Development and Validation of a Multimodal Artificial Intelligence Model for Early Prediction of Acute Myocardial Infarction

Development and Validation of a Multimodal Artificial Intelligence Model for Early Prediction of Acute Myocardial Infarction

Acute myocardial infarction (AMI) remains a leading cause of mortality worldwide. Although early revascularization has markedly improved short-term outcomes, the incidence of major adverse cardiovascular events after the index event remains unacceptably high, posing a formidable clinical challenge. Contemporary risk-stratification instruments rely predominantly on a restricted set of conventional clinical variables and therefore fail to capture the full spectrum of individual pathophysiological complexity. To overcome these limitations, the present investigation aims to develop a post-AMI prognostic model that integrates comprehensive multimodal data.

Inclusion Criteria:

1. Age ≥18 years old
2. Presented with myocardial infarction, including NSTEMI and STEMI
3. Undergoing successful percutaneous coronary intervention.

Exclusion Criteria:

1.Poor imaging quality or a data missing rate >30%