xpertsystems/hconc003-sample

这是一个合成的前列腺癌队列数据集，名为“HC-ONC-003 — 前列腺癌合成队列（样本）”，来自XpertSystems.ai合成数据工厂的肿瘤学垂直领域。数据集包含500名患者的队列数据（100列）和10,000行的纵向PSA面板数据（6列），覆盖了从非常低风险到转移性风险的完整NCCN风险谱。数据包括：PSA轨迹（基线、最低点、倍增时间、速度、BCR）、格里森/等级组病理学及共病理特征（筛状结构、IDC-P、PNI、LVI、ECE、SVI、PSM、淋巴结）、多模式治疗序列（主动监测、RP、EBRT/IMRT/SBRT/近距离放疗、ADT联合NHA、化疗、HIFU、Lu-PSMA、Ra-223、PARP、sipuleucel-T）、分子生物标志物（BRCA1/2/ATM/CDK12 HRR突变、AR-V7、PSMA表达、MSI-H、TMB、AR扩增、NEPC）、长期结果（BCR、MFS、CRPC进展、CSS、OS——基于Weibull校准到里程碑试验）、EPIC-26生活质量领域，以及每位患者20次半年一次的PSA面板（10年随访）。数据集完全合成，无真实患者数据，无PHI，无重新识别风险，适用于分析、建模、演示和教育。

A fully synthetic prostate cancer cohort dataset, named HC-ONC-003 — Prostate Cancer Synthetic Cohort (sample), from the XpertSystems.ai Synthetic Data Factory — Oncology vertical. It includes a 500-patient cohort (100 columns) and a 10,000-row longitudinal PSA panel (6 columns), spanning the complete NCCN risk spectrum from Very Low to Metastatic. Data covers: PSA trajectories (baseline, nadir, doubling time, velocity, BCR), Gleason/Grade Group pathology with co-pathology features (cribriform, IDC-P, PNI, LVI, ECE, SVI, PSM, lymph node), multimodal treatment sequences (active surveillance, RP, EBRT/IMRT/SBRT/brachytherapy, ADT with NHA, chemo, HIFU, Lu-PSMA, Ra-223, PARP, sipuleucel-T), molecular biomarkers (BRCA1/2/ATM/CDK12 HRR mutations, AR-V7, PSMA expression, MSI-H, TMB, AR amplification, NEPC), long-term outcomes (BCR, MFS, CRPC progression, CSS, OS — Weibull-calibrated to landmark trials), EPIC-26 quality-of-life domains, and a 20-visit semi-annual PSA panel per patient (10 years of follow-up). The dataset is 100% synthetic, with no real patient data, no PHI, and no re-identification risk, designed for analytics, modeling, demos, and education.