《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (1): 316-324.DOI: 10.11772/j.issn.1001-9081.2021010200

• 前沿与综合应用 • 上一篇    

基于多影像中心磁共振成像数据的半监督膝盖异常分类

吴洁1,2, 张师天3, 谢海滨1,2, 杨光1,2()   

  1. 1.华东师范大学 物理与电子科学学院, 上海 200241
    2.上海市磁共振重点实验室(华东师范大学), 上海 200062
    3.爱丁堡大学 物理与天文学院, 爱丁堡 EH8 9YL, 英国
  • 收稿日期:2021-02-03 修回日期:2021-04-28 接受日期:2020-04-29 发布日期:2021-05-07 出版日期:2022-01-10
  • 通讯作者: 杨光
  • 作者简介:吴洁(1995—),女,江苏扬州人,硕士研究生,主要研究方向:机器学习、医学图像处理
    张师天(1995—),男,江苏扬州人,硕士,主要研究方向:机器学习、非平衡态统计物理
    谢海滨(1967—),男,江苏泗洪人,工程师,博士,主要研究方向:机器学习、医学图像处理
    杨光(1968—),男,安徽蚌埠人,副研究员,博士,主要研究方向:机器学习、医学图像处理。
  • 基金资助:
    国家自然科学基金资助项目(61731009)

Semi-supervised knee abnormality classification based on multi-imaging center MRI data

Jie WU1,2, Shitian ZHANG3, Haibin XIE1,2, Guang YANG1,2()   

  1. 1.School of Physics and Electronic Science,East China Normal University,Shanghai 200241,China
    2.Shanghai Key Laboratory of Magnetic Resonance (East China Normal University),Shanghai 200062,China
    3.School of Physics and Astronomy,University of Edinburgh,Edinburgh EH8 9YL,UK
  • Received:2021-02-03 Revised:2021-04-28 Accepted:2020-04-29 Online:2021-05-07 Published:2022-01-10
  • Contact: Guang YANG
  • About author:WU Jie, born in 1995, M. S. candidate. Her research interests include machine learning, medical image processing.
    ZHANG Shitian, born in 1995, M. S. His research interests include machine learning, non-equilibrium statistical physics.
    XIE Haibin, born in 1967, Ph. D., engineer. His research interests include machine learning, medical image processing.
    YANG Guang, born in 1968, Ph. D., associate research fellow. His research interests include machine learning, medical image processing.
  • Supported by:
    National Natural Science Foundation of China(61731009)

摘要:

针对大量数据手工标记的繁重性和单一影像中心磁共振成像(MRI)数据的有限性问题,提出了一种利用多影像中心有标签与无标签MRI数据的用于磁共振的半监督学习(MRSSL)方法,并将其应用在膝盖异常分类任务中。首先,运用了数据扩增方法来提供模型所需的归纳偏置;接着,融合了分类损失项和一致性损失项来约束人工神经网络并使之从数据中提取出具有辨别力的特征;然后,将这些特征用于MRI膝盖异常分类。此外,也提出了对应的仅利用有标签数据的完全监督学习(MRSL)方法。在给出同样的有标签样本时,将MRSL与MRSSL进行了比较,结果表明MRSSL的模型分类性能与泛化性能明显优于MRSL。最后,将MRSSL与其他半监督学习方法进行了比较。结果表明数据扩增在性能提升中起到了重要作用,并且MRSSL凭借更强的MRI数据包容性取得了最优的膝盖异常分类性能。

关键词: 深度学习, 半监督学习, 磁共振成像, 多影像中心数据, 膝盖异常分类, MRNet数据集, OAI数据集

Abstract:

The manual labeling of abundant data is laborious and the amount of Magnetic Resonance Imaging (MRI) data from a single imaging center is limited. Concerning the above problems, a Magnetic Resonance Semi-Supervised Learning (MRSSL) method utilizing multi-imaging center labeled and unlabeled MRI data was proposed and applied to knee abnormality classification. Firstly, data augmentation was used to provide the inductive bias required by the model . Next, the classification loss and the consistency loss were combined to constraint an artificial neural network to extract the discriminative features from the data. Then, the features were used for the MRI knee abnormality classification. Additionally, the corresponding Magnetic Resonance Supervised Learning (MRSL) method only using labeled samples was proposed and compared with MRSSL for the same labeled samples. The results demonstrate that MRSSL surpasses MRSL in both model classification performance and model generalization ability. Finally, MRSSL was compared with other semi-supervised learning methods. The results indicate that data augmentation plays an important role on performance improvement, and with stronger inclusiveness for MRI data, MRSSL outperforms others on the knee abnormality classification.

Key words: deep learning, semi-supervised learning, Magnetic Resonance Imaging (MRI), multi-imaging center data, knee abnormality classification, MRNet dataset, OAI (OsteoArthritis Initiative) dataset

中图分类号: