No Masks Needed: Explainable AI for Deriving Segmentation from Classification

TL;DR

This paper presents a novel method that fine-tunes pre-trained models with explainable AI to improve medical image segmentation, achieving better results on key datasets compared to traditional approaches.

Contribution

It introduces a fine-tuning approach combined with explainable AI for medical image segmentation, addressing limitations of existing unsupervised methods in medical domains.

Findings

Improved segmentation accuracy on CBIS-DDSM, NuInsSeg, and Kvasir-SEG datasets.

Enhanced explainability of segmentation results.

Outperforms traditional methods in medical imaging benchmarks.

Abstract

Medical image segmentation is vital for modern healthcare and is a key element of computer-aided diagnosis. While recent advancements in computer vision have explored unsupervised segmentation using pre-trained models, these methods have not been translated well to the medical imaging domain. In this work, we introduce a novel approach that fine-tunes pre-trained models specifically for medical images, achieving accurate segmentation with extensive processing. Our method integrates Explainable AI to generate relevance scores, enhancing the segmentation process. Unlike traditional methods that excel in standard benchmarks but falter in medical applications, our approach achieves improved results on datasets like CBIS-DDSM, NuInsSeg and Kvasir-SEG.