Improving Error Detection in Deep Learning Based Radiotherapy Autocontouring Using Bayesian Uncertainty

Prerak Mody*, Nicolas F. Chaves-de-Plaza, Klaus Hildebrandt, Marius Staring

*Corresponding author for this work

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

Abstract

Bayesian Neural Nets (BNN) are increasingly used for robust organ auto-contouring. Uncertainty heatmaps extracted from BNNs have been shown to correspond to inaccurate regions. To help speed up the mandatory quality assessment (QA) of contours in radiotherapy, these heatmaps could be used as stimuli to direct visual attention of clinicians to potential inaccuracies. In practice, this is non-trivial to achieve since many accurate regions also exhibit uncertainty. To influence the output uncertainty of a BNN, we propose a modified accuracy-versus-uncertainty (AvU) metric as an additional objective during model training that penalizes both accurate regions exhibiting uncertainty as well as inaccurate regions exhibiting certainty. For evaluation, we use an uncertainty-ROC curve that can help differentiate between Bayesian models by comparing the probability of uncertainty in inaccurate versus accurate regions. We train and evaluate a FlipOut BNN model on the MICCAI2015 Head and Neck Segmentation challenge dataset and on the DeepMind-TCIA dataset, and observed an increase in the AUC of uncertainty-ROC curves by 5.6% and 5.9%, respectively, when using the AvU objective. The AvU objective primarily reduced false positives regions (uncertain and accurate), drawing less visual attention to these regions, thereby potentially improving the speed of error detection.

Original languageEnglish
Title of host publicationUncertainty for Safe Utilization of Machine Learning in Medical Imaging - 4th International Workshop, UNSURE 2022, Held in Conjunction with MICCAI 2022, Proceedings
EditorsCarole H. Sudre, Carole H. Sudre, Christian F. Baumgartner, Adrian Dalca, Adrian Dalca, William M. Wells III, Chen Qin, Ryutaro Tanno, Koen Van Leemput, Koen Van Leemput, William M. Wells III
PublisherSpringer
Pages70-79
Number of pages10
ISBN (Print)9783031167485
DOIs
Publication statusPublished - 2022
Event4th Workshop on Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, UNSURE 2022, held in conjunction with 25th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2022 - Singapore, Singapore
Duration: 18 Sep 202218 Sep 2022

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13563 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference4th Workshop on Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, UNSURE 2022, held in conjunction with 25th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2022
Country/TerritorySingapore
CitySingapore
Period18/09/2218/09/22

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

  • AvU loss
  • Bayesian uncertainty
  • Deep learning
  • FlipOut
  • Organs-at-Risk
  • Quality assessment
  • Radiotherapy
  • Uncertainty-ROC

Fingerprint

Dive into the research topics of 'Improving Error Detection in Deep Learning Based Radiotherapy Autocontouring Using Bayesian Uncertainty'. Together they form a unique fingerprint.

Cite this