计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 230400038-6.doi: 10.11896/jsjkx.230400038

• 图像处理&多媒体技术 • 上一篇    下一篇

面向3D肝脏CT图像分割的改进vnet模型

杨舒琪1, 韩俊玲1,2, 康晓东1, 杨靖怡1, 郭洪洋1, 李博3   

  1. 1 天津医科大学医学影像学院 天津 300202
    2 天津医科大学一中心临床学院 天津 300190
    3 天津医科大学三中心临床学院 天津 300170
  • 发布日期:2024-06-06
  • 通讯作者: 康晓东(yqishu@163.com)
  • 作者简介:(yqishu@163.com)
  • 基金资助:
    京津冀协同创新项目(17YEXTZC00020)

Improved vnet Model for 3D Liver CT Image Segmentation

YANG Shuqi1, HAN Junling1,2, KANG Xiaodong1, YANG Jingyi1, GUO Hongyang1, LI Bo3   

  1. 1 School of Medical Image,Tianjin Medical University,Tianjin 300202,China
    2 The First Central Clinical College of Tianjin Medical University,Tianjin 300190,China
    3 The Third Central Clinical College of Tianjin Medical University,Tianjin 300170,China
  • Published:2024-06-06
  • About author:YANG Shuqi,born in 2001.Her main research interests include medical image processing and so on.
    KANG Xiaodong,born in 1964,Ph.D,professor,is a member of CCF(No.11186S).His main research interests include medical image processing and medical information system integration.
  • Supported by:
    Beijing-Tianjin-Hebei Collaborative Innovation Project(17YEXTZC00020).

PDF (PC)

341

摘要： 分割3D医学影像是放疗计划的重要步骤。临床上,计算机断层扫描被广泛应用于肝脏及肝肿瘤的3D 医学影像图像分割。由于肝脏复杂的边缘结构及纹理特征,肝脏分割仍是一项具有挑战性的工作。针对这一问题,提出了一种面向3D肝脏CT图像精准分割的改进vnet模型。首先,将肝脏CT图像进行HU值截断和重采样,以完成三维数据集的预处理;同时,将vnet解码器和编码器中的卷积核替换为SG模块,即逐通道卷积和逐点卷积的组合,以减小网络模型的参数量。与vnet模型进行对比实验,结果表明该模型方法在肝脏分割数据集上的评估结果总体优越,Dice系数为94.93%,比vnet模型提高了3.49%,大大减少了模型的参数量;同时该方法在MSD脾脏分割数据集和新冠肺炎数据集上也表现出良好的鲁棒性并取得了优越的分割结果。

关键词: 图像分割, 肝脏, vnet, CT, 3D

Abstract: Segmentation of 3D medical images is an important step in radiotherapy planning.In clinical practice,computed tomography is widely used for 3D medical image segmentation of the liver and liver tumours.Due to the complex edge structure and texture features of the liver,liver segmentation is still a challenging task.To address this problem,an improved vnet model for accurate segmentation of 3D liver CT images is proposed.Firstly,the liver CT images are truncated and resampled with HU values to complete the preprocessing of the 3D dataset.Meanwhile,the convolution kernel in the vnet decoder and encoder is replaced with an SG module,which is a combination of depthwise convolution and pointwise convolution,to reduce the number of parameters in the network model.Comparative experiments with the vnet model show that the proposed method is generally superior in the evaluation of the liver segmentation dataset,with a Dice coefficient of 94.93%,an improvement of 3.49% over the vnet model,greatly reducing the number of parameters of the model,while the method also shows good robustness and achieves superior segmentation results on the MSD spleen segmentation dataset and COVID-19 dataset.

