Can machine learning be used to predict antidepressant use outcome longitudinally?

Supervisors:  Dr Rebecca Pearson, Bristol Medical School (honorary) Dr Ryan McConville, Department of Engineering Mathematics Dr Brit Davidson, Department of Electrical and Electronic Engineering Email: holly.fraser@bristol.ac.uk

‌General Profile:

I come from an interdisciplinary psychology and neuroscience background, and am interested in development of data science techniques to answer challenging questions within healthcare informatics. My interests include machine learning, network analysis, and artificial intelligence to model relationships within health and disease, and am particularly interested in data infrastructure around the mapping of long term health outcomes. My research uses the Avon Longitudinal Study of Parents and Children (ALSPAC) dataset to explore the long term health outcomes of people who take antidepressant medication, with a focus on employing novel machine learning techniques to gain insight into long-term management of mental health conditions and measurement of intervention efficacy over the lifespan. I am interested broadly in how we can employ intelligent systems to aid in characterisation of disease and diagnosis in mental health and other areas of clinical complexity, and am more specifically interested in antidepressant medication mechanisms of action and efficacy. I am also interested in exploring the relationship between hormonal contraceptive use and mental health outcomes specifically in women, and understanding hormonal phenotypes of depression, and how this might factor into medication efficacy.

Peripheral to my project, I am also broadly interested in the political landscape of health reform in the NHS, particularly in the factors affecting perceptions of technology and digitisation, and how a technological paradigm shift in healthcare may affect health systems and public policy at large