My research aims to develop an explainable machine learning framework to support early identification of cardiovascular risk in individuals with Type 2 Diabetes Mellitus (T2DM). Cardiovascular disease is a major complication of T2DM, yet early prediction remains challenging with traditional methods. This project will explore the use of large-scale health data from UK Biobank to build models that identify high-risk individuals, compare their performance with conventional risk scores, and apply explainable AI techniques (e.g., SHAP) to ensure transparency. The goal is to enable more accurate, timely, and personalized interventions, ultimately reducing preventable complications and improving long-term outcomes for people living with diabetes.
As part of this project, we are committed to responsible use of AI and ensuring that our research benefits the wider scientific and healthcare community. The outcomes of the study will be shared through peer-reviewed journal articles, conference presentations, and academic workshops.
To support transparency and reproducibility, we also plan to share relevant code, analysis pipelines, and trained models via trusted open platforms like GitHub-wherever it is permissible under UK Biobank’s data sharing policies. this contains how many character
No individual-level data from UK Biobank will be made public at any stage. All dissemination efforts will strictly follow the UK Biobank AI publication guidelines, ensuring ethical research conduct and meaningful public benefit.