Improved HER2 Tumor Segmentation with Subtype Balancing using Deep Generative Networks

TL;DR

This paper introduces a novel approach using deep generative networks, especially diffusion models, to create synthetic, subtype-conditioned images that improve HER2 tumor segmentation accuracy and reduce subtype recall variance.

Contribution

It demonstrates the effectiveness of diffusion models for generating realistic, subtype-conditioned histopathology images to enhance tumor segmentation performance.

Findings

Tumor Dice score increased from 0.833 to 0.854.

Variance between HER2 subtype recalls was nearly halved.

Diffusion models effectively inpaint HER2 tumor areas with modified subtypes.

Abstract

Tumor segmentation in histopathology images is often complicated by its composition of different histological subtypes and class imbalance. Oversampling subtypes with low prevalence features is not a satisfactory solution since it eventually leads to overfitting. We propose to create synthetic images with semantically-conditioned deep generative networks and to combine subtype-balanced synthetic images with the original dataset to achieve better segmentation performance. We show the suitability of Generative Adversarial Networks (GANs) and especially diffusion models to create realistic images based on subtype-conditioning for the use case of HER2-stained histopathology. Additionally, we show the capability of diffusion models to conditionally inpaint HER2 tumor areas with modified subtypes. Combining the original dataset with the same amount of diffusion-generated images increased the tumor Dice score from 0.833 to 0.854 and almost halved the variance between the HER2 subtype recalls. These results create the basis for more reliable automatic HER2 analysis with lower performance variance between individual HER2 subtypes.