An Elastic Interaction-Based Loss Function for Medical Image Segmentation

TL;DR

This paper introduces an elastic interaction-based loss function for medical image segmentation that improves boundary accuracy and connectivity, especially for small structures like blood vessels, outperforming traditional pixel-wise losses.

Contribution

The paper proposes a novel elastic interaction-based loss function that enhances boundary connectivity and segmentation accuracy in medical images, addressing limitations of pixel-wise loss functions.

Findings

Significant improvement over pixel-wise losses on retinal vessel datasets

Enhanced boundary connectivity and segmentation precision

Effective for small and complex structures in biomedical images

Abstract

Deep learning techniques have shown their success in medical image segmentation since they are easy to manipulate and robust to various types of datasets. The commonly used loss functions in the deep segmentation task are pixel-wise loss functions. This results in a bottleneck for these models to achieve high precision for complicated structures in biomedical images. For example, the predicted small blood vessels in retinal images are often disconnected or even missed under the supervision of the pixel-wise losses. This paper addresses this problem by introducing a long-range elastic interaction-based training strategy. In this strategy, convolutional neural network (CNN) learns the target region under the guidance of the elastic interaction energy between the boundary of the predicted region and that of the actual object. Under the supervision of the proposed loss, the boundary of the predicted region is attracted strongly by the object boundary and tends to stay connected. Experimental results show that our method is able to achieve considerable improvements compared to commonly used pixel-wise loss functions (cross entropy and dice Loss) and other recent loss functions on three retinal vessel segmentation datasets, DRIVE, STARE and CHASEDB1.