Approved Research

Multimodal AI-Powered Models for Type 2 Diabetes Risk Prediction: A Step towards Precision Medicine

Lay summary

Type 2 Diabetes (T2D) is a global concern, with almost half a billion people affected by it. The UK Biobank aims to improve the prevention, diagnosis, and treatment of various diseases, and our proposed research project aligns with this goal. Our research project seeks to develop an advanced artificial intelligence (AI) model that can predict the risk of T2D using multiple sources of data, including electronic health records, imaging data, and omics data.

Our project expects that by fusing multimodal data, we can capture the complexity of T2D and provide a more accurate prediction than existing models. Early prediction of T2D can play a vital role in preventing the onset and progression of the disease, improving quality of life, and reducing healthcare costs. Our goal is to develop a robust AI model that can provide personalized risk assessment and individualized preventative strategies. Our research project will last for three years, with the first year dedicated to data acquisition, preprocessing, and cleaning. The second year will focus on model development, validation, and interpretation, while the third year will be for writing and publishing our results. We aim to identify novel biomarkers for T2D risk prediction and inform future research in the field of precision medicine. Our project's public health impact is significant, as it has the potential to improve T2D prediction accuracy and personalize risk assessment, leading to improved preventive strategies and reduced healthcare costs.