CaCL: Class-aware Codebook Learning for Weakly Supervised Segmentation on Diffuse Image Patterns

TL;DR

This paper introduces CaCL, a novel weakly supervised segmentation method tailored for diffuse biomedical image patterns, leveraging a class-aware codebook learning approach within a multi-task VQ-VAE framework, demonstrating superior results.

Contribution

The paper presents a new codebook learning approach for weakly supervised segmentation of diffuse patterns, optimized for biomedical images, and integrates it into a multi-task VQ-VAE framework.

Findings

Achieved superior segmentation performance over baseline algorithms.

Effectively segments diffuse patterns like stains and fluorescence.

Demonstrates applicability to histological images of human duodenum.

Abstract

Weakly supervised learning has been rapidly advanced in biomedical image analysis to achieve pixel-wise labels (segmentation) from image-wise annotations (classification), as biomedical images naturally contain image-wise labels in many scenarios. The current weakly supervised learning algorithms from the computer vision community are largely designed for focal objects (e.g., dogs and cats). However, such algorithms are not optimized for diffuse patterns in biomedical imaging (e.g., stains and fluorescence in microscopy imaging). In this paper, we propose a novel class-aware codebook learning (CaCL) algorithm to perform weakly supervised learning for diffuse image patterns. Specifically, the CaCL algorithm is deployed to segment protein expressed brush border regions from histological images of human duodenum. Our contribution is three-fold: (1) we approach the weakly supervised segmentation from a novel codebook learning perspective; (2) the CaCL algorithm segments diffuse image patterns rather than focal objects; and (3) the proposed algorithm is implemented in a multi-task framework based on Vector Quantised-Variational AutoEncoder (VQ-VAE) via joint image reconstruction, classification, feature embedding, and segmentation. The experimental results show that our method achieved superior performance compared with baseline weakly supervised algorithms. The code is available at https://github.com/ddrrnn123/CaCL.