P. McGuire, United Kingdom

King's College London Department of Psychosis Studies

Moderator Of 1 Session

Sun, 11.04.2021
Session Time
10:00 - 11:30
Channel 4
Session Description
The Live Q&A of this session will take place in the Live Sessions auditorium. Please refer to the interactive programme for the exact time and channel.

Proposed by the EPA section on Neuroimaging -One of the major limitations of current therapeutic management of psychoses is the lack of predictive, personalised medicine tools that could inform clinicians’ as choice of treatment for individual patients, aiming to improve functional outcomes while preventing adverse metabolic side effects (e.g., weight gain, metabolic syndrome, diabetes). In current clinical practice, poor efficacy or adverse side effects of treatments can present months after commencement of treatment. Even if therapy is adjusted, it might already be too late for the patient to fully recover from such comorbidity. Therefore, prompt identification of a patient’s risk profile is essential for selecting an optimal preventative therapeutic strategy. In a personalised medicine approach to disease treatment and prevention of comorbidities, a patient would first undergo a comprehensive screening by a range of diagnostic tools, which would predict the patient’s mental, functional and somatic outcomes given various lines of treatment, and thus help identify the optimal treatment strategy. Recent research using molecular profiling approaches (such as metabolomics) and neuroimaging suggests that such prediction of patient outcomes, even in individuals at clinical high risk for psychosis, may be feasible. The aim of this Symposium is to cover recent advances in the domain of outcome prediction, with specific focus on use of high-dimensional and multi-modal data such as from ‘omics’ and neuroimaging.

Session Icon
Pre-Recorded with Live Q&A, Section

Presenter Of 1 Presentation

Symposium: Predicting the Outcomes in Psychosis: Recent Advances in Molecular Profiling, Neuroimaging and Machine Learning (ID 225) No Topic Needed

Live Q&A