Tubular Structure Segmentation via Multi-Scale Reverse Attention Sparse Convolution

Zeng, Xueqiang1,2; Guo, Yingwei2,3; Zaman, Asim2,4; Hassan, Haseeb2; Lu, Jiaxi1,2; Xu, Jiaxuan5; Yang, Huihui1,2; Miao, Xiaoqiang2,3; Cao, Anbo1,2; Yang, Yingjian2,3; Chen, Rongchang6,7,8; Kang, Yan1,2,3,9

2023-07-01

10.3390/diagnostics13132161

DIAGNOSTICS   影响因子和分区

2075-4418

Cerebrovascular and airway structures are tubular structures used for transporting blood and gases, respectively, providing essential support for the normal activities of the human body. Accurately segmenting these tubular structures is the basis of morphology research and pathological detection. Nevertheless, accurately segmenting these structures from images presents great challenges due to their complex morphological and topological characteristics. To address this challenge, this paper proposes a framework UARAI based on the U-Net multi-scale reverse attention network and sparse convolution network. The framework utilizes a multi-scale structure to effectively extract the global and deep detail features of vessels and airways. Further, it enhances the extraction ability of fine-edged features by a joint reverse attention module. In addition, the sparse convolution structure is introduced to improve the features' expression ability without increasing the model's complexity. Finally, the proposed training sample cropping strategy reduces the influence of block boundaries on the accuracy of tubular structure segmentation. The experimental findings demonstrate that the UARAI-based metrics, namely Dice and IoU, achieve impressive scores of 90.31% and 82.33% for cerebrovascular segmentation and 93.34% and 87.51% for airway segmentation, respectively. Compared to commonly employed segmentation techniques, the proposed method exhibits remarkable accuracy and robustness in delineating tubular structures such as cerebrovascular and airway structures. These results hold significant promise in facilitating medical image analysis and clinical diagnosis, offering invaluable support to healthcare professionals.

cerebrovascular airway tubular structures multi-scale reverse attention sparse convolution

SCI

英语

通讯

National Key Research and Development Program of China["2022YFF0710800","2022YFF0710802"] ; National Natural Science Foundation of China[62071311] ; special program for key fields of colleges and universities in Guangdong Province (biomedicine and health) of China[2021ZDZX2008] ; Stable Support Plan for Colleges and Universities in Shenzhen of China[SZWD2021010]

General & Internal Medicine

Medicine, General & Internal

WOS:001031003100001

MDPI

Web of Science

1.Shenzhen Univ, Sch Appl Technol, Shenzhen 518060, Peoples R China
2.Shenzhen Technol Univ, Coll Hlth Sci & Environm Engn, Shenzhen 518118, Peoples R China
3.Northeastern Univ, Coll Med & Biol Informat Engn, Shenyang 110169, Peoples R China
4.Shenzhen Univ, Med Sch, Sch Biomed Engn, Shenzhen 518060, Peoples R China
5.Guangzhou Med Univ, Affiliated Hosp 1, Natl Clin Res Ctr Resp Dis, State Key Lab Resp Dis,Natl Ctr Resp Med, Guangzhou 510120, Peoples R China
6.Shenzhen Peoples Hosp, Shenzhen Inst Resp Dis, Shenzhen 518001, Peoples R China
7.Jinan Univ, Clin Med Coll 2, Guangzhou 518001, Peoples R China
8.Southern Univ Sci & Technol, Affiliated Hosp 1, Shenzhen 518001, Peoples R China
9.Minist Educ, Engn Res Ctr Med Imaging & Intelligent Anal, Shenyang 110169, Peoples R China

Zeng, Xueqiang,Guo, Yingwei,Zaman, Asim,et al. Tubular Structure Segmentation via Multi-Scale Reverse Attention Sparse Convolution[J]. DIAGNOSTICS,2023,13(13).

Zeng, Xueqiang.,Guo, Yingwei.,Zaman, Asim.,Hassan, Haseeb.,Lu, Jiaxi.,...&Kang, Yan.(2023).Tubular Structure Segmentation via Multi-Scale Reverse Attention Sparse Convolution.DIAGNOSTICS,13(13).

Zeng, Xueqiang,et al."Tubular Structure Segmentation via Multi-Scale Reverse Attention Sparse Convolution".DIAGNOSTICS 13.13(2023).