The capacity of visualising in vivo the cerebral structures that are involved in man during the performance of cognitive tasks are modifying human cognitive sciences in depth. For decades, study of human cognition was relying on the lesions method that investigates anatomical lesions and the corresponding neuropsychological symptoms. The development of functional imaging techniques, milestoned by increasing progress in space-time resolution, has led to the birth of a new field: cognitive neuroscience with imaging as the main tool. This field brings a new vision regarding relations between brain and mind as defined in psychology or in neuropsychology.
Functional neuroimaging is a multiple research field, where signal and image processing is necessary to rebuild and interpret data that are collected by instruments. Availability of reliable and robust analysis methods and of tools of easy use is a considerable asset for all cerebral imaging users, allowing them to interpret and present their results. In imaging centers of international level, large teams are developping signal and image processing techniques to deliver users within their own centers, then across international community. These tools often become community standards and make these centers world famous.
This workshops objective is to present the works of the IMPARABL team (IMAgerie PARamétrique de lActivation et du métaBolisme cérébraL) that stand on the interface between mathematics, brain imaging and its applications to cognitive and clinical neuroscience.
The IMPARABL team implements mathematical methods to analyze and quantify neuronal processes such as:
- the localisation of distributed neuronal networks in fMRI using generalised linear model and Bayes theory.
- the localisation of distributed neuronal networks in EEG using maximum entropy on the mean method, hierarchical Bayesian model and graphical models.
- the temporal properties of distributed neuronal networks using biophysical models of hemodynamics, metabolics and neural activity.
- the study of functional connectivity of neuronal circuits involved in cognitive processes using (fMRI, fMRI/EEG) acquisitions and mathematical models (conditionnal graphs and Bayesian theory of graphs, bootstrap).
The IMPARABL team implements programming tools to visualise and perform multimodal fusion in order to :
- integrate visualisation and analysis programs within the brainVisa system.
- develop user friendly interfaces to analyse multimodal data.
The IMPARABL team deals with cognitive and clinical applications :
Based on the above described methodology as a whole, the IMPARABL team focuses on the cerebral plasticity space and time modelling that is associated with :
- the training, strengthening and retention of motor abilities
- surgery of intracerebral low grade glioma.
The IMPARABL team of the Institut national de la Santé Et de la Recherche Médicale (INSERM), unit 678, is part of the agreement between the Centre de Recherches Mathématiques (CRM) and the Unité de Neuroimagerie fonctionnelle(UNF) of the Centre de Recherche de lInstitut Universitaire de Gériatrie de Montréal (CRIUGM).