MRC DTP in Precision Medicine

Neurocomputational signatures of response to antipsychotic treatment in first episode psychosis.

Supervisors 

Dr Filippo Queirazza, Health and Wellbeing, University of Glasgow 

Professor Marios Philiastides, Collage of Psychology and Neuroscience, University of Glasgow 

Professor Stephen Lawrie, Centre for Clinical Brain Sciences, University of Edinburgh

 

Summary

Patients suffering from a first episode psychosis (FEP) will typically present with hallucinations and delusions. The mainstay treatment for these patients is an antipsychotic medication. Unfortunately, only around 55-70% of patients will respond to antipsychotics in the first 12 months. Early identification of treatment response is therefore crucial to curtail patients’ suffering, healthcare costs and improve long-term outcome. At present there is no clinical biomarker that helps predict treatment response.

A novel and promising approach to biomarker discovery in psychiatry research is the development of theory driven biomarkers, that are embedded in the mechanisms underpinning core clinical manifestations of psychiatric illnesses. Here, we will use computational modelling and state-of-the-art brain neuroimaging (EEG) to illuminate the neurocomputational mechanisms underlying core psychotic symptoms such as auditory hallucinations in FEP patients. Crucially, we will capitalise on these mechanisms to classify response to antipsychotic medications at the individual level. To this end we will use advanced machine learning classification algorithms applied to neuroimaging data. Overall, this project involves computational modelling, advanced analysis of neuroimaging data and the use of state-of-the-art machine learning techniques, providing training in key skills for translational neuroscience and precision medicine.