Unsupervised Anomaly Localization using Variational Auto-Encoders
This addresses the need for assumption-free anomaly detection in medical imaging to assist radiologists, though it is incremental as it builds on existing VAE-based localization methods.
The paper tackled the problem of unsupervised anomaly localization in medical images by proposing a method that complements reconstruction-based localization with a KL-divergence term, resulting in outperforming state-of-the-art VAE-based methods across many hyperparameter settings and showing competitive max performance on datasets including FashionMNIST and over 1000 healthy and 250 brain tumor patient images.
An assumption-free automatic check of medical images for potentially overseen anomalies would be a valuable assistance for a radiologist. Deep learning and especially Variational Auto-Encoders (VAEs) have shown great potential in the unsupervised learning of data distributions. In principle, this allows for such a check and even the localization of parts in the image that are most suspicious. Currently, however, the reconstruction-based localization by design requires adjusting the model architecture to the specific problem looked at during evaluation. This contradicts the principle of building assumption-free models. We propose complementing the localization part with a term derived from the Kullback-Leibler (KL)-divergence. For validation, we perform a series of experiments on FashionMNIST as well as on a medical task including >1000 healthy and >250 brain tumor patients. Results show that the proposed formalism outperforms the state of the art VAE-based localization of anomalies across many hyperparameter settings and also shows a competitive max performance.