TrialAlign

🤗 [Panacea Model] • 💻 [Github Repo] • 📃 [Panacea Paper]

TrialAlign datasets are the continual pretraining corpus for "Panacea: A foundation model for clinical trial search, summarization, design, and recruitment". TrialAlign contains 793,279 clinical trial design documents worldwide and 1,113,207 clinical study papers.

Load the dataset

pip install -U datasets fsspec

from datasets import load_dataset
ds = load_dataset('linjc16/ctgov', split='train')

Statistics of TrialAlign

Citation

If you find our work useful, please consider citing Panacea:

@article{lin2024panacea,
  title={Panacea: A foundation model for clinical trial search, summarization, design, and recruitment},
  author={Lin, Jiacheng and Xu, Hanwen and Wang, Zifeng and Wang, Sheng and Sun, Jimeng},
  journal={arXiv preprint arXiv:2407.11007},
  year={2024}
}