On the Pitfalls of Using the Residual Error as Anomaly Score

Felix Meissen, Benedikt Wiestler, Georgios Kaissis, Daniel Rueckert
Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, PMLR 172:914-928, 2022.

Abstract

Many current state-of-the-art methods for anomaly localization in medical images rely on calculating a residual image between a potentially anomalous input image and its ("healthy") reconstruction. As the reconstruction of the unseen anomalous region should be erroneous, this yields large residuals as a score to detect anomalies in medical images. However, this assumption does not take into account residuals resulting from imperfect reconstructions of the machine learning models used. Such errors can easily overshadow residuals of interest and therefore strongly question the use of residual images as scoring function. Our work explores this fundamental problem of residual images in detail. We theoretically define the problem and thoroughly evaluate the influence of intensity and texture of anomalies against the effect of imperfect reconstructions in a series of experiments.

Cite this Paper

BibTeX
@InProceedings{pmlr-v172-meissen22a, title = {On the Pitfalls of Using the Residual Error as Anomaly Score}, author = {Meissen, Felix and Wiestler, Benedikt and Kaissis, Georgios and Rueckert, Daniel}, booktitle = {Proceedings of The 5th International Conference on Medical Imaging with Deep Learning}, pages = {914--928}, 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/meissen22a/meissen22a.pdf}, url = {https://proceedings.mlr.press/v172/meissen22a.html}, abstract = {Many current state-of-the-art methods for anomaly localization in medical images rely on calculating a residual image between a potentially anomalous input image and its ("healthy") reconstruction. As the reconstruction of the unseen anomalous region should be erroneous, this yields large residuals as a score to detect anomalies in medical images. However, this assumption does not take into account residuals resulting from imperfect reconstructions of the machine learning models used. Such errors can easily overshadow residuals of interest and therefore strongly question the use of residual images as scoring function. Our work explores this fundamental problem of residual images in detail. We theoretically define the problem and thoroughly evaluate the influence of intensity and texture of anomalies against the effect of imperfect reconstructions in a series of experiments.} }
Endnote
%0 Conference Paper %T On the Pitfalls of Using the Residual Error as Anomaly Score %A Felix Meissen %A Benedikt Wiestler %A Georgios Kaissis %A Daniel Rueckert %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-meissen22a %I PMLR %P 914--928 %U https://proceedings.mlr.press/v172/meissen22a.html %V 172 %X Many current state-of-the-art methods for anomaly localization in medical images rely on calculating a residual image between a potentially anomalous input image and its ("healthy") reconstruction. As the reconstruction of the unseen anomalous region should be erroneous, this yields large residuals as a score to detect anomalies in medical images. However, this assumption does not take into account residuals resulting from imperfect reconstructions of the machine learning models used. Such errors can easily overshadow residuals of interest and therefore strongly question the use of residual images as scoring function. Our work explores this fundamental problem of residual images in detail. We theoretically define the problem and thoroughly evaluate the influence of intensity and texture of anomalies against the effect of imperfect reconstructions in a series of experiments.
APA
Meissen, F., Wiestler, B., Kaissis, G. & Rueckert, D.. (2022). On the Pitfalls of Using the Residual Error as Anomaly Score. Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 172:914-928 Available from https://proceedings.mlr.press/v172/meissen22a.html.

Related Material