Octopath Databases

There is a newer version of the record available.

Published July 18, 2025 | Version v1
Dataset Open

OMG-Octo Atypical: A refinement of the original OMG-Octo database to incorporate atypical mitoses

Authors/Creators

  • 1. ROR icon University College London
  • 2. ROR icon University of Oxford
  • 3. ROR icon University Hospital of Zurich
  • 4. ROR icon University of Bern
  • 5. ROR icon Institute of Cancer Research

Description

In this study, we pre-trained our model on the MIDOG 2025 atypical mitoses database (AMi-Br: 832 atypical and 2,888 normal mitotic figures and MIDOG-Atypical (1748 atypical and 10191 normal mitoses), From this model, we inferred to find atypical mitoses within our OMG-Octo dataset and found 10678 potential candidates. Dedicated pathologists reviewed 3024 of them (1378 atypical, 394 apoptotic,  379 normal MF,  399 Noise and 462 Unsure). The results are uploaded here under a CC BY.0 NC ND 4.0 International License. 

Publications and upates will follow. 

Files

AtypicalMitoses.zip

Files (24.6 MB)

Name Size Download all
md5:d76628df50125af84ebf0df53e050e6a
24.6 MB Preview Download

Additional details

Funding

UK Research and Innovation
UKRI Future Leaders Fellowship MR/T040785/1
UK Research and Innovation
AI-based diagnosis for improving classification of bone and soft tissue tumours across the UK EP/Y020030/1

Software

Repository URL
https://github.com/SZY1234567/OMG-Net
Programming language
Python
Development Status
Active