Alzheimer's Counterfactual Twin Cohort
FREE SAMPLE INCLUDED. This project holds the full technical documentation and a free evaluation sample (500 people, all 27 scenarios, all tables, hidden truth included). It is not the full dataset. Full dataset: https://sentineldata.com.ua/dataset/alzheimers-counterfactual-twin-cohort Zenodo record (same package): https://doi.org/10.5281/zenodo.24227228 About the dataset: ten thousand synthetic people aged 50 to 75 are followed for up to 25 years through Alzheimer's disease, and each person is lived 27 times: once untreated and 26 times under different medical strategies, with the same body and the same random numbers. The untreated course of every person is known exactly, so the effect of any strategy on that person is a subtraction of two of their own trajectories. Next to the noisy clinical record (MMSE, CDR-SB, plasma p-tau217, amyloid PET) the data carry the hidden truth behind it: true amyloid, tau, inflammation, clearance and damage every quarter. The full description, with the disease model, validation, benchmarks and limitations, is on the project wiki. Sample license: CC BY-NC 4.0. All data are synthetic and describe no real person. Version 1.0.
Authors
- Chekerskyi Maksym (ORCID: https://orcid.org/0009-0004-9749-4867)
Publication Details
- Journal
- Open Science Framework
- Published
- 2026-11-14
- DOI
- https://doi.org/10.17605/osf.io/epk8q
- Primary Topic
- Alzheimer's disease research and treatments
- Type
- article
- Field-Weighted Citation Impact
- 0.00