Key words: Image segmentation, Liver, vnet, CT, 3D

中图分类号: 

  • TP391
[1]LIU Z,SONG Y Q,SHENG V S,et al.Liver CT sequence segmentation based with improved U-Net and graph cut[J].Expert Systems with Applications,2019,126(15):54-63.
[2]MOGHBEL M,MASHOHOR S,MAHMUD R,et al.Review ofliver segmentation and computer assisted detection/diagnosis methods in computed tomography[J].Artificial Intelligence Review,2018,50(4):497-537.
[3]WANG L D,WANG J,HU C J,et al.Liver Segmentation by Using an Optimal Framework for CT Images[J].Chinese Journal of Computers,2016,39(7):1477-1489.
[4]LIU Z,ZHANG X L,SONG Y Q,et al.Combining improved U-Net and Morphsnakes for liver segmentation[J].Journal of Image and Graphics,2018,23(8):1254-1262.
[5]KANG X D.Medical Image Processing[M].Beijing:People’s Medical Publishing House,2009:99-100.
[6]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.
[7]RONNEBERGER O,FISCHER P,BROX T.U-net:Convolu-tional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-assisted Intervention.Cham:Springer,2015:234-241.
[8]TRAN S T,CHENG C H,LIU D G.A Multiple Layer U-Net,Un-Net,for Liver and Liver Tumor Segmentation in CT[J].IEEE Access,2020,9:3752-3764.
[9]ÇIÇEK Ö,ABDULKADIR A,LIENKAMP S S,et al.3D U-Net:learning dense volumetric segmentation from sparse annotation[C]//Medical Image Computing and Computer-Assisted Intervention(MICCAI 2016).Athens,Greece,2016:424-432.
[10]MILLETARI F,NAVAB N,AHMADI S A.V-net:Fully convolutional neural networks for volumetric medical image segmentation[C]//2016 Fourth International Conference on 3D Vision(3DV).2016:565-571.
[11]ZHANG L,ZHANG J M,SHEN P Y,et al.Block Level Skip Connections Across Cascaded V-Net for Multi-Organ Segmentation[J].IEEE Transactions on Medical Imaging,2020,39(9):2782-2793.
[12]IMRAN A-A-Z,HATAMIZADEH A,ANANTH S P,et al.Automatic segmentation of pulmonary lobes using a progressive dense V-network[C]//Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support.Granada,Spain,2018:282-290.
[13]ZHANG L,ZHANG J M,SHEN P Y,et al.Block Level Skip Connections Across Cascaded V-Net for Multi-Organ Segmentation[J].IEEE Transactions on Medical Imaging,2020,39(9):2782-2793.
[14]WANG T,YANG J,JI Z,et al.Probabilistic diffusion for interactive image segmentation[J].IEEE Transactions on Image Processing,2018,28(1):330-342.
[15]HU T,ZHU Y X,TIAN L,et al.Lightweight convolutionalneural network architecture for mobile platforms[J].Computer Engineering,2019,45(1):17-22.
[16]HOWARD A G,ZHU M,CHEN B,et al.Mobilenets:efficient convolutional neural networks for mobile vision applications[EB/OL].[2017-11-20].https://arxiv.org/abs/1704.04861.
[17]SHEN H Y,WU Y.MSFA-NET-based liver CT image segmentation method[J].Journal of Frontiers of Computer Science and Technology,2023,17(3):646-656.
[18]MA J L,DENG Y Y,MA Z P.Review of the deep learningmethod of liver tumor CT image[J].Journal of Image andGraphics,2020,25(10):2024-2046.
[19]TAHA A A,HANBURY A.Metrics for evaluating 3D medical image segmentation:analysis,selection,and tool[J].BMC Me-dical Imaging,2015,15(1):1-28.
[20]YAN Q,WANG W W.Division of liver and liver tumors based on conditional energy confrontation[J].Computer Engineering and Applications,2021,57(11):179-184.
[21]QIN W,WU J,HAN F,et al.Superpixel-based and boundary-sensitive convolutional neural network for automated liver segmentation[J].Physics in Medicine & Biology,2018,63(9):095017
[22]HAN X.Automatic Liver Lesion Segmentation Using A DeepConvolutional Neural Network Method[J].arXiv:1704.07239,2017.
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