New Imaging and GABA Spectroscopy Modalities to Localize the Epileptic Zone – EPI-CATCHER

In drug-resistant epileptic patients who may benefit from a surgical resection, a poor localization of the epileptic zone (EZ) may lead to neurological sequelae. We have recently developed two new approaches to identify the EZ in a mouse model of focal epilepsy: the first is based on GABA NMR spectroscopy and the second on quantitative multiparametric imaging. We hypothesize that one of these methods or the integration of both could help detecting more accurately the EZ in epileptic patients. We therefore propose in this project (i) to evaluate at the preclinical level the influence on our localization methods of antiepileptic drugs usually taken by epileptic patients, (ii) to develop mathematical tools based on machine learning integrating the data , and (iii) to conduct a clinical evaluation involving healthy volunteers and patients with mesial-temporal lobe epilepsy.
The project is structured in two WorkPackages (WP):

The first WP will characterize the impact of the most commonly used anti-epileptic drugs (AED) in a mouse model of mesiotemporal epilepsy on the proposed mqMRI and GABA-MRS biomarkers. EEG recordings and biological analyses (histology, molecular biology) will also be performed.

The second WP will implement computer tools for data analysis and acquisition of reference data in healthy volunteers and in epileptic patients who are candidates for surgical resection (mesial-temporal lobe epilepsy). Existing machine learning tools will be adapted to the mqMRI and GABA-MRS data to produce a tool capable of contouring the EZ.

At the end of this project, we expect to develop methods for acquiring and analysing MRI data to better delimit the EZ and thus fit into the pathway of patients with mesiotemporal epilepsy who are candidates for surgery. This method could then be evaluated in other forms of epilepsy but also in other brain diseases.