Reminder: this project studies markers of emotional regulation in a large sample of patients under MRI. The goal is to identify sub-categories of bipolar disorder and emotional dysregulation according to the disease’s progression stage, onset age, presence of psychotic symptoms, association with psychiatric or somatic disorders, and clinical variables such as childhood stress level. Innovative multivariate methods of analysis link clinical variables to brain function. This involves identifying biological correlates of emotional dysregulation, a mechanism present in several conditions such as bipolar disorder and emotionally labile personality disorder.

Since 2017, Dr Camille Piguet of the HUG in Geneva and Dr Josselin Houenou, Neurospin & CHU Henri Mondor in Paris and their teams have combined their skills and data. They were able to collect rest state data from 181 subjects, which allows them to identify significant correlations between the variability of the MRI signal and this dimension of emotional dysregulation. Some regions in particular, such as the prefrontal ventromedial cortex and the hippocampus on the right, seem to be associated with these difficulties. This year, these results were presented at the annual congress of the World Federation of Societies of Biological Psychiatry (June 2019, Vancouver), as part of a symposium organized by Professor Benicio Frey (Mc Master University, Canada).

In addition, the database has been extended, in the future this will make it possible to analyse in particular the structural connectivity and depth of brain furrows, always in relation to clinical variables. Finally, thanks to “machine learning” techniques – artificial intelligence – we will be able to analyze this data in order to establish a diagnosis at the individual level (personalized medicine) and identify homogeneous subgroups of patients on the basis of their “functional/structural” connectivity profile.