Unsupervised Anomaly Localization with Structural Feature-Autoencoders

TL;DR

This paper introduces a novel unsupervised anomaly localization method using structural feature-autoencoders that improve detection accuracy in medical images by leveraging multi-channel feature spaces and structural similarity loss.

Contribution

It proposes a new feature-mapping and autoencoder framework that enhances anomaly detection by capturing structural and contrast differences, outperforming existing intensity-based methods.

Findings

Significant performance improvement on brain MRI datasets

Effective detection of anomalies beyond intensity differences

Utilization of structural similarity loss enhances localization accuracy

Abstract

Unsupervised Anomaly Detection has become a popular method to detect pathologies in medical images as it does not require supervision or labels for training. Most commonly, the anomaly detection model generates a "normal" version of an input image, and the pixel-wise $l^p$-difference of the two is used to localize anomalies. However, large residuals often occur due to imperfect reconstruction of the complex anatomical structures present in most medical images. This method also fails to detect anomalies that are not characterized by large intensity differences to the surrounding tissue. We propose to tackle this problem using a feature-mapping function that transforms the input intensity images into a space with multiple channels where anomalies can be detected along different discriminative feature maps extracted from the original image. We then train an Autoencoder model in this space using structural similarity loss that does not only consider differences in intensity but also in contrast and structure. Our method significantly increases performance on two medical data sets for brain MRI. Code and experiments are available at https://github.com/FeliMe/feature-autoencoder