Patched Diffusion Models for Unsupervised Anomaly Detection in Brain MRI

Finn Behrendt, Debayan Bhattacharya, Julia Krüger, Roland Opfer, Alexander Schlaefer
Medical Imaging with Deep Learning, PMLR 227:1019-1032, 2024.

Abstract

The use of supervised deep learning techniques to detect pathologies in brain MRI scans can be challenging due to the diversity of brain anatomy and the need for annotated data sets. An alternative approach is to use unsupervised anomaly detection, which only requires sample-level labels of healthy brains to create a reference representation. This reference representation can then be compared to unhealthy brain anatomy in a pixel-wise manner to identify abnormalities. To accomplish this, generative models are needed to create anatomically consistent MRI scans of healthy brains. While recent diffusion models have shown promise in this task, accurately generating the complex structure of the human brain remains a challenge. In this paper, we propose a method that reformulates the generation task of diffusion models as a patch-based estimation of healthy brain anatomy, using spatial context to guide and improve reconstruction. We evaluate our approach on data of tumors and multiple sclerosis lesions and demonstrate a relative improvement of 25.1% compared to existing baselines.

Cite this Paper

BibTeX
@InProceedings{pmlr-v227-behrendt24a, title = {Patched Diffusion Models for Unsupervised Anomaly Detection in Brain MRI}, author = {Behrendt, Finn and Bhattacharya, Debayan and Kr\"uger, Julia and Opfer, Roland and Schlaefer, Alexander}, booktitle = {Medical Imaging with Deep Learning}, pages = {1019--1032}, year = {2024}, editor = {Oguz, Ipek and Noble, Jack and Li, Xiaoxiao and Styner, Martin and Baumgartner, Christian and Rusu, Mirabela and Heinmann, Tobias and Kontos, Despina and Landman, Bennett and Dawant, Benoit}, volume = {227}, series = {Proceedings of Machine Learning Research}, month = {10--12 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v227/behrendt24a/behrendt24a.pdf}, url = {https://proceedings.mlr.press/v227/behrendt24a.html}, abstract = {The use of supervised deep learning techniques to detect pathologies in brain MRI scans can be challenging due to the diversity of brain anatomy and the need for annotated data sets. An alternative approach is to use unsupervised anomaly detection, which only requires sample-level labels of healthy brains to create a reference representation. This reference representation can then be compared to unhealthy brain anatomy in a pixel-wise manner to identify abnormalities. To accomplish this, generative models are needed to create anatomically consistent MRI scans of healthy brains. While recent diffusion models have shown promise in this task, accurately generating the complex structure of the human brain remains a challenge. In this paper, we propose a method that reformulates the generation task of diffusion models as a patch-based estimation of healthy brain anatomy, using spatial context to guide and improve reconstruction. We evaluate our approach on data of tumors and multiple sclerosis lesions and demonstrate a relative improvement of 25.1% compared to existing baselines.} }
Endnote
%0 Conference Paper %T Patched Diffusion Models for Unsupervised Anomaly Detection in Brain MRI %A Finn Behrendt %A Debayan Bhattacharya %A Julia Krüger %A Roland Opfer %A Alexander Schlaefer %B Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2024 %E Ipek Oguz %E Jack Noble %E Xiaoxiao Li %E Martin Styner %E Christian Baumgartner %E Mirabela Rusu %E Tobias Heinmann %E Despina Kontos %E Bennett Landman %E Benoit Dawant %F pmlr-v227-behrendt24a %I PMLR %P 1019--1032 %U https://proceedings.mlr.press/v227/behrendt24a.html %V 227 %X The use of supervised deep learning techniques to detect pathologies in brain MRI scans can be challenging due to the diversity of brain anatomy and the need for annotated data sets. An alternative approach is to use unsupervised anomaly detection, which only requires sample-level labels of healthy brains to create a reference representation. This reference representation can then be compared to unhealthy brain anatomy in a pixel-wise manner to identify abnormalities. To accomplish this, generative models are needed to create anatomically consistent MRI scans of healthy brains. While recent diffusion models have shown promise in this task, accurately generating the complex structure of the human brain remains a challenge. In this paper, we propose a method that reformulates the generation task of diffusion models as a patch-based estimation of healthy brain anatomy, using spatial context to guide and improve reconstruction. We evaluate our approach on data of tumors and multiple sclerosis lesions and demonstrate a relative improvement of 25.1% compared to existing baselines.
APA
Behrendt, F., Bhattacharya, D., Krüger, J., Opfer, R. & Schlaefer, A.. (2024). Patched Diffusion Models for Unsupervised Anomaly Detection in Brain MRI. Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 227:1019-1032 Available from https://proceedings.mlr.press/v227/behrendt24a.html.

Related Material