Alwadee, Ebtihal, Sun, Xianfang  ORCID: https://orcid.org/0000-0002-6114-0766, Qin, Yipeng ORCID: https://orcid.org/0000-0002-1551-9126 and Langbein, Frank ORCID: https://orcid.org/0000-0002-3379-0323
2024.
Assessing and enhancing the robustness of brain tumor segmentation using a probabilistic deep learning architecture [Abstract].
Proceedings of the 2024 ISMRM & ISMRT Annual Meeting
(4526)
, pp. 1-6.
10.58530/2024/4526
|
Preview |
PDF
- Presentation
Download (394kB) | Preview |
Abstract
Motivation: Motivated by the challenge of enhancing the robustness of deep neural network decisions against variable noise in MRI-based brain tumor segmentation. Goal(s): This study aims to evaluate the efficacy of probabilistic bottlenecks in enhancing segmentation robustness. Approach: Our approach simulates structured perturbations at increasing strength to assess their impact on segmentation performance utilizing the Wasserstein distance between per-sample Dice score distributions and the sensitivity with respect to the perturbation strength. Results: Results show probabilistic bottlenecks significantly increase robustness to Gaussian noise, yet offer limited improvement towards Gaussian blur, with varying results for other perturbations, highlighting the perturbation-specific nature of network resilience. Impact: This study provides a tool to assess and guard against various perturbations in deep learning.It specifically demonstrates that probabilistic bottlenecks boost robustness of performance withrespect to certain noise types, but not all.
| Item Type: | Short Communication |
|---|---|
| Date Type: | Published Online |
| Status: | Published |
| Schools: | Professional Services > Advanced Research Computing @ Cardiff (ARCCA) Schools > Computer Science & Informatics |
| Publisher: | International Society for Magnetic Resonance in Medicine |
| ISSN: | 1524-6965 |
| Related URLs: | |
| Date of First Compliant Deposit: | 17 May 2024 |
| Date of Acceptance: | 26 January 2024 |
| Last Modified: | 11 Feb 2026 14:50 |
| URI: | https://orca.cardiff.ac.uk/id/eprint/167780 |
Actions (repository staff only)
 |
Edit Item |
Altmetric
Altmetric