Tumor Growth & Chemotherapy Response Cohort — Sample (100 Patients × 300 Regimens)

Full dataset page: https://sentineldata.com.ua/dataset/tumor-growth-chemotherapy-response-cohort-10k This is a free stratified sample of the full dataset (v1.0) — not the complete dataset. The full dataset covers 10,000 patients × 3 counterfactual arms. This Zenodo record contains only 100 patients (first 100 patient IDs, sorted) × all 3 arms — a true row-filtered subset of the delivered dataset parquet files, not a re-generation. Sample Composition: Patients: 100 Regimens (3 arms): 300 Outcomes: 300 Profile rows — MTD arm: 24,500 Profile rows — dose-dense arm: 24,500 Profile rows — metronomic arm: 24,500 Total profile rows in sample: 73,500 Dataset Architecture & Specifications The Tumor Growth & Chemotherapy Response Cohort is a fully synthetic, patient-level oncology dataset built on mechanistic tumor growth and drug-response models. Each of the 10,000 patients (four cancer types, 2,500 each) is simulated three times over — once per counterfactual chemotherapy schedule — plus given an oracle label for which arm would have worked best. No real patient data is used anywhere in the generation. Synthetic Patients 10,000 (breast, lung, colorectal, lymphoma — 2,500 each) Counterfactual Arms 3 per patient: MTD (q21d), dose-dense (q14d, equal total dose), metronomic (daily, 1/3 MTD intensity) + oracle optimal-arm label Regimens 30,000 Profile Rows 7,350,000 Data Schema Apache ... Full description with all tables is on the Wiki page of this project. Full dataset and purchase: https://sentineldata.com.ua/dataset/tumor-growth-chemotherapy-response-cohort-10k . Zenodo copy: https://doi.org/10.5281/zenodo.22538549