Machine learning for predictive multi-modal markers of neuropsychiatric syndromes

The project seeks to develop reliable machine learning tools for clinical diagnosis of neuropsychiatric disorders, in particular the presence of psychotic symptoms, using ‘deep sampling’ with multimodal brain imaging data (structural, functional, and diffusion MRI, simultaneous near infrared spectroscopy (NIRS)-electroencephalogryphy (EEG)), biomarker data, and genetic data. This involves psychiatric populations including first and multiple episode psychosis. Experimental settings include resting-state and social interaction paradigms. This project is a great opportunity to impact clinical practice through better technology.