EUVIMEDEUVIMED entdecken
Crossref

CPCLNet: Advancing Cross-Modal Few-Shot Medical Image Segmentation with Clustering-Inspired Prototype Consistency Learning

Peng Zhang, Chen Liu, Yida Wu, Heyang Zhao, Zeyu Liu, Guanglei Zhang, Hongming Shan, Wenjian Wang

IEEE Transactions on Medical Imaging · 2026 · 1-1

DOI: 10.1109/tmi.2026.3753036 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2026. Verfasst von Peng Zhang und 7 weiteren Autor:innen. Gefördert von: National Natural Science Foundation of China, Beijing Natural Science Foundation, Fundamental Research Program of Shanxi Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cross-modal few-shot medical image segmentation (FSMIS) provides a solution for alleviating annotation scarcity and improving generalization across imaging modalities. However, severe modality discrepancy can disrupt support-query matching, contaminate support-derived foreground and background prototypes, and produce noisy affinity maps, making target-modality segmentation unreliable. To address this problem, we propose CPCLNet, a clustering-inspired prototype consistency learning network for cross-modal FSMIS. CPCLNet formulates cross-modal segmentation as a prototype consistency learning problem and progressively improves feature-space consistency, prototype reliability, and affinity-map coherence. Specifically, a parameter-sharing feature encoder provides a common representation basis across modalities, and a cross-modal distribution alignment module performs episode-specific support-query alignment. A clustering-inspired representative prototype descriptor further suppresses unstable support responses and constructs compact, modality-robust prototypes. Finally, an RWKV-based affinity map prediction module propagates prototype-guided correspondence across long-range spatial contexts with linear complexity. Extensive experiments on three imaging modalities, demonstrate that CPCLNet outperforms existing methods, achieving an average Dice improvement of 3.34 percentage points over the second-best competitor across CT→MRI and CT→PET tasks under both 1-shot and 5-shot settings. Further integration with MedSAM yields consistent improvements, indicating CP-CLNet's complementary potential with foundation models.

Förderung

  • National Natural Science Foundation of China
  • Beijing Natural Science Foundation
  • Fundamental Research Program of Shanxi Province
  • Special Project for Patent Transformation in Shanxi Province
  • 2025 High-Quality Industrial Development Special Program for Scientific Intelligence “Baituan Baixiang”

Zitieren

APA

Zhang, P., Liu, C., Wu, Y., Zhao, H., Liu, Z., Zhang, G., Shan, H., & Wang, W. (2026). CPCLNet: Advancing Cross-Modal Few-Shot Medical Image Segmentation with Clustering-Inspired Prototype Consistency Learning. IEEE Transactions on Medical Imaging, 1-1. https://doi.org/10.1109/tmi.2026.3753036

Vancouver

Zhang P, Liu C, Wu Y, Zhao H, Liu Z, Zhang G, et al. CPCLNet: Advancing Cross-Modal Few-Shot Medical Image Segmentation with Clustering-Inspired Prototype Consistency Learning. IEEE Transactions on Medical Imaging. 2026:1-1. doi:10.1109/tmi.2026.3753036

BibTeX
@article{Zhang2026,
  author = {Zhang, Peng and Liu, Chen and Wu, Yida and Zhao, Heyang and Liu, Zeyu and Zhang, Guanglei and Shan, Hongming and Wang, Wenjian},
  title = {CPCLNet: Advancing Cross-Modal Few-Shot Medical Image Segmentation with Clustering-Inspired Prototype Consistency Learning},
  journal = {IEEE Transactions on Medical Imaging},
  year = {2026},
  pages = {1--1},
  doi = {10.1109/tmi.2026.3753036}
}
Kostenlos starten →