An Edge-Enhanced Network for Polyp Segmentation

被引:3
|
作者
Tong, Yao [1 ,2 ]
Chen, Ziqi [3 ]
Zhou, Zuojian [1 ,2 ]
Hu, Yun [1 ,2 ]
Li, Xin [4 ]
Qiao, Xuebin [2 ,5 ]
机构
[1] Nanjing Univ Chinese Med, Sch Artificial Intelligence & Informat Technol, Nanjing 210023, Peoples R China
[2] Nanjing Univ Chinese Med, Jiangsu Prov Engn Res Ctr TCM Intelligence Hlth Se, Nanjing 210023, Peoples R China
[3] Tsinghua Univ, Vanke Sch Publ Hlth, Beijing 100084, Peoples R China
[4] Hohai Univ, Coll Comp Sci & Software Engn, Nanjing 211100, Peoples R China
[5] Nanjing Univ Chinese Med, Sch Elderly Care Serv & Management, Nanjing 210023, Peoples R China
来源
BIOENGINEERING-BASEL | 2024年 / 11卷 / 10期
基金
中国国家自然科学基金;
关键词
polyp segmentation; convolutional neural network; edge enhancement; attention mechanism;
D O I
10.3390/bioengineering11100959
中图分类号
Q81 [生物工程学(生物技术)]; Q93 [微生物学];
学科分类号
071005 ; 0836 ; 090102 ; 100705 ;
摘要
Colorectal cancer remains a leading cause of cancer-related deaths worldwide, with early detection and removal of polyps being critical in preventing disease progression. Automated polyp segmentation, particularly in colonoscopy images, is a challenging task due to the variability in polyp appearance and the low contrast between polyps and surrounding tissues. In this work, we propose an edge-enhanced network (EENet) designed to address these challenges by integrating two novel modules: the covariance edge-enhanced attention (CEEA) and cross-scale edge enhancement (CSEE) modules. The CEEA module leverages covariance-based attention to enhance boundary detection, while the CSEE module bridges multi-scale features to preserve fine-grained edge details. To further improve the accuracy of polyp segmentation, we introduce a hybrid loss function that combines cross-entropy loss with edge-aware loss. Extensive experiments show that the EENet achieves a Dice score of 0.9208 and an IoU of 0.8664 on the Kvasir-SEG dataset, surpassing state-of-the-art models such as Polyp-PVT and PraNet. Furthermore, it records a Dice score of 0.9316 and an IoU of 0.8817 on the CVC-ClinicDB dataset, demonstrating its strong potential for clinical application in polyp segmentation. Ablation studies further validate the contribution of the CEEA and CSEE modules.
引用
收藏
页数:19
相关论文
empty
未找到相关数据