A Heteroscedastic Uncertainty Model for Decoupling Sources of MRI Image Quality

Richard Shaw, Carole H. Sudre, Sébastien Ourselin, M. Jorge Cardoso
; Proceedings of the Third Conference on Medical Imaging with Deep Learning, PMLR 121:733-742, 2020.

Abstract

Quality control (QC) of medical images is essential to ensure that downstream analyses such as segmentation can be performed successfully. Currently, QC is predominantly performed visually at significant time and operator cost. We aim to automate the process by formulating a probabilistic network that estimates uncertainty through a heteroscedastic noise model, hence providing a proxy measure of task-specific image quality that is learnt directly from the data. By augmenting the training data with different types of simulated k-space artefacts, we propose a novel cascading CNN architecture based on a student-teacher framework to decouple sources of uncertainty related to different k-space augmentations in an entirely self-supervised manner. This enables us to predict separate uncertainty quantities for the different types of data degradation. While the uncertainty measures reflect the presence and severity of image artefacts, the network also provides the segmentation predictions given the quality of the data. We show models trained with simulated artefacts provide informative measures of uncertainty on real-world images and we validate our uncertainty predictions on problematic images identified by human-raters.

Cite this Paper


BibTeX
@InProceedings{pmlr-v121-shaw20a, title = {A Heteroscedastic Uncertainty Model for Decoupling Sources of MRI Image Quality}, author = {Shaw, Richard and Sudre, Carole H. and Ourselin, S\'{e}bastien and Cardoso, M. Jorge}, pages = {733--742}, year = {2020}, editor = {Tal Arbel and Ismail Ben Ayed and Marleen de Bruijne and Maxime Descoteaux and Herve Lombaert and Christopher Pal}, volume = {121}, series = {Proceedings of Machine Learning Research}, address = {Montreal, QC, Canada}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v121/shaw20a/shaw20a.pdf}, url = {http://proceedings.mlr.press/v121/shaw20a.html}, abstract = {Quality control (QC) of medical images is essential to ensure that downstream analyses such as segmentation can be performed successfully. Currently, QC is predominantly performed visually at significant time and operator cost. We aim to automate the process by formulating a probabilistic network that estimates uncertainty through a heteroscedastic noise model, hence providing a proxy measure of task-specific image quality that is learnt directly from the data. By augmenting the training data with different types of simulated k-space artefacts, we propose a novel cascading CNN architecture based on a student-teacher framework to decouple sources of uncertainty related to different k-space augmentations in an entirely self-supervised manner. This enables us to predict separate uncertainty quantities for the different types of data degradation. While the uncertainty measures reflect the presence and severity of image artefacts, the network also provides the segmentation predictions given the quality of the data. We show models trained with simulated artefacts provide informative measures of uncertainty on real-world images and we validate our uncertainty predictions on problematic images identified by human-raters.} }
Endnote
%0 Conference Paper %T A Heteroscedastic Uncertainty Model for Decoupling Sources of MRI Image Quality %A Richard Shaw %A Carole H. Sudre %A Sébastien Ourselin %A M. Jorge Cardoso %B Proceedings of the Third Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2020 %E Tal Arbel %E Ismail Ben Ayed %E Marleen de Bruijne %E Maxime Descoteaux %E Herve Lombaert %E Christopher Pal %F pmlr-v121-shaw20a %I PMLR %J Proceedings of Machine Learning Research %P 733--742 %U http://proceedings.mlr.press %V 121 %W PMLR %X Quality control (QC) of medical images is essential to ensure that downstream analyses such as segmentation can be performed successfully. Currently, QC is predominantly performed visually at significant time and operator cost. We aim to automate the process by formulating a probabilistic network that estimates uncertainty through a heteroscedastic noise model, hence providing a proxy measure of task-specific image quality that is learnt directly from the data. By augmenting the training data with different types of simulated k-space artefacts, we propose a novel cascading CNN architecture based on a student-teacher framework to decouple sources of uncertainty related to different k-space augmentations in an entirely self-supervised manner. This enables us to predict separate uncertainty quantities for the different types of data degradation. While the uncertainty measures reflect the presence and severity of image artefacts, the network also provides the segmentation predictions given the quality of the data. We show models trained with simulated artefacts provide informative measures of uncertainty on real-world images and we validate our uncertainty predictions on problematic images identified by human-raters.
APA
Shaw, R., Sudre, C.H., Ourselin, S. & Cardoso, M.J.. (2020). A Heteroscedastic Uncertainty Model for Decoupling Sources of MRI Image Quality. Proceedings of the Third Conference on Medical Imaging with Deep Learning, in PMLR 121:733-742

Related Material