SUPRA: Superpixel Guided Loss for Improved Multi-modal Segmentation in Endoscopy

TL;DR

This paper introduces SUPRA, a superpixel-guided loss function that enhances the generalization of deep learning models for multi-modal endoscopic image segmentation, effectively addressing domain shift issues in clinical settings.

Contribution

The paper proposes SLICLoss and SUPRA, a novel superpixel-based augmentation method, to improve multi-modal segmentation performance under domain shift in endoscopy.

Findings

SLICLoss combined with BCE improves model generalization.

SUPRA yields nearly 20% performance gain on target domain.

Method validated on EndoUDA dataset with multiple modalities.

Abstract

Domain shift is a well-known problem in the medical imaging community. In particular, for endoscopic image analysis where the data can have different modalities the performance of deep learning (DL) methods gets adversely affected. In other words, methods developed on one modality cannot be used for a different modality. However, in real clinical settings, endoscopists switch between modalities for better mucosal visualisation. In this paper, we explore the domain generalisation technique to enable DL methods to be used in such scenarios. To this extend, we propose to use super pixels generated with Simple Linear Iterative Clustering (SLIC) which we refer to as "SUPRA" for SUPeRpixel Augmented method. SUPRA first generates a preliminary segmentation mask making use of our new loss "SLICLoss" that encourages both an accurate and color-consistent segmentation. We demonstrate that SLICLoss when combined with Binary Cross Entropy loss (BCE) can improve the model's generalisability with data that presents significant domain shift. We validate this novel compound loss on a vanilla U-Net using the EndoUDA dataset, which contains images for Barret's Esophagus and polyps from two modalities. We show that our method yields an improvement of nearly 20% in the target domain set compared to the baseline.