City Research Online

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

Marimont, S. N., Siomos, V. & Tarroni, G. ORCID: 0000-0002-0341-6138 (2024). MIM-OOD: Generative Masked Image Modelling for Out-of-Distribution Detection in Medical Images. In: Deep Generative Models. Third MICCAI Workshop, DGM4MICCAI 2023, 8-12 Oct 2023, Vancouver, Canada. doi: 10.1007/978-3-031-53767-7_4


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 (9.5 s vs 244 s).

Publication Type: Conference or Workshop Item (Paper)
Additional Information: This version of the contribution has been accepted for publication, after peer review but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: Use of this Accepted Version is subject to the publisher’s Accepted Manuscript terms of use
Publisher Keywords: out-of-distribution detection, unsupervised learning, masked, image modelling
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QM Human anatomy
R Medicine > RC Internal medicine
Departments: School of Science & Technology
School of Science & Technology > Computer Science
SWORD Depositor:
[thumbnail of MIM_OOD_Paper.pdf] Text - Accepted Version
This document is not freely accessible until 20 February 2025 due to copyright restrictions.

To request a copy, please use the button below.

Request a copy


Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email


Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login