USegMix: Unsupervised Segment Mix for Efficient Data Augmentation in Pathology Images

TL;DR

USegMix is an unsupervised data augmentation technique for pathology images that creates realistic synthetic images by segment replacement and blending, improving cancer classification accuracy.

Contribution

It introduces a novel two-phase unsupervised method combining superpixels, SAM, and diffusion models for realistic pathology image augmentation.

Findings

Enhanced cancer classification performance.

Effective generation of diverse, realistic pathology images.

Applicable to colorectal and prostate cancer datasets.

Abstract

In computational pathology, researchers often face challenges due to the scarcity of labeled pathology datasets. Data augmentation emerges as a crucial technique to mitigate this limitation. In this study, we introduce an efficient data augmentation method for pathology images, called USegMix. Given a set of pathology images, the proposed method generates a new, synthetic image in two phases. In the first phase, USegMix constructs a pool of tissue segments in an automated and unsupervised manner using superpixels and the Segment Anything Model (SAM). In the second phase, USegMix selects a candidate segment in a target image, replaces it with a similar segment from the segment pool, and blends them by using a pre-trained diffusion model. In this way, USegMix can generate diverse and realistic pathology images. We rigorously evaluate the effectiveness of USegMix on two pathology image datasets of colorectal and prostate cancers. The results demonstrate improvements in cancer classification performance, underscoring the substantial potential of USegMix for pathology image analysis.