Research Question and Objective
Our goal is to develop and apply advanced machine learning (ML) models to predict the likelihood of treatment resistance (TR) and treatment outcomes in major psychiatric disorders. We also aim to understand the relationship between multi-omics, real world evidence data such as clinical, and environmental factors in individuals suffering from TR during the treatment of schizophrenia (SCZ), bipolar disorder (BD), and major depressive disorder (MDD). Specifically, our focus includes:
* Multi-omics integrating, include proteomics and genetics.
* Real world evidence data such as electronic health records data
* Environmental and lifestyle factors influencing treatment response.
By utilizing ML techniques and algorithms, we seek to enable early identification of patients at risk of TR and facilitate personalized treatment approaches to improve outcomes in severe mental illnesses.
Scientific Rationale
Major psychiatric disorders such as SCZ, BD, and MDD are among the leading causes of disability worldwide. SCZ, BD and MDD share a relevant genetic component, which provides the possibility for trans-diagnostic analyses of the biological bases of such disease (Analysis of shared heritability in common disorders of the brain | Science).
Treatment resistance (TR) is a significant challenge in the treatment of such psychiatric disorders. TR affects 20-60% of patients and resulting in poor outcomes, including increased suicide risk and healthcare costs (Treatment resistance in psychiatry: state of the art and new directions – PMC). The traditional trial-and-error approach to treatment is inefficient, leading to delays in finding effective interventions. This situation highlights the need for predictive models to guide early, personalized, and effective treatments.
Recent advances in genetics and multi-omics research provide opportunities to identify biomarkers for TR and treatment outcomes.