AI-Driven Dynamic Prediction of Non-Target Lesion Progression After PCI: A Chinese Multicenter Cohort Study

DeVelopment of an Artificial Intelligence-Driven Dynamic Prediction and Precision Stratification System for Coronary Non-Target LesIon ProgressiON After Percutaneous Coronary Intervention: A Multimodal Data-Enabled Multicenter Cohort Study in China

Coronary artery disease remains a leading cause of global mortality. Although percutaneous coronary intervention (PCI) improves patient outcomes, the long-term risk of major adverse cardiovascular events (MACE) driven by the progression of non-target lesions (NTLs) remains substantial and continues to increase, while current risk stratification tools remain inadequate for predicting NTL progression. This multicenter cohort study aims to develop an artificial intelligence (AI)-driven system for the dynamic prediction and precision stratification of NTL progression after PCI. Utilizing comprehensive multimodal data from 52,577 Chinese patients-including clinical profiles, multi-omics blood biomarkers, and coronary imaging-the research pursues three primary objectives: (1) to identify and validate 2-3 specific biomarkers for NTL progression risk using multi-omics approaches; (2) to construct an integrated risk assessment and early-warning system by applying machine learning to multimodal data for predicting NTL progression and MACE; and (3) to establish metabolic and imaging-based subtypes to create a precision management system that optimizes secondary prevention strategies by identifying specific high-risk populations. This study is expected to provide a novel tool for accurate identification of high-risk patients and personalized post-PCI management, ultimately aiming to improve long-term prognosis.

Inclusion Criteria:

1. Aged 18 years or older;
2. Scheduled for or having undergone PCI;
3. Baseline plasma sample obtainable;
4. Informed consent obtained

Exclusion Criteria:

1. Pregnancy or lactation;
2. Severe hepatic or renal dysfunction;
3. Active autoimmune disease;
4. Missing critical data