Normal and Atypical Mitosis Image Classifier using Efficient Vision Transformer
Xuan Qi, Dominic Labella, Thomas Sanford, Maxwell Lee · 2025
Read Normal and Atypical Mitosis Image Classifier using Efficient Vision Transformer by Xuan Qi, Dominic Labella, Thomas Sanford, Maxwell Lee (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-02589.txt.
Description
Xuan Qi and three co-authors wrote a 2025 paper on mitotic figures. These are images of dividing cells. The task is to sort them as normal or atypical. It forms part of the MIDOG 2025 challenge. Atypical division patterns can signal aggressive cancer behaviour.
Their classifier uses EfficientViT-L2. It mixes convolutional and transformer parts for speed and accuracy. Training data merged two public sets. The sets cover seven cancer types. Atypical cases form a small minority. To test generalisation, the team held out one cancer type at a time. They also ensembled five model folds. Stain deconvolution supplied augmentation. It separates dye colours in the images. The test model posted strong balanced accuracy and ranking scores. Performance held across cancer types.
Catalog record
| CREATOR | Xuan Qi, Dominic Labella, Thomas Sanford, Maxwell Lee |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 9 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-02589