Skip to content

Dataset · anamnesis-data

Clinical Trials Registry Version History

anamnesis-data/clinical_trials_history

Full version history of 611k clinical trials in English in parquet format.

What's inside

A structured medical dataset by Anamnesis Data preserving all historical record versions from clinical trial registries, rather than just final statuses. The corpus covers 611,592 studies (4,485,603 versions) in English, organized into 79 tables spanning over 544 million rows. Data is partitioned into folders in parquet format.

What you can do with it

  • Training models to predict clinical trial completion timelines and success (time-series and tabular classification)
  • Analyzing protocol and date changes in pharmaceutical and biotech research
Tasks
Tabular classification, Tabular regression, Time series forecasting
Languages
English
Size
100M–1B rows
Downloads this month
59.3K
Likes
14
Access
on request on Hugging Face

How to load it

from datasets import load_dataset

ds = load_dataset("anamnesis-data/clinical_trials_history")