Development and Prospective Validation of an AI Model for Prognosis in ITP Patients Undergoing Coronary Revascularization

This study employs a dual-cohort design to develop and validate a prognostic model for Major Adverse Cardiovascular Events (MACE) following revascularization in immune thrombocytopenia (ITP) patients with Coronary Artery Disease (CAD). The model will be developed and trained using a retrospective multi-center cohort (development/training cohort). Its performance will then be prospectively validated in a separate, consecutively enrolled prospective cohort (validation cohort). The goal is to create an AI-based tool to assist in personalized risk assessment and decision-making for this high-risk population.

Study Design: This is a dual-phase, multi-center observational study. Phase 1 (Retrospective Cohort): A retrospective cohort will serve as the development and training set. Data from eligible patients treated in the past will be collected to identify predictors and develop the initial AI prediction model. Phase 2 (Prospective Cohort): A prospective, observational cohort will serve as the validation set. Consecutively eligible patients will be enrolled and followed forward in time. The model derived from Phase 1 will be applied to this cohort to evaluate its predictive accuracy and clinical utility. Treatment Groups: Within both cohorts, patients will be categorized based on actual clinical care: 1. Revascularization Group: Patients undergoing PCI or CABG. 2. Medical Therapy Group: Patients managed with guideline-directed medical therapy alone. Objective: To compare MACE risk between groups and to develop and validate a model predicting MACE specifically in the revascularization group.