EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

RPFeaNet: Rethinking Deep Progressive Prompt-Guided Feature Interaction Fusion Network for Medical Ultrasound Image Segmentation

Lei Zhu iD, Yuqing Du iD

Sensors · 14.04.2026 · 26 (8) · 2394

DOI: 10.3390/s26082394 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 14.04.2026. Verfasst von Lei Zhu und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechanics and Engineering Science, Peking University, Beijing 100871, China. Hauptthemen laut MEDLINE: Ultrasonography, Deep Learning. Die Arbeit nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Although ultrasound image segmentation has advanced significantly with deep learning, existing methods still suffer from a lack of prior knowledge guidance, partly due to the low-contrast, speckle-noise-corrupted nature from clinical ultrasound sensors. This paper proposes a novel ultrasound segmentation framework (RPFeaNet) that extracts progressive prompts from a low-to-high level prompt generation mechanism. Furthermore, the high-level prompt-guided feature interaction module (HPGFIM) fuses progressive prompt via Mamba blocks and stage-wise condition injection. The dynamic selective-frequency decoder (DSFD) combines dynamically selecting a strategy with the fusion of high-frequency details to suppress noise and refine edge details. Extensive experiments on six datasets demonstrate that RPFeaNet achieves state-of-the-art performance compared to existing methods, validating its strong generalization and robustness across diverse clinical ultrasound scenarios.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

UltrasonographyDeep LearningImage Processing, Computer-AssistedDatasets as TopicHumans

Stichwörter der Autor:innen

Mambafrequency selectionprogressive promptultrasound image segmentation

Autor:innen und Institutionen

  • School of Mechanics and Engineering Science, Peking University, Beijing 100871, China.
  • Academy of Artificial Intelligence, Beijing Institute of Petrochemical Technology, Beijing 102617, China.

ORCID-Profile: Lei Zhu, Yuqing Du

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Zhu, L., & Du, Y. (2026). RPFeaNet: Rethinking Deep Progressive Prompt-Guided Feature Interaction Fusion Network for Medical Ultrasound Image Segmentation. Sensors, 26(8), 2394. https://doi.org/10.3390/s26082394

Vancouver

Zhu L, Du Y. RPFeaNet: Rethinking Deep Progressive Prompt-Guided Feature Interaction Fusion Network for Medical Ultrasound Image Segmentation. Sensors. 2026;26(8):2394. doi:10.3390/s26082394

BibTeX
@article{Zhu2026,
  author = {Zhu, Lei and Du, Yuqing},
  title = {RPFeaNet: Rethinking Deep Progressive Prompt-Guided Feature Interaction Fusion Network for Medical Ultrasound Image Segmentation},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {8},
  pages = {2394},
  doi = {10.3390/s26082394}
}
Kostenlos starten →