GMSRF-Net: An improved generalizability with global multi-scale residual fusion network for polyp segmentation

TL;DR

This paper introduces GMSRF-Net, a novel polyp segmentation network that enhances generalizability across datasets by using multi-scale residual fusion, cross-scale attention, and feature selection modules, outperforming existing methods.

Contribution

The paper proposes GMSRF-Net, a new architecture with multi-scale fusion, attention, and feature selection modules to improve generalizability in polyp segmentation across diverse datasets.

Findings

GMSRF-Net outperforms state-of-the-art by 8.34% and 10.31% in dice coefficient on unseen datasets.

The proposed modules improve model robustness across different imaging protocols.

Experiments validate the effectiveness of multi-scale fusion and attention mechanisms.

Abstract

Colonoscopy is a gold standard procedure but is highly operator-dependent. Efforts have been made to automate the detection and segmentation of polyps, a precancerous precursor, to effectively minimize missed rate. Widely used computer-aided polyp segmentation systems actuated by encoder-decoder have achieved high performance in terms of accuracy. However, polyp segmentation datasets collected from varied centers can follow different imaging protocols leading to difference in data distribution. As a result, most methods suffer from performance drop and require re-training for each specific dataset. We address this generalizability issue by proposing a global multi-scale residual fusion network (GMSRF-Net). Our proposed network maintains high-resolution representations while performing multi-scale fusion operations for all resolution scales. To further leverage scale information, we design cross multi-scale attention (CMSA) and multi-scale feature selection (MSFS) modules within the GMSRF-Net. The repeated fusion operations gated by CMSA and MSFS demonstrate improved generalizability of the network. Experiments conducted on two different polyp segmentation datasets show that our proposed GMSRF-Net outperforms the previous top-performing state-of-the-art method by 8.34% and 10.31% on unseen CVC-ClinicDB and unseen Kvasir-SEG, in terms of dice coefficient.