an Inria associate team shared between Inria VisAGeS research team and the Computational Radiology Laboratory at the Boston Children’s hospital (Harvard Medical School). This associate team aims at better understanding the behavior of normal and pathological Central Nervous System (CNS) organs and systems. Pathologies of particular interest to us are multiple sclerosis, dementias, and pediatric diseases such as pediatric multiple sclerosis or tuberous sclerosis.

A major challenge is to characterize the future course of the pathological processes in each patient as early as possible in order to predict the progression of the disease and/or adverse neurological outcomes, and to develop better techniques for both monitoring response to therapy and for altering therapy (duration, dose and nature) in response to patient-specific changes in imaging characteristics. At term, this project will allow to introduce objective figures to correlate qualitative and quantitative phenotypic markers coming from the clinic and image analysis, mostly at the early stage of the pathologies. This will allow for the selection or adaptation of the treatment for patients at an early stage of the disease.

Research directions

  • Development of specific shape and appearance models, construction of atlases better adapted to a patient or a group of patients in order to better characterize the pathology
  • Advanced segmentation and modeling methods dealing with longitudinal and multidimensional data (vector or tensor fields), especially with the integration of new prior models to control the integration of multiscale data and aggregation of models
  • New models and probabilistic methods to create water diffusion maps from diffusion MRI, and their robust comparison between patients and control subjects to better understand the brain white matter structure and its alteration by pathologies
  • New machine learning procedures for classification and labeling of multidimensional features (from scalar to tensor fields and/or geometric features)

Partners

Comments are closed.