CRIS: Collaborative Refinement Integrated with Segmentation for Polyp Segmentation

TL;DR

The paper presents two LaTeX class files, cas-sc.cls and cas-dc.cls, designed to streamline Elsevier journal article formatting within the publisher's updated workflow. It emphasizes flexible front matter handling, including long front matter with the longmktitle option, and supports complex annotations such as author marks, affiliations, corresponding author indicators, and footnotes. The work also documents theorem-style environments and extended list macros to enhance formatting, with practical examples and templates illustrating end-to-end usage. Overall, the contribution aims to simplify submission formatting and ensure compatibility with Elsevier’s systems across different column layouts.

Abstract

Accurate detection of colorectal cancer and early prevention heavily rely on precise polyp identification during gastrointestinal colonoscopy. Due to limited data, many current state-of-the-art deep learning methods for polyp segmentation often rely on post-processing of masks to reduce noise and enhance results. In this study, we propose an approach that integrates mask refinement and binary semantic segmentation, leveraging a novel collaborative training strategy that surpasses current widely-used refinement strategies. We demonstrate the superiority of our approach through comprehensive evaluation on established benchmark datasets and its successful application across various medical image segmentation architectures.