xpertsystems/hccar006-sample

HC-CAR-006 — 外周动脉疾病合成队列（样本）是一个完全合成的、统计上真实的患者数据集，专注于外周动脉疾病（PAD）。该数据集涵盖疾病分期、血流动力学评估、症状/生活质量工具、伤口护理、血运重建手术和通畅性结果、截肢风险、主要不良肢体事件（MALE）、指南指导的药物治疗和心血管风险标志物。数据集设计为可直接用于分析、建模、演示和教育，同时保持100%合成，无真实患者数据、无个人健康信息（PHI）、无重新识别风险。数据集包含500名患者×115列，分为9个模块：人口统计学与合并症、疾病分期、血流动力学、跛行与生活质量、伤口护理、血运重建、截肢/MALE/WIfI/PREVENT III、医疗管理、实验室与心血管结果。数据格式为CSV，样本许可证为CC-BY-NC-4.0，完整产品为商业许可证。数据集经过严格校准，基于行业标准注册表和指南（如全球血管指南、ESC指南、AHA/ACC指南等），并提供了详细的验证报告。用例包括风险模型原型设计、通畅性生存分析、指南指导药物治疗差距分析、肢体挽救队列定义、健康经济建模、EHR ETL原型设计、教学培训以及血管SaaS和设备的演示验证。数据集还包含加载示例（如使用pandas或Hugging Face datasets库）和代码示例（如截肢风险模型和生存曲线分析）。局限性包括：样本为横截面数据（非纵向序列）、MACE率和死亡率被高估、技术成功率略低于现代标准、他汀治疗与LDL水平无真实相关性、无缺失数据建模等。完整商业产品提供更多患者、纵向访问、可配置队列混合、改进的数据耦合和定制注册表导出。

HC-CAR-006 — Peripheral Artery Disease Synthetic Cohort (sample) is a fully synthetic, statistically realistic cohort of patients with peripheral artery disease (PAD) covering disease staging, hemodynamic assessment, symptom/QoL instruments, wound care, revascularization procedures and patency outcomes, amputation risk, Major Adverse Limb Events (MALE), guideline-directed medical therapy, and cardiovascular risk markers. Built to be drop-in usable for analytics, modeling, demos, and education while remaining 100% synthetic — no real patient data, no PHI, no re-identification risk. The dataset includes 500 patients × 115 columns across 9 modules: Demographics & Comorbidities, Disease Staging, Hemodynamics, Claudication & QoL, Wound Care, Revascularization, Amputation / MALE / WIfI / PREVENT III, Medical Management, and Labs & CV Outcomes. Format is CSV, with sample license CC-BY-NC-4.0 and full product under commercial license. It is calibrated against industry-grade registries and guidelines (e.g., Global Vascular Guidelines, ESC PAD Guideline, AHA/ACC PAD Guideline) and includes a validation report. Use cases include risk model prototyping, patency survival analysis, guideline-directed medical therapy gap analytics, limb salvage cohort definition, health economics modeling, EHR ETL prototyping, teaching/training, and demo/proof-of-concept for vascular solutions. Loading examples are provided for pandas and Hugging Face datasets, along with code snippets for amputation risk modeling and survival curve analysis. Limitations include: sample is cross-sectional (not longitudinal), inflated MACE and mortality rates, technical success rates slightly below modern standards, no correlation between statin therapy and LDL levels, no missingness modeling, etc. The full commercial product offers more patients, longitudinal visits, configurable cohort mix, improved data coupling, and custom registry exports.