Ensembled Cold-Diffusion Restorations for Unsupervised Anomaly Detection

TL;DR

This paper introduces a novel unsupervised anomaly detection method combining generative cold-diffusion with synthetic anomaly generation, achieving state-of-the-art results in Brain MRI datasets by ensemble restoration techniques.

Contribution

It proposes a new cold-diffusion pipeline trained to restore normal images from synthetic corruptions, integrating a novel anomaly score and synthetic anomaly generation method called DAG.

Findings

Surpasses prior state-of-the-art in Brain MRI anomaly detection

Effective ensemble of restorations conditioned on abnormality levels

Introduces a novel synthetic anomaly generation procedure (DAG)

Abstract

Unsupervised Anomaly Detection (UAD) methods aim to identify anomalies in test samples comparing them with a normative distribution learned from a dataset known to be anomaly-free. Approaches based on generative models offer interpretability by generating anomaly-free versions of test images, but are typically unable to identify subtle anomalies. Alternatively, approaches using feature modelling or self-supervised methods, such as the ones relying on synthetically generated anomalies, do not provide out-of-the-box interpretability. In this work, we present a novel method that combines the strengths of both strategies: a generative cold-diffusion pipeline (i.e., a diffusion-like pipeline which uses corruptions not based on noise) that is trained with the objective of turning synthetically-corrupted images back to their normal, original appearance. To support our pipeline we introduce a novel synthetic anomaly generation procedure, called DAG, and a novel anomaly score which ensembles restorations conditioned with different degrees of abnormality. Our method surpasses the prior state-of-the art for unsupervised anomaly detection in three different Brain MRI datasets.