Cover Image

Test-time augmentation with synthetic data addresses distribution shifts in spectral imaging

Purpose: Surgical scene segmentation is crucial for providing context-aware surgical assistance. Recent studies highlight the significant advantages of hyperspectral imaging (HSI) over traditional RGB data in enhancing segmentation performance. Nevertheless, the current hyperspectral imaging (HSI) d...

Full description

Saved in:
Bibliographic Details
Main Authors: Qasim, Ahmad Bin (Author) , Motta, Alessandro (Author) , Studier-Fischer, Alexander (Author) , Sellner, Jan (Author) , Ayala, Leonardo (Author) , Hübner, Marco (Author) , Bressan, Marc (Author) , Özdemir, Berkin (Author) , Kowalewski, Karl-Friedrich (Author) , Nickel, Felix (Author) , Seidlitz, Silvia (Author) , Maier-Hein, Lena (Author)
Format: Article (Journal)
Language:English
Published: 14 March 2024
In: International journal of computer assisted radiology and surgery
Year: 2024, Volume: 19, Issue: 6, Pages: 1021-1031
ISSN:1861-6429
DOI:10.1007/s11548-024-03085-3
Online Access:Verlag, kostenfrei, Volltext: https://doi.org/10.1007/s11548-024-03085-3
Get full text
Author Notes:Ahmad Bin Qasim, Alessandro Motta, Alexander Studier-Fischer, Jan Sellner, Leonardo Ayala, Marco Hübner, Marc Bressan, Berkin Özdemir, Karl Friedrich Kowalewski, Felix Nickel, Silvia Seidlitz, Lena Maier-Hein
Description
Summary:Purpose: Surgical scene segmentation is crucial for providing context-aware surgical assistance. Recent studies highlight the significant advantages of hyperspectral imaging (HSI) over traditional RGB data in enhancing segmentation performance. Nevertheless, the current hyperspectral imaging (HSI) datasets remain limited and do not capture the full range of tissue variations encountered clinically. Methods: Based on a total of 615 hyperspectral images from a total of 16 pigs, featuring porcine organs in different perfusion states, we carry out an exploration of distribution shifts in spectral imaging caused by perfusion alterations. We further introduce a novel strategy to mitigate such distribution shifts, utilizing synthetic data for test-time augmentation. Results: The effect of perfusion changes on state-of-the-art (SOA) segmentation networks depended on the organ and the specific perfusion alteration induced. In the case of the kidney, we observed a performance decline of up to 93% when applying a state-of-the-art (SOA) network under ischemic conditions. Our method improved on the state-of-the-art (SOA) by up to 4.6 times. Conclusion: Given its potential wide-ranging relevance to diverse pathologies, our approach may serve as a pivotal tool to enhance neural network generalization within the realm of spectral imaging.
Item Description:Gesehen am 29.07.2024
Physical Description:Online Resource
ISSN:1861-6429
DOI:10.1007/s11548-024-03085-3

Similar Items