Using data from the UK Biobank (UKBB), we aim to improve our ability to predict metabolic diseases utilizing advanced deep-learning models. These models will help generate new incident cases, balance the data, and improve the accuracy of our predictions. With the extensive omics data available from the UKBB, we plan to identify and validate candidate biomarkers associated with metabolic diseases, providing insights into the underlying mechanisms and causes of these conditions.
Our project will also focus on developing risk prediction models using Explainable AI (XAI). This approach will not only predict the risk of metabolic diseases but also provide personalized interpretations, helping both clinicians and patients understand the factors contributing to the risk.
The project is planned to last for three years, with the possibility of an extension.
This research will have several public health benefits. It will train young researchers in handling larger datasets and using sophisticated analytical techniques. The findings are expected to lead to high-impact publications and deepen our understanding of metabolic pathways and disease mechanisms. Ultimately, the project aims to contribute to clinical practice by improving disease prediction and prevention, thereby enhancing public health outcomes.