# Pathology AI benchmarks (CAMELYON, PANDA, TCGA slide tasks)

Source: https://onco.cc/collections/pathology-benchmarks/  
OnCo record `pathology-benchmarks` (Collection). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Pathology AI benchmarks are the open challenge datasets on which every pathology model is scored: CAMELYON16 and 17 for lymph node metastasis detection, PANDA for prostate grading with 11,000 biopsies, and TCGA slide-level tasks used to compare foundation models. Licences are mostly CC BY-NC-SA or set per challenge.

## Summary

Pathology AI benchmarks are the exam papers on which every pathology model is graded. The open challenge datasets include CAMELYON16 and CAMELYON17 for lymph node metastasis detection, PANDA for prostate grading, and TCGA-derived slide-level tasks used to compare foundation models. PANDA, with 11,000 biopsies, remains the largest public prostate grading set, and leaderboards such as eva and HEST aggregate results across models. The datasets are maintained by Grand Challenge, Radboud UMC and partners, under CC BY-NC-SA or per-challenge terms. The collection is referenced by the AI in oncology roadmap, which traces the field from pattern readers through foundation models to agents in the workflow.

## Fields

- Kind: Collection
- Last checked: 2026-09-08
- Tags: data; benchmark
- URL: https://panda.grand-challenge.org
- Holds: Open challenge datasets for lymph node metastasis (CAMELYON16/17), prostate grading (PANDA), and TCGA-derived slide-level tasks used to compare foundation models.
- Licence: CC BY-NC-SA and per challenge

## Sources

- Pathology AI benchmarks (CAMELYON, PANDA, TCGA slide tasks): https://panda.grand-challenge.org

## Connected records

- roadmaps: [AI in oncology roadmap: pattern readers → foundation models → agents in the workflow](https://onco.cc/roadmaps/ai-oncology-roadmap/)

---
JSON: https://onco.cc/api/v1/entities/pathology-benchmarks.json