AI approaches towards personalised medicine and Intergenerational life-course evolution of diabetes

Purpose and goal:

Diabetes, a global health crisis, is projected to affect 783 million people by 2045. We aim to apply AI-ML methods to refine diabetes sub-classification in Swedish and Indian populations, identifying those needing intensive treatment to prevent complications. Our objective is to then identify T2D and subtype specific biomarkers to improve treatment preferences. We then aim to modeling life-course trajectories to help uncover pre-clinical pathways towards primordial prevention.

Expected results and effects:

We will identify individuals at highest risk of diabetes and comorbidities. We will determine the applicability of Swedish study-derived coordinates for diabetes subgrouping in Indians. By comparing T2D and subtype-specific biomarkers, we’ll gain insights into distinct etiologies and treatment preferences across these diverse populations. Modeling life-course trajectories will provide invaluable insights towards primordial prevention.

Approach and implementation:

Polygenic scores pertaining to diabetes traits and comorbidities and birth parameters will be assessed and compared in both populations. Novel clustering approaches using clinical measures and genetics will be explored. -Omics biomarkers in T2D and subgroups will be testes to better understand pathophysiology and possible implications for treatment. Life course modeling will be performed for promoting primordial-primary prevention strategies.