计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 231000003-6.doi: 10.11896/jsjkx.231000003
王利彬, 王树梅
WANG Libin, WANG Shumei
摘要： 为了缩小编码器- 解码器结构存在的语义差距,提出了一种基于注意力机制的医学图像分割算法。首先,使用CBAM注意力模块,通过注意力机制模块增强模型进行医学图像的特征提取;其次,将CBAM模块输出的特征图作为文中所提出的特征细化模块的输入,用于恢复由于下采样所丢失的血管细节信息;最后,使用一种尺度注意力模块,将不同尺度的特征图所具有的特征结合起来形成最终的预测。通过与当下流行的眼底血管分割算法进行对比,所提算法在DRIVE数据集上的mIoU最高提升了2～3个百分点,最接近的也提升了0.4个百分点,证明了所提模型能够有效提升分割精度,对于恢复细微血管像素有着较好的效果。
中图分类号:
| [1]FRAZ M M,REMAGNINO P,HOPPE A,et al.Blood vesselsegmentation methodologies in retinal images-a survey[J].Computer Methods and Programs in Biomedicine,2012,108(1):407-433. [2]CELEBI M E,QUAN W,HWANG S,et al.Lesion Border Detection in Dermoscopy Images Using Ensembles of Thresholding Methods[J].Skin Research and Technology,2013,19(1):252-258. [3]HEMANN T,MEINZER H P.Statistical shape models for 3D medical image segmentation:A review[J].Medical Image Analysis,2009,13(4):543-563. [4]OLIVA D,CUEVAS E,PAJARES G,et al.Multilevel Thresholding Segmentation Based on Harmony Search Optimization[J].Journal of Applied Mathematics,2013,2013:1-12. [5]ABED S,ALI M H,AI-SHAYEJI M.An adaptive edge detec-tion operator for noisy images based on a total variation approach restoration[J].Computer Systems Science and Engineering,2017,32(1):21-33. [6] RICCI E,PERFETTI R.Retinal blood vessel segmentationusing line operators and support vector classification[J].IEEE Transactions on Medical Imaging,2007,26(10):1357-1365. [7]KAUR N,SHARMA M.Brain tumor detection using self-adaptive K-means clustering[C]//Proceedings of the International Conference on Energy,Communication,Data Analytics and Soft Computing.Los Alamitos:IEEE Computer Society Press,2017:1861-1865. [8] RONNEBERGER O,FISCHER P,BROX T.U-Net:convolu-tional networks for biomedical image segmentation[C]//Proceedings of International Conference on Medical Image Computing and Computer-Assisted Intervention.Heidelberg:Springer,2015:234-241. [9]OKTAY O,SCHLEMPER J,FOLGOC L L,et al.AttentionU-Net:learning where to look for the pancreas[OL].[2021-06-21].https://arxiv.org/abs/1804.03999. [10] GU R,WANG G,SONG T,et al.CA-Net:ComprehensiveAttention Convolutional Neural Networks for Explainable Medical Image Segmentation[J].IEEE Transactions on Medical Imaging,2020,40(2):699-711. [11] CHEN D Q,ZHANG F,HAO P Y,et al.2D Medical Image Segmentation Combining Multi-Scale Channel Attention and Boundary Enhancement[J].Journal of Computer-Aided Design & Computer Graphics,2022,34(11):1742-1752. [12] HOU X D,LI Z Y,NIU J Y,et al.Retinal Vessel Segmentation Based on Attention Mechanism and Multi-Path U-Net[J]. Journal of Computer-Aided Design & Computer Graphics,2023,35(1):55-65. [13]WOO S,PARK J,LEE J Y,et al.Cbam:Convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision(ECCV).2018:3-19. [14] HU J,SHEN L,SUN G.Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:7132-7141. [15] STAAL J,ABRAMOFF M D,NIEMEIJER M,et al.Ridge-based vessel segmentation in color images of the retina[J].IEEE Transactions on Medical Imaging,2004,23(4):501-509. [16]ZHOU Z,SIDDIQUEE M M R,TAJBAKHSH N,et al.UNet++:Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation[J].IEEE Transactions on Medical Imaging,2020,39(6):1856-1867. [17] ZHANG Z,FU H,DAI H,et al.Et-net:A generic edge-atten-tion guidance network for medical image segmentation[C]//Medical Image Computing and Computer Assisted Intervention-MICCAI 2019:22nd International Conference,Shenzhen,China,October 13-17,2019,Proceedings,Part I 22.Springer International Publishing,2019:442-450. [18] GUO C,SZEMENYEI M,YI Y,et al.Sa-unet:Spatial attentionu-net for retinal vessel segmentation[C]//2020 25th International Conference on Pattern Recognition(ICPR).IEEE,2021:1236-1242. [19] JHA D,SMEDSRUD P H,RIEGLER M A,et al.Resunet++:An advanced architecture for medical image segmentation[C]//2019 IEEE International Symposium on Multimedia(ISM).IEEE,2019:225-2255. [20] LIU W,YANG H,TIAN T,et al.Full-resolution network and dual-threshold iteration for retinal vessel and coronary angiograph segmentation[J].IEEE Journal of Biomedical and Health Informatics,2022,26(9):4623-4634. [21] ALOM M Z,HASAN M,YAKOPCIC C,et al.Recurrent residual convolutional neural network based on u-net(r2u-net)for medical image segmentation[J].arXiv:1802.06955,2018. |
|
||
首页