Position Regression for Unsupervised Anomaly Detection

Florentin Bieder, Julia Wolleb, Robin Sandkühler, Philippe C. Cattin
Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, PMLR 172:160-172, 2022.

Abstract

In recent years, anomaly detection has become an essential field in medical image analysis. Most current anomaly detection methods for medical images are based on image reconstruction. In this work, we propose a novel anomaly detection approach based on coordinate regression. Our method estimates the position of patches within a volume, and is trained only on data of healthy subjects. During inference, we can detect and localize anomalies by considering the error of the position estimate of a given patch. We apply our method to 3D CT volumes and evaluate it on patients with intracranial haemorrhages and cranial fractures. The results show that our method performs well in detecting these anomalies. Furthermore, we show that our method requires less memory than comparable approaches that involve image reconstruction. This is highly relevant for processing large 3D volumes, for instance, CT or MRI scans.

Cite this Paper

BibTeX
@InProceedings{pmlr-v172-bieder22a, title = {Position Regression for Unsupervised Anomaly Detection}, author = {Bieder, Florentin and Wolleb, Julia and Sandk\"uhler, Robin and Cattin, Philippe C.}, booktitle = {Proceedings of The 5th International Conference on Medical Imaging with Deep Learning}, pages = {160--172}, year = {2022}, editor = {Konukoglu, Ender and Menze, Bjoern and Venkataraman, Archana and Baumgartner, Christian and Dou, Qi and Albarqouni, Shadi}, volume = {172}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v172/bieder22a/bieder22a.pdf}, url = {https://proceedings.mlr.press/v172/bieder22a.html}, abstract = {In recent years, anomaly detection has become an essential field in medical image analysis. Most current anomaly detection methods for medical images are based on image reconstruction. In this work, we propose a novel anomaly detection approach based on coordinate regression. Our method estimates the position of patches within a volume, and is trained only on data of healthy subjects. During inference, we can detect and localize anomalies by considering the error of the position estimate of a given patch. We apply our method to 3D CT volumes and evaluate it on patients with intracranial haemorrhages and cranial fractures. The results show that our method performs well in detecting these anomalies. Furthermore, we show that our method requires less memory than comparable approaches that involve image reconstruction. This is highly relevant for processing large 3D volumes, for instance, CT or MRI scans.} }
Endnote
%0 Conference Paper %T Position Regression for Unsupervised Anomaly Detection %A Florentin Bieder %A Julia Wolleb %A Robin Sandkühler %A Philippe C. Cattin %B Proceedings of The 5th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2022 %E Ender Konukoglu %E Bjoern Menze %E Archana Venkataraman %E Christian Baumgartner %E Qi Dou %E Shadi Albarqouni %F pmlr-v172-bieder22a %I PMLR %P 160--172 %U https://proceedings.mlr.press/v172/bieder22a.html %V 172 %X In recent years, anomaly detection has become an essential field in medical image analysis. Most current anomaly detection methods for medical images are based on image reconstruction. In this work, we propose a novel anomaly detection approach based on coordinate regression. Our method estimates the position of patches within a volume, and is trained only on data of healthy subjects. During inference, we can detect and localize anomalies by considering the error of the position estimate of a given patch. We apply our method to 3D CT volumes and evaluate it on patients with intracranial haemorrhages and cranial fractures. The results show that our method performs well in detecting these anomalies. Furthermore, we show that our method requires less memory than comparable approaches that involve image reconstruction. This is highly relevant for processing large 3D volumes, for instance, CT or MRI scans.
APA
Bieder, F., Wolleb, J., Sandkühler, R. & Cattin, P.C.. (2022). Position Regression for Unsupervised Anomaly Detection. Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 172:160-172 Available from https://proceedings.mlr.press/v172/bieder22a.html.

Related Material