Skip to main content Skip to navigation

Predictors of Psychosis in the Clinical High Risk Population: A multimodal data science and brain imaging approach

Primary Supervisor: Dr Jack Rogers, School of Psychology

Secondary supervisor: Professor Rachel Upthegrove

PhD project title: Predictors of Psychosis in the Clinical High Risk Population: A multimodal data science and brain imaging approach

University of Registration: University of Birmingham

Project outline:

Prospectively identifying young people at clinical high risk (CHR) of psychotic disorders could allow the development of treatments that attenuate, postpone, or even prevent the onset of psychosis and improve clinical and functional outcomes. Yet, young people are currently identified as being CHR based on symptom profiles and clinical history only, with approximately one third of CHR individuals converting to a psychotic disorder over a 3-year period. Whilst considerably higher than in the general population and other clinical populations(Fusar-Poli et al., 2013), there remains substantial heterogeneity in the CHR population and associated clinical trajectories. Many different approaches are being used to enhance the predictive accuracy of CHR status, including clinical, neurocognitive, neuroimaging, neurobiological and genetic data. The ‘Trajectories and Predictors in the Clinical High Risk for Psychosis Population: Prediction Scientific Global Consortium (PRESCIENT)’ study aims to recruit a large sample of CHR patients aged 12 to 30 years to develop targeted prediction models that will substantially benefit clinical care, healthcare services and research in this field. The PRESCIENT network includes Orygen, the world’s largest translational research institute in youth mental health, as well as eight partnered sites across Europe and Asia, including the Institute for Mental Health, University of Birmingham and Forward Thinking Birmingham. This PhD study will utilise recruitment sources, prospective data collection and analytical approaches adopted as part of the PRESCIENT study to develop, test and validate predictive models of CHR status as a marker of illness severity and prognosis. This PhD programme will build on the PRESCIENT study to explore mediators of innate immune response and ask the question of whether these peripheral immune-related measures are associated with changes in brain-related biochemical and neurobiological mechanistic processes in CHR individuals.

Novel data will include electroencephalography (EEG) to provide a temporally precise measure of changes in neurophysiological activity using established paradigms in psychosis and CHR research. Dynamic changes in brain structure and function between CHR and later stages of psychosis will be assessed using measures of brain volume, magnetic resonance spectroscopy and functional connectivity. The predictive utility of biospecimens (blood/saliva samples) on disease progression in CHR individuals will be assessed, with alterations of immune protein levels (e.g. circulating interleukin-6 concentrations) and oxidative stress markers (e.g. glutathione) previously reported in patients with psychotic and mood disorders (Upthegrove & Khandaker, 2020; Murray et al., 2021). Data assessment will include clinical measures capturing the broad range of symptomology relevant to CHR status and outcomes, and neurocognitive assessment (e.g. attention tasks) found to be most relevant in previous CHR research. Multivariate (i.e. multiple assessment domains or levels of analysis) data-driven approaches utilising supervised machine learning will be used with the aim of validating clinical and neurobiological markers of CHR for psychosis with previous findings supplementing clinical ratings with structural neuroimaging shown to improve the predictive accuracy for functional outcomes in CHR individuals (Koutsouleris et al., 2018).

In addition to PRESCIENT study data the student will collect and analyse primary data exploring functional and structural brain changes and clinical symptom profiles in CHR individuals. The student will also test the external validity of prediction models generated from PRESCIENT and primary data against existing models in the field (e.g. PRONIA). The student working on this project will have the opportunity to collaborate with researchers from the global PRESCIENT network on predictive models of CHR and train with the Orygen centre.

References:

  1. Fusar-Poli P, Borgwardt S, Bechdolf A, et al. (2013) The psychosis high-risk state: a comprehensive state-of-the-art review. JAMA psychiatry;70(1):107-120.
  2. Upthegrove R, Khandaker GM. (2020) Cytokines,Oxidative Stress and Cellular Markers of Inflammation in Schizophrenia. Current topics in behavioral neurosciences; 44:49-66.
  3. Murray AJ, Rogers JC, Katshu MZUH, Liddle PF, & Upthegrove R (2021) Oxidative Stress and the Pathophysiology and Symptom Profile of Schizophrenia Spectrum Disorders. Psychiatry: 12:703452.
  4. Koutsouleris N, Kambeitz-Ilankovic L, Ruhrmann S, et al. (2018) Prediction Models of Functional Outcomes for Individuals in the Clinical High-Risk State for Psychosis or With Recent-Onset Depression: A Multimodal, Multisite Machine Learning Analysis. JAMA Psychiatry;75(11):1156-1172.

BBSRC Strategic Research Priority: Understanding the Rules of Life: Immunology & Neuroscience and behaviour

    Techniques that will be undertaken during the project:

    • Advanced multivariate analyses (e.g., machine learning, AI-based techniques, network analysis).
    • Working with varied neuroimaging methods (e.g. electroencephalography (EEG), magnetoencephalography (MEG), structural and functional MRI, magnetic resonance spectroscopy (MRS))
    • Computer programming (e.g. MatLab, R, Python, Shell)
    • Data science and working with big data

    Contact: Dr Jack Rogers, University of Birmingham