HINT / TrialBench

Clinical trial outcome prediction benchmarks built on ClinicalTrials.gov (17-21k trials).

Composite

76.5

Experimental validation

Clinical

Stages

Phase IIPhase IIIClinical Development

Modalities

cross-modality

Task types

trial-outcome-prediction

Size

trials: 21,000
drugs: 12,000
diseases: 5,000

License

MIT

First release

2022

Last updated

2024-07

Official site

→ project page

Leaderboard

→ leaderboard

Dataset

→ dataset

Code / GitHub

→ repository

HuggingFace

→ HF

Paper

Hierarchical Interaction Network for Clinical Trial Outcome Prediction · Fu T, Huang K, Xiao C, Glass L, Sun J · 2022 · paper · doi:10.1016/j.patter.2022.100445 · 200 citations

Flags

none

Experts

Tianfan Fu, Marinka Zitnik

Groups

Fu Lab (RPI), Sun Lab (Georgia Tech), Zitnik Lab

Hosted by

TrialBench / HINT / TOP, Therapeutics Data Commons (TDC)

Related benchmarks

Trial Outcome Prediction (TOP), CT-Outcome (TrialBench v2)

Rubric (7-criterion)

rigor

coverage

maintenance

adoption

quality

accessibility

industry_relevance

Notes

Limited by ClinicalTrials.gov quality.

← Back to all benchmarks

Compare:

Open comparison →