Abstract:Coronary heart disease (CHD) is a major global public health issue that poses a serious threat to human health. Despite advances in treatment improving the prognosis for some patients, hospital mortality rates remain largely unchanged, and significant disparities in treatment outcomes persist across regions and medical institutions. Accurate risk prediction models play a critical role in identifying high-risk populations, formulating personalized intervention strategies, and improving patient outcomes. Over the past 30 years, more than a dozen CHD prognosis prediction models have been developed worldwide. These models differ significantly in terms of target populations, predicted outcomes, selected predictors, and follow-up durations, posing various limitations and requirements for clinical application. This article systematically reviews CHD prognosis prediction models developed globally, analyzes their characteristics in terms of algorithm design, data processing, and clinical application. It also focuses on optimizing data integration methods, enhancing model interpretability, and improving cross-population validation strategies. The aim is to provide a scientific basis for precise prevention and management of CHD, offer robust support for clinical decision-making and public health management, and serve as a valuable reference for the development of medical models driven by multi-modal data.
