KarelDO/BioDEX

BioDEX: Large-Scale Biomedical Adverse Drug Event Extraction for Real-World Pharmacovigilance.

36
/ 100
Emerging

Pairs 256k+ expert-annotated drug safety reports with PubMed articles (19k full-text, 65k abstracts) to enable structured adverse event extraction. Built on encoder-decoder architectures (FLAN-T5) fine-tuned for report generation from clinical text, with modular Python objects for seamless data manipulation and published benchmark datasets (BioDEX-ICSR, BioDEX-Reactions) on Hugging Face. Supports both full ICSR report extraction and reaction-only classification tasks via in-context learning approaches.

No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

60

Forks

7

Language

Jupyter Notebook

License

MIT

Last pushed

Jan 29, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/KarelDO/BioDEX"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.