xpertsystems/hccar009-sample
收藏资源简介：
HC-CAR-009 心肌病纵向合成队列（样本）是一个完全合成的纵向数据集，模拟了500名心肌病患者在10年内的半年度随访。数据集涵盖6种心肌病亚型（扩张型心肌病DCM、肥厚型心肌病HCM、缺血性心肌病ICM、限制型心肌病RCM、致心律失常性右室心肌病ARVC、左室致密化不全LVNC），包含多模态数据：每次随访的超声心动图、每两年一次的心脏磁共振成像、生物标志物、症状评分（如KCCQ、6分钟步行试验、峰值摄氧量）、遗传学信息（致病基因、变异分类、多基因风险评分）、7种治疗策略（包括指南指导的药物治疗、设备治疗、移植）、合并症以及纵向结果（如心衰住院、心源性猝死、主要不良心血管事件、移植、治疗依从性）。数据集以CSV格式提供，共101列，分为9个模块：患者与访视元数据、超声心动图、心脏磁共振成像、生物标志物、症状与功能状态、遗传学、治疗反应、合并症与生命体征。该数据集基于真实注册研究和指南校准，适用于分类模型、纵向轨迹建模、风险预测、治疗优化分析等，但作为合成数据，存在一些生成器特性（如遗传阳性率偏高、亚型交叉等），需注意其局限性。样本数据采用CC-BY-NC-4.0许可，完整商业产品需联系XpertSystems.ai获取商业许可。
HC-CAR-009 — Cardiomyopathy Longitudinal Synthetic Cohort (sample) is a fully synthetic longitudinal panel dataset simulating 500 cardiomyopathy patients across 10 years of semi-annual follow-up. It covers 6 phenotypic subtypes (Dilated Cardiomyopathy DCM, Hypertrophic Cardiomyopathy HCM, Ischemic Cardiomyopathy ICM, Restrictive Cardiomyopathy RCM, Arrhythmogenic Right Ventricular Cardiomyopathy ARVC, Left Ventricular Non-Compaction LVNC) with rich multimodality data: echocardiography at every visit, cardiac MRI every 2 years, biomarkers, symptoms (KCCQ, 6-minute walk test, peak VO2), genetics (causal genes, variant class, polygenic risk score), 7 treatment arms (including guideline-directed medical therapy, device therapy, transplant), comorbidities, and longitudinal outcomes (e.g., heart failure hospitalization, sudden cardiac death, major adverse cardiovascular events, transplant, treatment adherence). The dataset is provided in CSV format with 101 columns across 9 modules: Patient & Visit Metadata, Echocardiography, Cardiac MRI, Biomarkers, Symptoms & Functional Status, Genetics, Treatment Response, Comorbidities & Vitals. Calibrated against real-world registries and guidelines, it is suitable for subtype classification models, longitudinal trajectory modeling, risk prediction, treatment optimization analytics, etc. As synthetic data, it has generator quirks (e.g., enriched genetic positivity, subtype crossover) that require consideration. The sample is licensed under CC-BY-NC-4.0, while the full commercial product requires a license from XpertSystems.ai.
