Learning of Inter-Label Geometric Relationships Using Self-Supervised Learning: Application To Gleason Grade Segmentation

TL;DR

This paper introduces a self-supervised learning method for prostate cancer histopathology image segmentation that learns inter-label geometric relationships, reducing the need for extensive manual annotations and improving robustness.

Contribution

It proposes a novel self-supervised approach using shape restoration and latent variable sampling to synthesize images and enhance segmentation with limited labeled data.

Findings

Outperforms existing image synthesis methods for segmentation.

Integrating geometry and diversity improves image quality.

Achieves comparable performance to fully supervised methods with limited labels.

Abstract

Segmentation of Prostate Cancer (PCa) tissues from Gleason graded histopathology images is vital for accurate diagnosis. Although deep learning (DL) based segmentation methods achieve state-of-the-art accuracy, they rely on large datasets with manual annotations. We propose a method to synthesize for PCa histopathology images by learning the geometrical relationship between different disease labels using self-supervised learning. We use a weakly supervised segmentation approach that uses Gleason score to segment the diseased regions and the resulting segmentation map is used to train a Shape Restoration Network (ShaRe-Net) to predict missing mask segments in a self-supervised manner. Using DenseUNet as the backbone generator architecture we incorporate latent variable sampling to inject diversity in the image generation process and thus improve robustness. Experiments on multiple histopathology datasets demonstrate the superiority of our method over competing image synthesis methods for segmentation tasks. Ablation studies show the benefits of integrating geometry and diversity in generating high-quality images, and our self-supervised approach with limited class-labeled data achieves similar performance as fully supervised learning.