文献类型：期刊文献

英文题名：PDCA-Net: Parallel dual-channel attention network for polyp segmentation

作者：Chen, Gang[1];Zhang, Minmin[2];Zhu, Junmin[1];Meng, Yao[3]

第一作者：Chen, Gang

通信作者：Meng, Y[1]

机构：[1]Lanzhou Univ, Hosp & Clin Med Sch 2, Lanzhou 730030, Peoples R China;[2]Gansu Univ Chinese Med, Sch Clin Med, Lanzhou 730000, Peoples R China;[3]Gansu Privincial Hosp, Dept Geriatr, Lanzhou 730000, Peoples R China

第一机构：Lanzhou Univ, Hosp & Clin Med Sch 2, Lanzhou 730030, Peoples R China

通信机构：[1]corresponding author), Gansu Privincial Hosp, Dept Geriatr, Lanzhou 730000, Peoples R China.

年份：2025

卷号：101

外文期刊名：BIOMEDICAL SIGNAL PROCESSING AND CONTROL

收录：;EI(收录号：20244817442407);Scopus(收录号：2-s2.0-85210124005);WOS:【SCI-EXPANDED(收录号:WOS:001370300800001)】；

基金：This work was in part supported by the Natural science Foundation project of Gansu Provincial Science and Technology Department (No. 23JRRA1630, 23JRRA0959) , the Scientific research project of health industry of Gansu province (No. GSWSKY-2019-32) .

语种：英文

外文关键词：Polyp segmentation; Channel attention; Spatial attention; Adaptive associative mapping

摘要：Accurate segmentation of polyps in colonoscopy images is crucial for the diagnosis and cure of colorectal cancer. Although various deep learning methods have been proposed and have shown promising performance, accurately distinguishing between polyp and mucosal boundaries remains a challenge. In this work, we propose a Parallel Dual-Channel Attention Network (PDCA-Net) for polyp segmentation. This method utilizes the mapping transformations to adaptively encapsulate the global dependency from superpixel into pixels, enhancing the model's ability to localize foreground and background regions. Specifically, we first design a parallel spatial and channel attention fusion module to capture the global dependencies at the superpixel level from the spatial and channel dimensions. Furthermore, an adaptive associative mapping module is proposed to encapsulate the global dependencies of superpixels into each pixel through a coarse-to-fine learning strategy. Extensive experiments demonstrate that the proposed PDCA-Net effectively improves the segmentation performance and achieves new state-of-the-art results (i.e., 0.815, 0.936, 0.945, and 0.838 mDice, 0.744, 0.891, 0.900, and 0.765 mIoU on the ETIS, Kvasir-SEG, CVC-ClinicDB, and CVC-ColonDB). Our code is available at https://github.com/lzucg/PDCA-Net.