Using multimodal normative modelling to predict clinical outcomes in early psychosis

Background: Individuals at the early stages of psychosis already suffer from alterations in their brain anatomy and function, and behaviour (e.g. cognitive skills such as memory and attention). However, it is not clear how these changes interact with each other and, more importantly, how this interaction may help predict who will become unwell. Approach: In this study, I propose to address this issue in three main steps. First, I will identify previously hidden brain-behaviour relationships in healthy individuals. Second, I will develop a model that captures the normative pattern of these brain-behaviour relationships and estimate by how much each individual patient deviates from this pattern. Third, I will then use these deviations to make predictions about each patient's illness course. Impact: Findings from my study will help to better understand brain-behaviour changes in psychosis and how we can use this information to bring personalised care to psychosis patients.