Myocardial infarction (MI) remains a major threat to human health. Although interventional treatment techniques have advanced rapidly, many patients still experience major adverse cardiovascular events (MACE) and require hospital readmission after discharge. Artificial intelligence (AI) based on wearable device data has shown great potential in the diagnosis and management of cardiovascular diseases.

This study aims to explore the clinical value of wearable device-based data analysis and AI-driven risk stratification models in post-discharge management of acute myocardial infarction (AMI) patients.

This prospective, open-label, randomized controlled study aims to evaluate the clinical benefits of wearable device-based AI risk models in post-discharge management of AMI patients. A total of 200 patients who have undergone PCI and provided informed consent will be enrolled, including those with both preserved and reduced left ventricular ejection fraction (LVEF).

Participants will be randomly assigned to either the control group or the intervention group in a 1:1 ratio. All patients will be equipped with a wearable smartwatch and continuously monitored for 3 months after discharge. Data collected will include physiological signals, sleep and activity parameters. In both groups, patients will receive weekly telephone follow-ups and monthly office visits to record symptoms, medication use, and adverse events.

In the intervention group, wearable data and AI analytical results will be made available to both patients and their physicians. These insights will be discussed during follow-ups and used to support lifestyle modification, medication adjustment, and clinical decision-making. In the control group, AI data will be collected but not shared or used for clinical management during the study period.

The primary study endpoint is the time to first unplanned hospital readmission within 3 months, including readmissions due to chest pain, heart failure, arrhythmia, recurrent myocardial infarction, or death. The secondary endpoints include: Change in Kansas City Cardiomyopathy Questionnaire-12 (KCCQ-12) score from baseline to 3 months; change in left ventricular ejection fraction (LVEF) measured by echocardiography between baseline and 3 months.

The investigators hypothesize that AI-assisted, wearable-based monitoring and feedback will improve early detection of adverse cardiovascular events, reduce unplanned hospitalizations, increase LVEF in patients with reduced LVEF at discharge, and enhance quality of life compared with standard post-discharge care.

Inclusion Criteria:

• Adults aged 18 to 75 years.
• Confirmed diagnosis of acute myocardial infarction (AMI), including both ST-segment elevation myocardial infarction (STEMI) and non-ST-segment elevation myocardial infarction (NSTEMI).
• Underwent successful percutaneous coronary intervention (PCI) during index hospitalization.
• Hemodynamically stable at the time of hospital discharge.
• Willing and able to wear a smartwatch continuously for the study period.
• Compatible with the data collection application and have stable internet access.

Exclusion Criteria:

• Planned staged or elective PCI or any coronary revascularization scheduled within 3 months after discharge.
• Unable to tolerate or contraindicated for wearing metal or electronic monitoring devices.
• Pregnant or breastfeeding women.
• Residence in an area without stable network connectivity or inability to use a smartphone for data upload and communication.
• Severe comorbidities that limit 3-month survival or follow-up.