MIM-OOD: Generative Masked Image Modelling for Out-of-Distribution Detection in Medical Images

TL;DR

MIM-OOD introduces a fast, effective method for unsupervised out-of-distribution detection in medical images by replacing autoregressive models with specialized transformers for token identification and in-painting.

Contribution

The paper presents MIM-OOD, a novel approach using masked image modeling with transformers to improve speed and accuracy in OOD detection for medical images.

Findings

MIM-OOD outperforms autoregressive models in accuracy (DICE 0.458 vs 0.301)

MIM-OOD achieves a 25x speedup in inference time

Effective detection of brain MRI anomalies

Abstract

Unsupervised Out-of-Distribution (OOD) detection consists in identifying anomalous regions in images leveraging only models trained on images of healthy anatomy. An established approach is to tokenize images and model the distribution of tokens with Auto-Regressive (AR) models. AR models are used to 1) identify anomalous tokens and 2) in-paint anomalous representations with in-distribution tokens. However, AR models are slow at inference time and prone to error accumulation issues which negatively affect OOD detection performance. Our novel method, MIM-OOD, overcomes both speed and error accumulation issues by replacing the AR model with two task-specific networks: 1) a transformer optimized to identify anomalous tokens and 2) a transformer optimized to in-paint anomalous tokens using masked image modelling (MIM). Our experiments with brain MRI anomalies show that MIM-OOD substantially outperforms AR models (DICE 0.458 vs 0.301) while achieving a nearly 25x speedup (9.5s vs 244s).