Patients with coronary heart disease who take dual antiplatelet therapy face two serious risks: thrombosis and major bleeding. This study aims to develop better ways to predict these risks and guide personalized treatment.
The investigators will use a large, long-term follow-up study of Chinese patients with coronary heart disease. This research plans to discover new biomarkers related to clot and bleeding risk. The study will combine information from proteins, metabolites, sugars attached to proteins, genes, and medical images. Using machine learning methods, the investigators will identify the most important markers and test them in the patient group of this study.
The investigators will then build new risk prediction models that include these new markers together with traditional risk scores (such as GRACE, PARIS, and Precise-DAPT). This study will check whether these new models are better than existing ones at predicting who will develop clots or bleeding and at helping doctors decide on the best treatment for each patient.
The new aspects of this research are: (1) using advanced multi-omics technology to find novel markers specifically for Chinese patients; (2) combining clinical, biological, and imaging data to improve prediction accuracy; and (3) using machine learning to create more precise risk models.
The goal is to provide doctors with a more accurate tool to assess each patient's risk of clots and bleeding. This will help them choose the safest and most effective antiplatelet treatment, reduce serious complications, and improve patient care.