WMCAF-Net: A Wavelet-Enhanced Mamba and Cross-Scale Adaptive Fusion Network for Breast Ultrasound Image Segmentation
Electronics · 08.11.2026 · 15 (21) · 5182
DOI: 10.3390/electronics15215182 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 08.11.2026. Verfasst von Xicheng Fu und 5 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Hunan Provincial Natural Science Foundation of China, Hunan Provincial University Key Laboratory of Artificial Intelligence Application and System Security, the 14th Five-Year Plan Key Disciplines and Application-oriented Special Disciplines of Hunan Province.
Abstract (Originalsprache)
Breast ultrasound image segmentation remains challenging because of low contrast, speckle noise, and blurred lesion boundaries. Although Mamba can model long-range dependencies efficiently, directly processing noisy ultrasound features may introduce unreliable contextual responses. Moreover, conventional skip connections can transmit redundant background information and hinder accurate boundary recovery. To address these issues, we propose WMCAF-Net, a Wavelet-Enhanced Mamba and Cross-Scale Adaptive Fusion Network. The Wave-Mamba module decomposes features into low- and high-frequency components using wavelet transform. Low-frequency components are processed by a visual state-space module to capture stable global structures, while high-frequency components are enhanced by directional convolutions to strengthen boundary and texture representation. The Cross-Scale Adaptive Fusion Module improves encoder–decoder feature fusion through channel-response recalibration and adaptive spatial selection. On BUSI, WMCAF-Net achieves 75.47% IoU and 83.47% Dice, surpassing the strongest competing methods by 2.08 and 1.97 percentage points, respectively. Under five-fold cross-validation on Dataset B, it achieves 72.89% IoU and 82.27% Dice, surpassing the strongest competing methods by 1.53 and 2.57 percentage points, respectively. Furthermore, bidirectional cross-dataset evaluation was conducted to analyze the segmentation performance of WMCAF-Net when transferred across different ultrasound datasets.
Förderung
- the Hunan Provincial Natural Science Foundation of China
- Hunan Provincial University Key Laboratory of Artificial Intelligence Application and System Security
- the 14th Five-Year Plan Key Disciplines and Application-oriented Special Disciplines of Hunan Province
- the Science and Technology Innovation Program of Hunan Province
Zitieren
APA
Fu, X., Deng, H., Lei, J., Lei, Y., Xu, L., & Wang, Q. (2026). WMCAF-Net: A Wavelet-Enhanced Mamba and Cross-Scale Adaptive Fusion Network for Breast Ultrasound Image Segmentation. Electronics, 15(21), 5182. https://doi.org/10.3390/electronics15215182
Vancouver
Fu X, Deng H, Lei J, Lei Y, Xu L, Wang Q. WMCAF-Net: A Wavelet-Enhanced Mamba and Cross-Scale Adaptive Fusion Network for Breast Ultrasound Image Segmentation. Electronics. 2026;15(21):5182. doi:10.3390/electronics15215182
BibTeX
@article{Fu2026,
author = {Fu, Xicheng and Deng, Hongwei and Lei, Jingshi and Lei, Yi and Xu, Lei and Wang, Qiangfei},
title = {WMCAF-Net: A Wavelet-Enhanced Mamba and Cross-Scale Adaptive Fusion Network for Breast Ultrasound Image Segmentation},
journal = {Electronics},
year = {2026},
volume = {15},
number = {21},
pages = {5182},
doi = {10.3390/electronics15215182}
}
EUVIMED