Detecting Mitoses with a Convolutional Neural Network for MIDOG 2022 Challenge
This work addresses the problem of automating mitosis detection for pathologists, but it is incremental as it builds on existing CNN and class activation map techniques.
The paper tackled mitosis detection in pathology images using a single CNN with class activation maps for localization, achieving an F1 score of 0.7323 in preliminary tests and 0.6847 in final tests for the MIDOG 2022 challenge.
This work presents a mitosis detection method with only one vanilla Convolutional Neural Network (CNN). Our method consists of two steps: given an image, we first apply a CNN using a sliding window technique to extract patches that have mitoses; we then calculate each extracted patch's class activation map to obtain the mitosis's precise location. To increase the model performance on high-domain-variance pathology images, we train the CNN with a data augmentation pipeline, a noise-tolerant loss that copes with unlabeled images, and a multi-rounded active learning strategy. In the MIDOG 2022 challenge, our approach, with an EfficientNet-b3 CNN model, achieved an overall F1 score of 0.7323 in the preliminary test phase, and 0.6847 in the final test phase (task 1). Our approach sheds light on the broader applicability of class activation maps for object detections in pathology images.