Detecting Mitosis against Domain Shift using a Fused Detector and Deep Ensemble Classification Model for MIDOG Challenge
This work addresses domain shift in medical imaging for pathologists, but it is incremental as it builds on existing deep learning methods with specific enhancements.
The paper tackled the problem of mitotic figure detection in H&E stained images, which suffers from performance deterioration due to color variation, by proposing a two-stage framework combining a detector and deep ensemble classification with stain normalization and data augmentation, achieving an F1 score of 0.7550 on the MIDOG challenge testing set.
Mitotic figure count is an important marker of tumor proliferation and has been shown to be associated with patients' prognosis. Deep learning based mitotic figure detection methods have been utilized to automatically locate the cell in mitosis using hematoxylin \& eosin (H\&E) stained images. However, the model performance deteriorates due to the large variation of color tone and intensity in H\&E images. In this work, we proposed a two stage mitotic figure detection framework by fusing a detector and a deep ensemble classification model. To alleviate the impact of color variation in H\&E images, we utilize both stain normalization and data augmentation, aiding model to learn color irrelevant features. The proposed model obtains an F1 score of 0.7550 on the preliminary testing set released by the MIDOG challenge.