Stain-Robust Mitotic Figure Detection for the Mitosis Domain Generalization Challenge

TL;DR

This paper presents a hybrid deep learning approach for robust mitotic figure detection across different scanners, demonstrating high generalization performance in the MIDOG challenge.

Contribution

The study introduces a stain normalization and hybrid detection model that improves robustness and generalization in mitotic figure detection across unseen scanner data.

Findings

F1-score of 0.786 on training data

F1-score of 0.765 on preliminary test set

Model generalizes well to unseen scanners

Abstract

The detection of mitotic figures from different scanners/sites remains an important topic of research, owing to its potential in assisting clinicians with tumour grading. The MItosis DOmain Generalization (MIDOG) challenge aims to test the robustness of detection models on unseen data from multiple scanners for this task. We present a short summary of the approach employed by the TIA Centre team to address this challenge. Our approach is based on a hybrid detection model, where mitotic candidates are segmented on stain normalised images, before being refined by a deep learning classifier. Cross-validation on the training images achieved the F1-score of 0.786 and 0.765 on the preliminary test set, demonstrating the generalizability of our model to unseen data from new scanners.