Thesis title: Big Data Analytics for Early Diagnosis of Major Depressive Disorder
Supervisors: Dr Elaine Murray, Professor Martin McGinnity, Professor Tony Bjourson
In the UK, depression is predominantly diagnosed in primary care by general practitioners (GP’s), but diagnosis and treatment selection is largely subjective, and reliant on patient self-report and clinical judgment and experience. DSM-IV guidelines maintain that 5 out of 9 specific symptoms must be present for a minimum of two weeks, for a diagnosis of depression. Despite these standard guidelines, diagnosis and subsequent determination of episode severity is complex and variable. Based on DSM criteria there are over 200 possible ways to meet the criteria of symptoms of depression, and very little information to inform the best choice of treatment. Moderate to severe depression is predominantly treated with antidepressants, but this treatment is far from straightforward. Over 20 antidepressant medications are approved for clinical use and there is currently no empirical evidence to support treatment selection.
There are currently no validated biological markers for depression or response to treatment with antidepressant medications, but a number of candidate biomarkers have emerged. Development of a novel biomarker panel that could be integrated with clinical, physiological, behavioural, and environmental data to develop a decision tool which would allow clinicians to effectively diagnose and stratify patients with depression to determine the most appropriate medication for each individual would shorten the duration of untreated depression, help maintain compliance and ensure better treatment outcomes.My project focuses on the development of big data autonomous learning, computational intelligence techniques for major depressive disorder, combining the expertise in intelligent systems of the ISRC and the mental health and stratified medicine expertise of NI Centre for Stratified Medicine.