您的位置:山东大学 -> 科技期刊社 -> 《山东大学学报(工学版)》

山东大学学报 (工学版) ›› 2020, Vol. 50 ›› Issue (6): 76-81.doi: 10.6040/j.issn.1672-3961.0.2020.230

• • 上一篇    

融合残差块注意力机制和生成对抗网络的海马体分割

张月芳,邓红霞*,呼春香,钱冠宇,李海芳   

  1. 太原理工大学信息与计算机学院, 山西 晋中 030600
  • 发布日期:2020-12-15
  • 作者简介:张月芳(1996— ),女,山西吕梁人,硕士研究生,主要研究方向为医学图像处理. E-mail:792331696@qq.com. *通信作者简介:邓红霞(1976— ),女,山西太原人,副教授,博士,主要研究方向为计算机视觉. E-mail: denghongxia@tyut.edu.cn
  • 基金资助:
    国家自然科学基金资助项目(61873178,61976150);山西省自然科学基金资助项目(201801D21135);山西省重点研发计划(国际科技合作)(201803D421047)

Hippocampal segmentation combining residual attention mechanism and generative adversarial networks

ZHANG Yuefang, DENG Hongxia*, HU Chunxiang, QIAN Guanyu, LI Haifang   

  1. School of Information and Computer, Taiyuan University of Technology, Jinzhong 030600, Shanxi, China
  • Published:2020-12-15

摘要: 研究一种基于改进的生成对抗网络的深度学习方法对海马体进行分割。提出不同的卷积配置,以捕获由分割网络获得的信息。提出以Pixel2Pixel为基本架构的生成对抗网络模型,生成模型结合残差网络以及注意力机制的编解码结构以捕获更多细节信息。判别网络采用卷积神经网络对生成模型的分割结果和专家分割结果进行判别。经过生成模型和对抗模型不断地传递其损失,使生成模型达到分割海马体的最优状态。使用来自ADNI数据集130名健康受试者的T1加权MRI扫描和相关海马标签作为训练和测试数据,以相似度系数作为评价指标,准确率达到89.46%。试验结果表明,该网络模型可以实现高效地自动分割海马体,对于阿尔茨海默症等疾病的正确诊断具有重要的现实意义。

关键词: 磁共振图像, 生成对抗网络, 残差网络, 注意力机制, 海马体

Abstract: This research discussed a deep learning method based on an improved generative adversarial network to segment the hippocampus. Different convolutional configurations were proposed to capture the information obtained by the segmentation network. The generative adversarial network based on Pixel2Pixel was proposed. The generator was a codec structure combining residual network and attention mechanism to capture more detailed information. The discriminator used a convolutional neural network to discriminate the segmentation results of the generated model and the expert segmentation results. Through generator and discriminator continuously transmitted losses, the generator reached the optimal state of segmenting the hippocampus. Using the T1-weighted MRI scans and related hippocampus labels of 130 healthy subjects from the ADNI data set as training and test data, and the similarity coefficient as the evaluation index, the accuracy rate reached 89.46%. Results showed that the network model could achieve efficient automatic segmentation of hippocampus, which had important practical significance for the correct diagnosis of diseases such as Alzheimer's disease.

Key words: magnetic resonance imaging, generative adversarial network, residual network, attention mechanism, hippocampus

中图分类号: 

  • TP3
[1] 廖南星,周世斌,张国鹏,程德强. 基于类激活映射-注意力机制的图像描述方法[J]. 山东大学学报 (工学版), 2020, 50(4): 28-34.
[2] 蔡国永,贺歆灏,储阳阳. 基于空间注意力和卷积神经网络的视觉情感分析[J]. 山东大学学报 (工学版), 2020, 50(4): 8-13.
[3] 李春阳,李楠,冯涛,王朱贺,马靖凯. 基于深度学习的洗衣机异常音检测[J]. 山东大学学报 (工学版), 2020, 50(2): 108-117.
[4] 朱安,徐初. 一种使用并行交错采样进行超分辨的方法[J]. 山东大学学报 (工学版), 2020, 50(2): 10-16,26.
[5] 黄劲潮. 深度残差特征与熵能量优化运动目标跟踪算法[J]. 山东大学学报 (工学版), 2019, 49(4): 14-23.
[6] 刘振丙,方旭升,杨辉华,蓝如师. 基于多尺度残差神经网络的阿尔茨海默病诊断分类[J]. 山东大学学报 (工学版), 2018, 48(6): 1-7, 18.
[7] 王海军,柳明. 克服灰度不均匀性的脑MR图像分割及去偏移场模型[J]. 山东大学学报(工学版), 2011, 41(3): 36-41.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!