Unsupervised abnormality detection through mixed structure regularization (MSR) in deep sparse autoencoders.

Journal: Medical physics
Published Date:

Abstract

PURPOSE: The purpose of this study is to introduce and evaluate the mixed structure regularization (MSR) approach for a deep sparse autoencoder aimed at unsupervised abnormality detection in medical images. Unsupervised abnormality detection based on identifying outliers using deep sparse autoencoders is a very appealing approach for computer-aided detection systems as it requires only healthy data for training rather than expert annotated abnormality. However, regularization is required to avoid overfitting of the network to the training data.

Authors

  • Moti Freiman
    Philips Medical Systems Technologies Ltd., Advanced Technologies Center, Haifa, 3100202, Israel.
  • Ravindra Manjeshwar
    CT BU, Global Advanced Technology, Philips Healthcare, 100 Park Ave, Highland Hills, OH, 44122, USA.
  • Liran Goshen
    Philips Medical Systems Technologies Ltd., Advanced Technologies Center, Haifa, 3100202, Israel.