Pathology image question answering, one of the earliest medical VQA sets.

PathVQA is a benchmark in healthcare AI. Pathology image question answering, one of the earliest medical VQA sets. Scores are reported as: Percent accuracy and open-answer scoring.

Audited by MedCheckHe et al., 2020

Key facts

What it measuresPathology image question answering, one of the earliest medical VQA sets.
Who should careResearchers citing the medical VQA literature.
Use it whenHistorical comparison in pathology VQA.
Do not use it forCurrent model selection. It is small and old.
Score scalePercent accuracy and open-answer scoring.
Grader methodunknown
First released2020
LanguagesEnglish
Use casesimaging
CitationHe et al., 2020
Statusactive

Paper

Frequently asked questions

What does PathVQA measure?

Pathology image question answering, one of the earliest medical VQA sets.

Who should use PathVQA?

Researchers citing the medical VQA literature. Historical comparison in pathology VQA.

What should PathVQA not be used for?

Current model selection. It is small and old.

How are PathVQA scores reported?

Percent accuracy and open-answer scoring. Scores from different benchmarks are not comparable to each other.

Is PathVQA independent?

No conflicts of interest have been verified for PathVQA in this registry as of 2026-08-05. Absence of a recorded conflict means none has been verified, not that none exists.

Compare with

Asclepius

Spectrum evaluation for multimodal medical models across many imaging modalities and difficulty levels.

CheXpert

A large chest radiograph dataset with uncertainty labels and expert comparison. Predates language models and i

EndoBench

Multimodal evaluation for endoscopy image and video analysis.

GMAI-MMBench

Large multimodal benchmark for general medical AI, spanning many imaging modalities, departments and task type

MMMU (Health and Medicine)

The health and medicine subset of a broad expert-level multimodal reasoning benchmark.

OmniMedVQA

Very large medical visual question answering set spanning many modalities and anatomical regions.

Last verified 2026-08-05. Governance and conflict data compiled by Healthcare Discovery from primary sources, each linked above. Seed inventory from Ma et al., Beyond the Leaderboard, ACL 2026. How we verify · All 63 entities