Beyond Model Stochasticity: External Randomness for Diffusion-Based Unsupervised MRI Anomaly Detection
Journal of Imaging Informatics in Medicine · 17.11.2026
DOI: 10.1007/s10278-026-02500-2 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Imaging Informatics in Medicine, online erschienen am 17.11.2026. Verfasst von Changsuk Oh und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiological Sciences, David Geffen School of Medicine, University of California, Los Angeles, CA, 90095, USA. changsukoh@mednet.ucla.e…. Die Arbeit nennt 50 Literaturangaben. Gefördert von: Siemens Medical Solutions USA.
Abstract (Originalsprache)
Unsupervised anomaly detection (UAD) in MRI aims to identify abnormal regions through a model trained on normal MR images. Recently, reconstruction-based methods leveraging denoising diffusion probabilistic models (DDPMs) as base generative models (GMs) have gained attention. These approaches typically compute an anomaly score by measuring the difference between a query image and a single reconstructed counterpart (or low-diversity counterparts). However, these approaches typically struggle to differentiate erroneous reconstructions of normal regions from actual abnormalities. To address this problem, we propose a novel approach that enhances anomaly detection by leveraging statistical discrepancy from high-diversity reconstructions. We employ image inpainting with randomly sampled masks to generate diversified reconstructions. We can produce diverse reconstructions through image inpainting even when the input masks differ only slightly, depending on the information retained in the unmasked regions. By sampling multiple random input masks, we can obtain a diverse set of reconstructions, enabling more reliable distinction between normal and abnormal regions. In abnormal lesions, large discrepancies consistently appear across reconstructions, whereas normal regions exhibit irregular and less pronounced differences. Since existing anomaly definitions often fail to adequately capture statistical discrepancy in inpainted images, we introduce a new anomaly score. We define a pixel-wise anomaly score as the average of the normalized differences between the query image and the inpainted images generated from randomly sampled masks covering the pixel. Experimental results demonstrate that our method, with its reconstruction diversification strategy and novel anomaly score, substantially outperforms existing reconstruction-based approaches.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiological Sciences, David Geffen School of Medicine, University of California, Los Angeles, CA, 90095, USA. changsukoh@mednet.ucla.edu.
- Department of Radiation Oncology, David Geffen School of Medicine, University of California, Los Angeles, CA, 90095, USA. changsukoh@mednet.ucla.edu.
- Department of Radiological Sciences, David Geffen School of Medicine, University of California, Los Angeles, CA, 90095, USA.
- School of Communication and Information Engineering, Shanghai University, Shanghai, China.
- Department of Radiation Oncology, David Geffen School of Medicine, University of California, Los Angeles, CA, 90095, USA.
Förderung
- Siemens Medical Solutions USA
Zitieren
APA
Oh, C., He, Q., Zhao, K., Pang, K., Bang, C., Jank, E., Raj, V., Savjani, R., & Sung, K. (2026). Beyond Model Stochasticity: External Randomness for Diffusion-Based Unsupervised MRI Anomaly Detection. Journal of Imaging Informatics in Medicine. https://doi.org/10.1007/s10278-026-02500-2
Vancouver
Oh C, He Q, Zhao K, Pang K, Bang C, Jank E, et al. Beyond Model Stochasticity: External Randomness for Diffusion-Based Unsupervised MRI Anomaly Detection. Journal of Imaging Informatics in Medicine. 2026. doi:10.1007/s10278-026-02500-2
BibTeX
@article{Oh2026,
author = {Oh, Changsuk and He, Qiudi and Zhao, Kai and Pang, Kaifeng and Bang, Chulmin and Jank, Erika and Raj, Vinith and Savjani, Ricky and Sung, Kyunghyun},
title = {Beyond Model Stochasticity: External Randomness for Diffusion-Based Unsupervised MRI Anomaly Detection},
journal = {Journal of Imaging Informatics in Medicine},
year = {2026},
doi = {10.1007/s10278-026-02500-2}
}
EUVIMED