计算机应用 ›› 2021, Vol. 41 ›› Issue (1): 280-285.DOI: 10.11772/j.issn.1001-9081.2020060895

所属专题: 前沿与综合应用

• 前沿与综合应用 • 上一篇    下一篇

深度学习在主动脉中膜变性病理图像分类中的应用

孙中杰1,2, 万涛1,2, 陈东3, 汪昊3, 赵艳丽3, 秦曾昌4   

  1. 1. 北京航空航天大学 生物与医学工程学院, 北京 100191;
    2. 北京航空航天大学 生物医学工程高精尖创新中心, 北京 100191;
    3. 首都医科大学附属北京安贞医院 病理科, 北京 100029;
    4. 北京航空航天大学 自动化科学与电气工程学院, 北京 100191
  • 收稿日期:2020-06-28 修回日期:2020-11-11 出版日期:2021-01-10 发布日期:2020-12-30
  • 通讯作者: 万涛
  • 作者简介:孙中杰(1995-),男,山西临汾人,硕士研究生,主要研究方向:计算机辅助诊断、医学图像分析;万涛(1978-),女,江西南昌人,副教授,博士,主要研究方向:医学图像分析、医疗人工智能;陈东(1967-),女,安徽宿州人,主任医师,博士,主要研究方向:心血管病理;汪昊(1994-),女,山东济宁人,硕士研究生,主要研究方向:心肺血管病理、心脏肿瘤病理;赵艳丽(1992-),女,河北承德人,硕士研究生,主要研究方向:心血管病理;秦曾昌(1979-),男,黑龙江明水人,副教授,博士,CCF会员,主要研究方向:人工智能。
  • 基金资助:
    国家自然科学基金资助项目(61876197);北京市医院管理局临床技术创新项目(XMLX201814);北京市自然科学基金资助项目(7192105)。

Application of deep learning in histopathological image classification of aortic medial degeneration

SUN Zhongjie1,2, WAN Tao1,2, CHEN Dong3, WANG Hao3, ZHAO Yanli3, QIN Zengchang4   

  1. 1. School of Biomedical Science and Medical Engineering, Beihang University, Beijing 100191, China;
    2. Beijing Advanced Innovation Center for Biomedical Engineering, Beihang University, Beijing 100191 China;
    3. Department of Pathology, Beijing Anzhen Hospital, Capital Medical University, Beijing 100029, China;
    4. School of Automation Science and Electrical Engineering, Beihang University, Beijing 100191, China
  • Received:2020-06-28 Revised:2020-11-11 Online:2021-01-10 Published:2020-12-30
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61876197), the Clinical Technical Innovation Project of Beijing Municipal Administration of Hospitals (XMLX201814), the Beijing Municipal Natural Science Foundation (7192105).

摘要: 胸主动脉瘤和夹层(TAAD)是严重的心血管疾病之一,而中膜变性(MD)的组织学改变对疾病的诊断及早期干预具有重要的临床意义。针对病理图像的高度复杂性使得MD的诊断过程耗时费力且一致性差的问题,提出了一种基于深度学习的病理图像分类方法,并将其应用于四种MD病变类型以进行性能验证。该方法使用了一种改进的基于GoogLeNet的卷积神经网络模型,首先采用迁移学习来将先验知识应用于TAAD病理图像的表达,然后使用Focal loss和L2正则化来解决数据不平衡问题,从而进一步优化模型性能。实验结果表明,所提模型的平均四分类准确率达到98.78%,表现出较好的泛化性能。可见所提方法可以有效地提升病理学家的诊断效率。

关键词: 深度学习, 卷积神经网络, 计算机辅助诊断, 非炎性主动脉中膜变性, 病理图像

Abstract: Thoracic Aortic Aneurysm and Dissection (TAAD) is one of the life-threatening cardiovascular diseases, and the histological changes of Medial Degeneration (MD) have important clinical significance for the diagnosis and early intervention of TAAD. Focusing on the issue that the diagnosis of MD is time-consuming and prone to poor consistency because of the great complexity in histological images, a deep learning based classification method of histological images was proposed, and it was applied to four types of MD pathological changes to verify its performance. In the method, an improved Convolutional Neural Network (CNN) model was employed based on the GoogLeNet. Firstly, transfer learning was adopted for applying the prior knowledge to the expression of TAAD histopathological images. Then, Focal loss and L2 regularization were utilized to solve the data imbalance problem, so as to optimize the model performance. Experimental results show that the proposed model is able to achieve the average accuracy of four-class classification of 98.78%, showing a good generalizability. It can be seen that the proposed method can effectively improve the diagnostic efficiency of pathologists.

Key words: deep learning, Convolutional Neural Network (CNN), Computer-Aided Diagnosis (CAD), non-inflammatory aortic medial degeneration, histopathological image

中图分类号: