Imperial College London
Alternate

ProfessorDanielRueckert

Faculty of Engineering, Department of Computing

Professor of Visual Information Processing
 
 
 
//

Contact

 

+44 (0)20 7594 8333d.rueckert Website

 
 
//

Location

 

568Huxley BuildingSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@inproceedings{Chai:2021:10.1007/978-3-030-80432-9_23,
author = {Chai, S and Rueckert, D and Fetit, A},
doi = {10.1007/978-3-030-80432-9_23},
pages = {294--304},
publisher = {Springer Verlag},
title = {Reducing textural bias improves robustness of deep segmentation models},
url = {http://dx.doi.org/10.1007/978-3-030-80432-9_23},
year = {2021}
}
Download

RIS format (EndNote, RefMan)

TY  - CPAPER
AB  - 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.
AU  - Chai,S
AU  - Rueckert,D
AU  - Fetit,A
DO  - 10.1007/978-3-030-80432-9_23
EP  - 304
PB  - Springer Verlag
PY  - 2021///
SN  - 0302-9743
SP  - 294
TI  - Reducing textural bias improves robustness of deep segmentation models
UR  - http://dx.doi.org/10.1007/978-3-030-80432-9_23
UR  - http://hdl.handle.net/10044/1/90728
ER  -
Download