4.6 Article

Unsupervised Deep Anomaly Detection for Medical Images Using an Improved Adversarial Autoencoder

期刊

JOURNAL OF DIGITAL IMAGING
卷 35, 期 2, 页码 153-161

出版社

SPRINGER
DOI: 10.1007/s10278-021-00558-8

关键词

Deep anomaly detection; Medical images; Unsupervised; Adversarial autoencoders; Chain of convolution block

资金

  1. research project of Taizhou University [Z2018046]
  2. Science and Technology Program of Taizhou [2003gy12, 2003gy04]
  3. National Natural Science Foundation of China [61976149]
  4. Zhejiang Provincial Natural Science Foundation of China [LZ20F020002]
  5. Humanities and Social Science Project of the Chinese Ministry of Education [20YJAZH033]

向作者/读者索取更多资源

This paper proposes an unsupervised learning method for deep anomaly detection based on an improved adversarial autoencoder. By using a module called chain of convolutional block (CCB), the proposed method provides considerable advantages in capturing the distribution of normal samples in image space and latent vector space.
Anomaly detection has been applied in the various disease of medical practice, such as breast cancer, retinal, lung lesion, and skin disease. However, in real-world anomaly detection, there exist a large number of healthy samples, and but very few sick samples. To alleviate the problem of data imbalance in anomaly detection, this paper proposes an unsupervised learning method for deep anomaly detection based on an improved adversarial autoencoder, in which a module called chain of convolutional block (CCB) is employed instead of the conventional skip-connections used in adversarial autoencoder. Such CCB connections provide considerable advantages via direct connections, not only preserving both global and local information but also alleviating the problem of semantic disparity between the encoding features and the corresponding decoding features. The proposed method is thus able to capture the distribution of normal samples within both image space and latent vector space. By means of minimizing the reconstruction error within both spaces during training phase, higher reconstruction error during test phase is indicative of an anomaly. Our method is trained only on the healthy persons in order to learn the distribution of normal samples and can detect sick samples based on high deviation from the distribution of normality in an unsupervised way. Experimental results for multiple datasets from different fields demonstrate that the proposed method yields superior performance to state-of-the-art methods.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.6
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据