Development and Validation of a Cardiac Magnetic Resonance-Based Multimodal Deep Learning Model for Long-Term Outcome Prediction in ST-Segment Elevation Myocardial Infarction

Objective: This study aims to develop and test a novel artificial intelligence-based prediction model. This model will integrate cardiac magnetic resonance imaging and clinical data to predict the long-term risk of major adverse cardiovascular events in patients who have undergone emergency percutaneous coronary intervention for ST-segment elevation myocardial infarction. Description: This study plans to enroll patients with STEMI who have received primary PCI. Approximately one week after the procedure, patients will undergo a cardiac magnetic resonance scan. Concurrently, we will collect patients' basic information, blood test results during treatment, and procedural records. Thereafter, patients will be followed up regularly (every six months) to record the occurrence of any major adverse cardiac events, such as cardiovascular death, recurrent myocardial infarction, hospitalization for heart failure, or unplanned repeat revascularization.All collected data, including clinical data and analyzed cardiac MR images, will be used to construct a multimodal deep learning model. The model will learn to identify features associated with future cardiac problems. The accuracy of the model will be tested and validated in different patient groups. Potential Impact: If successful, this prediction tool could assist physicians in identifying high-risk patients earlier and more accurately, enabling closer monitoring and more timely interventions, ultimately improving the long-term prognosis for these patients.