A Case for the Score: Identifying Image Anomalies using Variational Autoencoder Gradients

Through training on unlabeled data, anomaly detection has the potential to impact computer-aided diagnosis by outlining suspicious regions. Previous work on deep-learning-based anomaly detection has primarily focused on the reconstruction error. We argue instead, that pixel-wise anomaly ratings deri...

Full description

Saved in:
Bibliographic Details
Main Authors Zimmerer, David, Petersen, Jens, Kohl, Simon A. A, Maier-Hein, Klaus H
Format Journal Article
LanguageEnglish
Published 28.11.2019
Subjects
Computer Science - Learning
Statistics - Machine Learning
Online AccessGet full text

Cover

More Information
Summary:Through training on unlabeled data, anomaly detection has the potential to impact computer-aided diagnosis by outlining suspicious regions. Previous work on deep-learning-based anomaly detection has primarily focused on the reconstruction error. We argue instead, that pixel-wise anomaly ratings derived from a Variational Autoencoder based score approximation yield a theoretically better grounded and more faithful estimate. In our experiments, Variational Autoencoder gradient-based rating outperforms other approaches on unsupervised pixel-wise tumor detection on the BraTS-2017 dataset with a ROC-AUC of 0.94.
DOI:10.48550/arxiv.1912.00003