Approved research

Application of deep learning for investigating various biomarkers related to retinal images

Lay summary

Aims: The primary objective of this research is to utilize fundus photography and OCT scan data to analyze and predict prospective or existing cardiovascular, cerebrovascular, or renal diseases. The central focus is developing a deep learning model based on retinal images, which will contribute to the anticipation of cardiovascular, cerebrovascular, or renal diseases risks and aid in the early diagnosis and prevention of these disorders. This study aims to harness the potential of retinal imaging and advanced deep-learning techniques to enhance our understanding of heart, brain, kidney health and improve clinical outcomes through proactive interventions.

Scientific raionale: This study investigates the application of deep learning technology through retinal imaging to examine the correlation between cardiovascular, cerebrovascular, or renal diseases and retinal status. Previous research has raised the hypothesis of a link between the retina and these non-ophthalmic diseases, highlighting the potential for using retinal images in predicting theses diseases. However, more studies are needed to explore this association comprehensively. This research endeavors to bridge this scientific gap by developing highly automated models and assessing their practicality in clinical settings. It seeks to advance our understanding of the relationship between retinal health and theses disorders while exploring the clinical applications of this innovative approach.

Project duration: 36 months

Public health impact: The paramount public health impact of this research lies in the early diagnosis and prevention of cardiovascular, cerebrovascular, or renal diseases using retinal images. The developed deep learning model allows for a more precise measurement of a patient's biological age, facilitating the provision of personalized preventive strategies. This approach is expected to enhance disease early detection and patient management efficiency within the public health system. Furthermore, this research can potentially promote a shift toward a prevention-oriented healthcare system, supporting healthier aging and alleviating the burden of chronic diseases.