Detecting Mitoses with a Convolutional Neural Network for MIDOG 2022 Challenge

TL;DR

This paper introduces a mitosis detection method using a single CNN with a sliding window and class activation maps, enhanced by data augmentation, noise-tolerant loss, and active learning, achieving competitive results in MIDOG 2022.

Contribution

The novel approach combines a simple CNN architecture with advanced training strategies for improved pathology image analysis.

Findings

Achieved F1 score of 0.7323 in preliminary test

Achieved F1 score of 0.6847 in final test

Demonstrated effectiveness of class activation maps for object detection

Abstract

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.