Chief Supervisor

Prof. HE Mingguang

Project Title

Synopsis

This project aims to investigate the association between ocular biomarkers —measured via high-resolution imaging techniques such as fundus photography and optical coherence tomography (OCT) —and individual health behaviors, leveraging artificial intelligence for integrated analysis. By examining structural changes of fundus vessels, the study seeks to identify reliable, non-invasive biomarkers reflective of modifiable health behaviors, including physical activity, dietary patterns, smoking, and alcohol consumption.

The integration of artificial intelligence, particularly deep learning approaches, enables the processing of complex multimodal oculomics data alongside behavioral metrics. This approach facilitates the development of robust predictive models that link behavioral patterns with early subclinical signs of chronic conditions such as diabetes, hypertension, and neurodegenerative disorders. The ultimate goal is to create a scalable digital health tool for continuous, non-invasive monitoring and personalized behavioral intervention strategies. This system aims to provide actionable insights to both individuals and healthcare providers, promoting early intervention and personalized strategies to prevent disease progression and improve long-term health outcomes.