计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (6A): 220400273-6.doi: 10.11896/jsjkx.220400273

• 图像处理&多媒体技术 • 上一篇    下一篇

基于SegFormer的超声影像图像分割

杨靖怡1, 李芳1,2, 康晓东1, 王笑天1, 刘汉卿1, 韩俊玲1   

  1. 1 天津医科大学医学影像学院 天津 300202;
    2 北京市化工职业病防治院 北京 100093
  • 出版日期:2023-06-10 发布日期:2023-06-12
  • 通讯作者: 康晓东(kxd2004@126.com)
  • 作者简介:(yangjy9912@163.com)
  • 基金资助:
    京津冀协同创新项目(17YEXTZC00020)

Ultrasonic Image Segmentation Based on SegFormer

YANG Jingyi1, LI Fang1,2, KANG Xiaodong1, WANG Xiaotian1, LIU Hanqing1, HAN Junling1   

  1. 1 School of Medical Image,Tianjin Medical University,Tianjin 300202,China;
    2 Beijing Chemical Occupational Disease Control Hospital,Beijing 100093,China
  • Online:2023-06-10 Published:2023-06-12
  • About author:YANG Jingyi,born in 1999,undergra-duates.Her main research interests include medical imaging diagnosis and so on. KANG Xiaodong,born in 1964,Ph.D,professor.His main research interests include medical image processing and medical information system integration.
  • Supported by:
    Beijing-Tianjin-Hebei Collaborative Innovation Project(17YEXTZC00020).

PDF (PC)

337

摘要： 超声影像分割既是医学影像图像处理的重要环节,也是临床诊断的常用技术手段。文中提出将SegFormer网络模型用于实现医学超声影像图像的精准分割。一方面,将超声标签图转化为单通道形式,并对其进行二值化处理,以完成对数据集图像的预处理;另一方面,采用迁移学习的方式载入预训练模型,用于微调已经训练好的模型参数,并选用带有动量的随机梯度下降优化器来加速收敛速度及减小震荡。与FCN,UNet和DeepLabV3的对比实验结果表明,该模型在乳腺结节超声影像数据集上的各项评估指标均为最优,mIoU,Acc,DSC和Kappa分别为81.32%,96.22%,88.91%和77.85%。实验结果还表明,该模型在不同超声影像数据集中表现出了良好的鲁棒性。

关键词: SegFormer, 图像分割, 超声影像, Transformer

Abstract: Ultrasonic image segmentation is not only an important part of medical image processing,but also a common technical means of clinical diagnosis.In this paper,the SegFormer network model is proposed to realize the accurate segmentation of medical ultrasound images.On the one hand,the ultrasonic label image is transformed into a single channel and processed by binarization to complete the preprocessing of the data set image;on the other hand,the pre-training model is loaded into the pre-training model to fine-tune the trained model parameters,and a random gradient descent optimizer with momentum is selected to accelerate the convergence speed and reduce the oscillation.Experimental results show that,compared with FCN,UNet and DeepLabV3,all the evaluation indexes of the proposed model are the best in the breast nodule ultrasound image data set,and the evaluation indexes of mIoU,Acc,DSC and Kappa is 81.32%,96.22%,88.91% and 77.85% respectively.The experimental results also show that the model is robust in different ultrasonic image data sets.

Key words: SegFormer, Image segmentation, Ultrasonic image, Transformer

中图分类号: 

  • TP391
[1]ZHOU T,DONG Y L,HUO B Q,et al.U-Net and its applications in medical image segmentation:a review[J].Journal of Image and Graphics,2021,26(9):2058-2077.
[2]KANG X D.Medical Image Processing[M].People’s Medical Publishing House,2009:200.
[3]CHEN C,QI F.Review on Development of Convolutional Neural Network and Its Application in Computer Vision[J].Computer Science,2019,46(3):63-73.
[4]GIRSHICK R,DONAHUE J,DARRELL T,et al.Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2014:580-587.
[5]LONG J,SHELHAMER E,DARRELL T.Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2015:3431-3440.
[6]YAP M H,GOYAL M,OSMAN F M,et al.Breast ultrasound lesions recognition:end-to-end deep learning approaches[J].Journal of Medical Imaging,2018,6(1):011007.
[7]RONNEBERGER O,FISCHER P,BROX T.U-net:Convolu-tional networks for biomedical image segmentation[C]//Proceedings of the 2015 International Conference on Medical Image Computing and Computer-assisted Intervention.Springer,2015:234-241.
[8]ZHUANG Z,LI N,JOSEPH RAJ A N,et al.An RDAU-NET model for lesion segmentation in breast ultrasound images[J].PloS One,2019,14(8):e0221535.
[9]IBTEHAZ N,RAHMAN M S J N N.MultiResUNet:Rethin-king the U-Net architecture for multimodal biomedical image segmentation[J].Neural Networks,2020,121:74-87.
[10]HE K,GKIOXARI G,DOLLÁR P,et al.Mask r-cnn[C]//Proceedings of the IEEE International Conference on Computer Vision.2017:2961-2969.
[11]WANG B,LI M X,LIU X.Ultrasound Image SegmentationMethod of Thyroid Nodules Based on the Improved U-Net Network[J].Journal of Electronics & Information Technology,2022,44(2):514-522.
[12]CARNEIRO G,NASCIMENTO J,FREITAS A.Robust leftventricle segmentation from ultrasound data using deep neural networks and efficient search methods[C]//Proceedings of the 2010 IEEE International Symposium on Biomedical Imaging.From Nano to Macro:IEEE,2010:1085-1088.
[13]ANAS E M A,NOURANIAN S,MAHDAVI S S,et al.Clinical target-volume delineation in prostate brachytherapy using residual neural networks[C]//Proceedings of the 2017 International Conference on Medical Image Computing and Computer-Assisted Intervention.Springer,2017:365-373.
[14]MILLETARI F,AHMADI S A,KROLL C,et al.Hough-CNN:deep learning for segmentation of deep brain regions in MRI and ultrasound[J].Computer Vision and Image Understanding,2017,164:92-102.
[15]CHEN L C,PAPANDREOU G,SCHROFF F,et al.Rethinking atrous convolution for semantic image segmentation[J].arXiv:1706.05587,2017.
[16]DOSOVITSKIY A,BEYER L,KOLESNIKOV A,et al.An image is worth 16x16 words:Transformers for image recognition at scale[J].arXiv:2010.11929,2020.
[17]CARION N,MASSA F,SYNNAEVE G,et al.End-to-end object detection with transformers[C]//Proceedings of the 2020 European Conference on Computer Vision.Springer,2020:213-229.
[18]WANG W,XIE E,LI X,et al.Pyramid vision transformer:Aversatile backbone for dense prediction without convolutions[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:568-578.
[19]TAY Y,DEHGHANI M,BAHRI D,et al.Efficient transfor-mers:A survey[J].arXiv:2009.06732,2020.
[20]XIE E,WANG W,YU Z,et al.SegFormer:Simple and efficient design for semantic segmentation with transformers[J].Advances in Neural Information Processing Systems,2021,34.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发