Efficiently correcting patch-based segmentation errors to control image-level performance in retinal images

Patrick Köhler, Jeremiah Fadugba, Philipp Berens, Lisa M. Koch
Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning, PMLR 250:841-856, 2024.

Abstract

Segmentation models which are deployed into clinical practice need to meet a quality standard for each image. Even when models perform well on average, they may fail at segmenting individual images with a sufficiently high quality. We propose a combined quality control and error correction framework to reach the desired segmentation quality in each image. Our framework recommends the necessary number of local patches for manual review and estimates the impact of the intervention on the Dice Score of the corrected segmentation. This allows to trade off segmentation quality against time invested into manual review. We select the patches based on uncertainty maps obtained from an ensemble of segmentation models. We evaluated our method on retinal vessel segmentation on fundus images, where the Dice Score increased substantially after reviewing only a few patches. Our method accurately estimated the review’s impact on the Dice Score and we found that our framework controls the quality standard efficiently, i.e. reviewing as little as necessary.

Cite this Paper

BibTeX
@InProceedings{pmlr-v250-kohler24a, title = {Efficiently correcting patch-based segmentation errors to control image-level performance in retinal images}, author = {K\"ohler, Patrick and Fadugba, Jeremiah and Berens, Philipp and Koch, Lisa M.}, booktitle = {Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning}, pages = {841--856}, year = {2024}, editor = {Burgos, Ninon and Petitjean, Caroline and Vakalopoulou, Maria and Christodoulidis, Stergios and Coupe, Pierrick and Delingette, Hervé and Lartizien, Carole and Mateus, Diana}, volume = {250}, series = {Proceedings of Machine Learning Research}, month = {03--05 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v250/main/assets/kohler24a/kohler24a.pdf}, url = {https://proceedings.mlr.press/v250/kohler24a.html}, abstract = {Segmentation models which are deployed into clinical practice need to meet a quality standard for each image. Even when models perform well on average, they may fail at segmenting individual images with a sufficiently high quality. We propose a combined quality control and error correction framework to reach the desired segmentation quality in each image. Our framework recommends the necessary number of local patches for manual review and estimates the impact of the intervention on the Dice Score of the corrected segmentation. This allows to trade off segmentation quality against time invested into manual review. We select the patches based on uncertainty maps obtained from an ensemble of segmentation models. We evaluated our method on retinal vessel segmentation on fundus images, where the Dice Score increased substantially after reviewing only a few patches. Our method accurately estimated the review’s impact on the Dice Score and we found that our framework controls the quality standard efficiently, i.e. reviewing as little as necessary.} }
Endnote
%0 Conference Paper %T Efficiently correcting patch-based segmentation errors to control image-level performance in retinal images %A Patrick Köhler %A Jeremiah Fadugba %A Philipp Berens %A Lisa M. Koch %B Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2024 %E Ninon Burgos %E Caroline Petitjean %E Maria Vakalopoulou %E Stergios Christodoulidis %E Pierrick Coupe %E Hervé Delingette %E Carole Lartizien %E Diana Mateus %F pmlr-v250-kohler24a %I PMLR %P 841--856 %U https://proceedings.mlr.press/v250/kohler24a.html %V 250 %X Segmentation models which are deployed into clinical practice need to meet a quality standard for each image. Even when models perform well on average, they may fail at segmenting individual images with a sufficiently high quality. We propose a combined quality control and error correction framework to reach the desired segmentation quality in each image. Our framework recommends the necessary number of local patches for manual review and estimates the impact of the intervention on the Dice Score of the corrected segmentation. This allows to trade off segmentation quality against time invested into manual review. We select the patches based on uncertainty maps obtained from an ensemble of segmentation models. We evaluated our method on retinal vessel segmentation on fundus images, where the Dice Score increased substantially after reviewing only a few patches. Our method accurately estimated the review’s impact on the Dice Score and we found that our framework controls the quality standard efficiently, i.e. reviewing as little as necessary.
APA
Köhler, P., Fadugba, J., Berens, P. & Koch, L.M.. (2024). Efficiently correcting patch-based segmentation errors to control image-level performance in retinal images. Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 250:841-856 Available from https://proceedings.mlr.press/v250/kohler24a.html.

Related Material