Reducing Textural Bias Improves Robustness of Deep Segmentation Models
Chai S, Rueckert D, Fetit AE (2021)
Publication Type: Conference contribution
Publication year: 2021
Journal
Publisher: Springer Science and Business Media Deutschland GmbH
Book Volume: 12722 LNCS
Pages Range: 294-304
Conference Proceedings Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Event location: Virtual, Online
ISBN: 9783030804312
DOI: 10.1007/978-3-030-80432-9_23
Abstract
Despite advances in deep learning, robustness under domain shift remains a major bottleneck in medical imaging settings. Findings on natural images suggest that deep neural models can show a strong textural bias when carrying out image classification tasks. In this thorough empirical study, we draw inspiration from findings on natural images and investigate ways in which addressing the textural bias phenomenon could bring up the robustness of deep segmentation models when applied to three-dimensional (3D) medical data. To achieve this, publicly available MRI scans from the Developing Human Connectome Project are used to study ways in which simulating textural noise can help train robust models in a complex semantic segmentation task. We contribute an extensive empirical investigation consisting of 176 experiments and illustrate how applying specific types of simulated textural noise prior to training can lead to texture invariant models, resulting in improved robustness when segmenting scans corrupted by previously unseen noise types and levels.
Involved external institutions
United Kingdom (GB)How to cite
APA:
Chai, S., Rueckert, D., & Fetit, A.E. (2021). Reducing Textural Bias Improves Robustness of Deep Segmentation Models. In Bartłomiej W. Papież, Mohammad Yaqub, Jianbo Jiao, Ana I. Namburete, J. Alison Noble (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 294-304). Virtual, Online: Springer Science and Business Media Deutschland GmbH.
MLA:
Chai, Seoin, Daniel Rueckert, and Ahmed E. Fetit. "Reducing Textural Bias Improves Robustness of Deep Segmentation Models." Proceedings of the 25th Annual Conference on Medical Image Understanding and Analysis, MIUA 2021, Virtual, Online Ed. Bartłomiej W. Papież, Mohammad Yaqub, Jianbo Jiao, Ana I. Namburete, J. Alison Noble, Springer Science and Business Media Deutschland GmbH, 2021. 294-304.
BibTeX: Download