Masked Autoencoders for Unsupervised Anomaly Detection in Medical Images

TL;DR

This paper introduces a novel unsupervised anomaly detection method for medical images using masked autoencoders to learn normal structures and a pseudo-abnormal module to generate training positives, outperforming existing methods.

Contribution

The work presents a new framework combining masked autoencoders with a pseudo-abnormal module for effective unsupervised anomaly detection in medical imaging.

Findings

Outperforms four state-of-the-art anomaly detection methods on BRATS2020 and LUNA16 datasets.

Effectively learns normal tissue structures with masked autoencoders.

Generates pseudo-abnormal samples to improve supervised training of the anomaly classifier.

Abstract

Pathological anomalies exhibit diverse appearances in medical imaging, making it difficult to collect and annotate a representative amount of data required to train deep learning models in a supervised setting. Therefore, in this work, we tackle anomaly detection in medical images training our framework using only healthy samples. We propose to use the Masked Autoencoder model to learn the structure of the normal samples, then train an anomaly classifier on top of the difference between the original image and the reconstruction provided by the masked autoencoder. We train the anomaly classifier in a supervised manner using as negative samples the reconstruction of the healthy scans, while as positive samples, we use pseudo-abnormal scans obtained via our novel pseudo-abnormal module. The pseudo-abnormal module alters the reconstruction of the normal samples by changing the intensity of several regions. We conduct experiments on two medical image data sets, namely BRATS2020 and LUNA16 and compare our method with four state-of-the-art anomaly detection frameworks, namely AST, RD4AD, AnoVAEGAN and f-AnoGAN.