EUVIMEDEUVIMED entdecken
Crossref

Learning Robust Shape Regularization for Generalizable Medical Image Segmentation

Kecheng Chen, Tiexin Qin, Victor Ho-Fun Lee, Hong Yan, Haoliang Li

IEEE Transactions on Medical Imaging · 2024-07 · 43 (7) · 2693-2706

DOI: 10.1109/tmi.2024.3371987 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2024-07. Verfasst von Kecheng Chen und 4 weiteren Autor:innen. Die Arbeit wurde bisher 28-mal zitiert (Crossref) und nennt 50 Literaturangaben. Gefördert von: Hong Kong Innovation and Technology Commission, Hong Kong Research Grants Council, CityU New Research Initiatives/Infrastructure Support from Central.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Generalizable medical image segmentation enables models to generalize to unseen target domains under domain shift issues. Recent progress demonstrates that the shape of the segmentation objective, with its high consistency and robustness across domains, can serve as a reliable regularization to aid the model for better cross-domain performance, where existing methods typically seek a shared framework to render segmentation maps and shape prior concurrently. However, due to the inherent texture and style preference of modern deep neural networks, the edge or silhouette of the extracted shape will inevitably be undermined by those domain-specific texture and style interferences of medical images under domain shifts. To address this limitation, we devise a novel framework with a separation between the shape regularization and the segmentation map. Specifically, we first customize a novel whitening transform-based probabilistic shape regularization extractor namely WT-PSE to suppress undesirable domain-specific texture and style interferences, leading to more robust and high-quality shape representations. Second, we deliver a Wasserstein distance-guided knowledge distillation scheme to help the WT-PSE to achieve more flexible shape extraction during the inference phase. Finally, by incorporating domain knowledge of medical images, we propose a novel instance-domain whitening transform method to facilitate a more stable training process with improved performance. Experiments demonstrate the performance of our proposed method on both multi-domain and single-domain generalization.

Förderung

  • Hong Kong Innovation and Technology Commission
  • Hong Kong Research Grants Council
  • CityU New Research Initiatives/Infrastructure Support from Central
  • Seed Fund for Collaborative Research, The University of Hong Kong

Zitieren

APA

Chen, K., Qin, T., Lee, V. H., Yan, H., & Li, H. (2024). Learning Robust Shape Regularization for Generalizable Medical Image Segmentation. IEEE Transactions on Medical Imaging, 43(7), 2693-2706. https://doi.org/10.1109/tmi.2024.3371987

Vancouver

Chen K, Qin T, Lee VH, Yan H, Li H. Learning Robust Shape Regularization for Generalizable Medical Image Segmentation. IEEE Transactions on Medical Imaging. 2024;43(7):2693-2706. doi:10.1109/tmi.2024.3371987

BibTeX
@article{Chen2024,
  author = {Chen, Kecheng and Qin, Tiexin and Lee, Victor Ho-Fun and Yan, Hong and Li, Haoliang},
  title = {Learning Robust Shape Regularization for Generalizable Medical Image Segmentation},
  journal = {IEEE Transactions on Medical Imaging},
  year = {2024},
  volume = {43},
  number = {7},
  pages = {2693--2706},
  doi = {10.1109/tmi.2024.3371987}
}
Kostenlos starten →