MIDOG Challenge
Mitosis detection benchmarked across scanners, tumours and species.
Mitotic count drives grading in several tumours, and it is one of the least reproducible things pathologists do — which is exactly why this benchmark measures generalisation rather than headline accuracy.
- Cost
- Free
- Hardware
- Needs a consumer GPU
- Offline
- Works without a connection
- Scanner
- Not required
- Your data
- Stays on your machine
- Licence
- unknown
- Maturity
- production
- From
- DeepMicroscopy
What it is
Challenge series built specifically around domain shift: models must count mitoses on scanners, tumour types and species they were not trained on. Successive editions have widened the domain gap deliberately.
Licence notes
Check the repository before any reuse.
Activity: 14 stars · last commit 2022-06-19 — refreshed nightly
Related
Catalogued 2026-08-02, last checked 2026-08-02. View the source record · Report an error