Enhancing Foreground Boundaries for Medical Image Segmentation

TL;DR

This paper introduces a boundary enhancement loss for medical image segmentation that improves boundary prediction accuracy without complex pre- or post-processing, validated by experimental results.

Contribution

The paper proposes a novel, lightweight boundary enhancement loss function to improve segmentation boundary accuracy in medical images.

Findings

Outperforms existing loss functions in boundary accuracy

Achieves comparable or better segmentation results

No additional pre- or post-processing required

Abstract

Object segmentation plays an important role in the modern medical image analysis, which benefits clinical study, disease diagnosis, and surgery planning. Given the various modalities of medical images, the automated or semi-automated segmentation approaches have been used to identify and parse organs, bones, tumors, and other regions-of-interest (ROI). However, these contemporary segmentation approaches tend to fail to predict the boundary areas of ROI, because of the fuzzy appearance contrast caused during the imaging procedure. To further improve the segmentation quality of boundary areas, we propose a boundary enhancement loss to enforce additional constraints on optimizing machine learning models. The proposed loss function is light-weighted and easy to implement without any pre- or post-processing. Our experimental results validate that our loss function are better than, or at least comparable to, other state-of-the-art loss functions in terms of segmentation accuracy.