Medical Images Anomaly Detection for Imbalanced Datasets with Multi-scale Normalizing Flow

Yufeng Xiao1, 2, 3, Xueting Huang1, 2, 3, Wei Liang1, 2, 3, Jingnian Liu1, 2, 3, Yuxiang Chen1, 2, 3, Rui Xie1, 2, 3, Kuanching Li1, 2, 3 and Nam Ling4

  1. School of Computer Engineering and Science, Hunan University of Science and Technology
    Xiangtan 411100, China
  2. Sanya Research Institute, Hunan University of Science and Technology
    Sanya 572024, China
  3. Hunan Key Laboratory for Service Computing and Novel Software Technology
    Xiangtan 411201, China
    hnxiaoyf@hnust.edu.cn, HXuetingmail@163.com, wliang@hnust.edu.cn, jingnianl@mail.hnust.edu.cn, chenyuxiang@hnust.edu.cn, xierui54@mail.hnust.edu.cn, aliric@hnust.edu.cn
  4. Department of Computer Science and Engineering, Santa Clara University
    USA
    nling@scu.edu

Abstract

Due to the substantial feature extraction and end-to-end learning capability, deep learning has been widely used in intelligent medical image detection. However, amount of parameters in these models relies on the number of labeled training data, which influences the performance. Due to this reason, we propose a novel unsupervised medical image detection model named Multi-Scale Normalizing Flow (MS-NF). First, a fusion backbone network is applied to extract the multi-scale feature maps, which capture the different scale features of the anomalies. Second, normalizing flow transfers the abnormal distribution into the normal distribution hidden in the latent space, which is used for anomaly detection. To further improve the detection performance, channel and spatial convolutional attention mechanisms are integrated to make the model focus on the anomalous region by a shared network. Experimental results obtained on brain tumor MRI and ISIC2018 datasets show that MS-NF improves the pixel-level AUC index by 9% compared to existing medical image detection models, also performing well on small-scale data with efficient training and inference.

Key words

Anomaly Detection, Unsupervised learning, Attention Mechanisms, Normalizing Flow, Medical Images

Digital Object Identifier (DOI)

https://doi.org/10.2298/CSIS240227001X

Publication information

Volume 22, Issue 1 (January 2025)
Year of Publication: 2025
ISSN: 2406-1018 (Online)
Publisher: ComSIS Consortium

Full text

DownloadAvailable in PDF
Portable Document Format

How to cite

Xiao, Y., Huang, X., Liang, W., Liu, J., Chen, Y., Xie, R., Li, K., Ling, N.: Medical Images Anomaly Detection for Imbalanced Datasets with Multi-scale Normalizing Flow. Computer Science and Information Systems, Vol. 22, No. 1, 219–238. (2025), https://doi.org/10.2298/CSIS240227001X