Enhancing Medical Image Segmentation: Optimizing Cross-Entropy Weights and Post-Processing with Autoencoders

TL;DR

This paper introduces a deep learning method that improves medical image segmentation accuracy by optimizing loss function weights and utilizing autoencoders for post-processing, specifically applied to dermatomyositis images.

Contribution

It presents a novel approach that enhances segmentation performance and benchmarks it across multiple challenging medical imaging tasks, focusing on autoimmune disease analysis.

Findings

Outperforms state-of-the-art by over 12% on dermatomyositis dataset

Optimizing loss weights significantly improves segmentation accuracy

Autoencoder-based post-processing enhances delineation of medical images

Abstract

The task of medical image segmentation presents unique challenges, necessitating both localized and holistic semantic understanding to accurately delineate areas of interest, such as critical tissues or aberrant features. This complexity is heightened in medical image segmentation due to the high degree of inter-class similarities, intra-class variations, and possible image obfuscation. The segmentation task further diversifies when considering the study of histopathology slides for autoimmune diseases like dermatomyositis. The analysis of cell inflammation and interaction in these cases has been less studied due to constraints in data acquisition pipelines. Despite the progressive strides in medical science, we lack a comprehensive collection of autoimmune diseases. As autoimmune diseases globally escalate in prevalence and exhibit associations with COVID-19, their study becomes increasingly essential. While there is existing research that integrates artificial intelligence in the analysis of various autoimmune diseases, the exploration of dermatomyositis remains relatively underrepresented. In this paper, we present a deep-learning approach tailored for Medical image segmentation. Our proposed method outperforms the current state-of-the-art techniques by an average of 12.26% for U-Net and 12.04% for U-Net++ across the ResNet family of encoders on the dermatomyositis dataset. Furthermore, we probe the importance of optimizing loss function weights and benchmark our methodology on three challenging medical image segmentation tasks