Context-encoding Variational Autoencoder for Unsupervised Anomaly Detection

TL;DR

This paper introduces the ceVAE, a novel unsupervised anomaly detection model that combines reconstruction and density-based scoring, significantly improving medical image anomaly detection performance.

Contribution

The paper proposes the context-encoding variational autoencoder (ceVAE), integrating density-based scoring with reconstruction to enhance anomaly detection accuracy and comparability.

Findings

Achieved ROC-AUC of 0.95 on BraTS-2017

Achieved ROC-AUC of 0.89 on ISLES-2015

Outperformed state-of-the-art methods significantly

Abstract

Unsupervised learning can leverage large-scale data sources without the need for annotations. In this context, deep learning-based auto encoders have shown great potential in detecting anomalies in medical images. However, state-of-the-art anomaly scores are still based on the reconstruction error, which lacks in two essential parts: it ignores the model-internal representation employed for reconstruction, and it lacks formal assertions and comparability between samples. We address these shortcomings by proposing the Context-encoding Variational Autoencoder (ceVAE) which combines reconstruction- with density-based anomaly scoring. This improves the sample- as well as pixel-wise results. In our experiments on the BraTS-2017 and ISLES-2015 segmentation benchmarks, the ceVAE achieves unsupervised ROC-AUCs of 0.95 and 0.89, respectively, thus outperforming state-of-the-art methods by a considerable margin.