Robust Detection Outcome: A Metric for Pathology Detection in Medical Images

Felix Meissen, Philip Müller, Georgios Kaissis, Daniel Rueckert
Medical Imaging with Deep Learning, PMLR 227:568-585, 2024.

Abstract

Detection of pathologies is a fundamental task in medical imaging and the evaluation of algorithms that can perform this task automatically is crucial. However, current object detection metrics for natural images do not reflect the specific clinical requirements in pathology detection sufficiently. To tackle this problem, we propose Robust Detection Outcome (RoDeO); a novel metric for evaluating algorithms for pathology detection in medical images, especially in chest X-rays. RoDeO evaluates different errors directly and individually, and reflects clinical needs better than current metrics. Extensive evaluation on the ChestX-ray8 dataset shows the superiority of our metrics compared to existing ones. We released the code at [https://github.com/FeliMe/RoDeO](https://github.com/FeliMe/RoDeO) and published RoDeO as pip package ($rodeometric$).

Cite this Paper

BibTeX
@InProceedings{pmlr-v227-meissen24a, title = {Robust Detection Outcome: A Metric for Pathology Detection in Medical Images}, author = {Meissen, Felix and M\"uller, Philip and Kaissis, Georgios and Rueckert, Daniel}, booktitle = {Medical Imaging with Deep Learning}, pages = {568--585}, 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/meissen24a/meissen24a.pdf}, url = {https://proceedings.mlr.press/v227/meissen24a.html}, abstract = {Detection of pathologies is a fundamental task in medical imaging and the evaluation of algorithms that can perform this task automatically is crucial. However, current object detection metrics for natural images do not reflect the specific clinical requirements in pathology detection sufficiently. To tackle this problem, we propose Robust Detection Outcome (RoDeO); a novel metric for evaluating algorithms for pathology detection in medical images, especially in chest X-rays. RoDeO evaluates different errors directly and individually, and reflects clinical needs better than current metrics. Extensive evaluation on the ChestX-ray8 dataset shows the superiority of our metrics compared to existing ones. We released the code at [https://github.com/FeliMe/RoDeO](https://github.com/FeliMe/RoDeO) and published RoDeO as pip package ($rodeometric$).} }
Endnote
%0 Conference Paper %T Robust Detection Outcome: A Metric for Pathology Detection in Medical Images %A Felix Meissen %A Philip Müller %A Georgios Kaissis %A Daniel Rueckert %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-meissen24a %I PMLR %P 568--585 %U https://proceedings.mlr.press/v227/meissen24a.html %V 227 %X Detection of pathologies is a fundamental task in medical imaging and the evaluation of algorithms that can perform this task automatically is crucial. However, current object detection metrics for natural images do not reflect the specific clinical requirements in pathology detection sufficiently. To tackle this problem, we propose Robust Detection Outcome (RoDeO); a novel metric for evaluating algorithms for pathology detection in medical images, especially in chest X-rays. RoDeO evaluates different errors directly and individually, and reflects clinical needs better than current metrics. Extensive evaluation on the ChestX-ray8 dataset shows the superiority of our metrics compared to existing ones. We released the code at [https://github.com/FeliMe/RoDeO](https://github.com/FeliMe/RoDeO) and published RoDeO as pip package ($rodeometric$).
APA
Meissen, F., Müller, P., Kaissis, G. & Rueckert, D.. (2024). Robust Detection Outcome: A Metric for Pathology Detection in Medical Images. Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 227:568-585 Available from https://proceedings.mlr.press/v227/meissen24a.html.

Related Material