Semantic Segmentation of Malignant Tumors in Breast Ultrasound Images Based on Lightweight U-Net

Yusen Li

doi:10.47297/wspbdsWSP2752-630526.20260601

PDF
Export
Share
Collection
Album

Semantic Segmentation of Malignant Tumors in Breast Ultrasound Images Based on Lightweight U-Net

Volume 6, Issue 1

Views: 0 Downloads: 0

Semantic Segmentation of Malignant Tumors in Breast Ultrasound Images Based on Lightweight U-Net
Semantic Segmentation of Malignant Tumors in Breast Ultrasound Images Based on Lightweight U-Net
Tianjin University of Traditional Chinese Medicine, Medical Information Engineering, Tianjin, 301617, China
Vol. 6, Issue 1, Pages: 116-121(2026)
Published：30 March 2026
DOI：10.47297/wspbdsWSP2752-630526.20260601
Accepted：

Scan QR Code

Li Yusen.Semantic Segmentation of Malignant Tumors in Breast Ultrasound Images Based on Lightweight U-Net[J].Biomedical Data Science,2026,06(01):116-121.
DOI：

Li Yusen.Semantic Segmentation of Malignant Tumors in Breast Ultrasound Images Based on Lightweight U-Net[J].Biomedical Data Science,2026,06(01):116-121. DOI： 10.47297/wspbdsWSP2752-630526.20260601.

摘要

Abstract

To address the challenges of high noise

blurred edges in breast ultrasound images

and the large parameter size and slow inference speed of traditional U-Net models

which fail to meet the demands of real-time clinical auxiliary diagnosis

a lightweight U-Net-based semantic segmentation method for malignant tumors in breast ultrasound images with multi-scale supervision is proposed. This algorithm adopts the classical U-Net architecture as its foundation

introduces multi-scale supervision strategies and channel attention mechanisms to enhance tumor feature extraction capabilities

while simultaneously employing structured pruning and quantized attention training (QAT) to achieve extreme model lightweighting. Experimental results demonstrate that the lightweight model maintains segmentation accuracy (Dice coefficient reaching 0.818) while reducing parameter size by approximately 76.8% and improving inference speed by about 77.7%. This approach provides efficient technical support for intelligent auxiliary diagnosis of malignant breast tumors

making it suitable for low-power

real-time diagnostic scenarios in primary healthcare institutions.

关键词

Keywords

references

Chen Maoshan , Lü Qing . Key points interpretation of "Analysis of global incidence and mortality of breast cancer based on population registration data from 2000 to 2020 " [J ] . China Journal of Cardiothoracic Surgery , 2022 , 29 ( 04 ): 401 - 406 .

Zeng H , Zheng R , Guo Y , et al . Cancer survival in China, 2003 .

Chen Wanqing , Zheng Rongshou . Incidence, mortality and survival status of female breast cancer in China [J ] . China Oncology Clinical , 2015 , 42 ( 13 ): 668 - 674 .

Bao Mingwen , Xu Xiaohong . Current status and application of new technologies in ultrasound diagnosis of breast cancer [J ] . Journal of Guangdong Medical University , 2009 , 27 ( 02 ): 204 - 205 .

He Xiaodong , Gu Suying . Current Status and Advances in Ultrasound Imaging for Breast Cancer [J ] . Chinese Journal of General Practice , 2010 , 8 ( 02 ): 231 - 233 .

Muhammad M , Zeebaree D , Brifcani AM , Muhammad M , Zeebaree D , Brifcani AM , et al . Region of interest segmentation based on clustering techniques for breast cancer ultrasound images:a review [J ] . J Appl Sci Technol Trends , 2020 , 1 ( 2 ): 78 - 91 .

Zhao Kefu , Zhang Lei . Ultrasound image segmentation of thyroid nodules based on convolutional neural networks [J ] . Modern Computer , 2021 ( 15 ): 54 - 60 .

Shi Jun , Wang Tiantong , Zhu Ziqi , et al . A review of medical image segmentation methods based on deep learning [J ] . China Journal of Image and Graphics , 2025 , 30 ( 06 ): 2161 - 2186 .

Xu Guangxian , Feng Chun , Ma Fei . A Review of Medical Image Segmentation Based on U-Net [J ] . Computer Science and Exploration , 2023 , 17 ( 08 ): 1776 - 1792 .

Bi Pengcheng , Luo Jianxin , Chen Weiwei . Research on Lightweight Convolutional Neural Network Technology [J ] . Computer Engineering and Applications , 2019 , 55 ( 16 ): 25 - 35 .

Lin Jingdong , Wu Xinyi , Chai Yi , Yin Hongpeng . Review on Structural Optimization of Convolutional Neural Networks [J ] . Journal of Automation Chemistry , 2020 , 46 ( 01 ): 24 - 37 .

Yang Chun , Zhang Ruiyao , Huang Long , Ti Shutong , Lin Jinhui , Dong Zhiwei , Chen Songlu , Liu Yan , Yin Xucheng . A Review of Quantification Methods for Deep Neural Network Models [J ] . Acta Engineering Scientiae , 2023 , 45 ( 10 ): 1613 - 1629 .

Zhou Xiaosong , Zhao Juanjuan . Multi-scale supervised U-Net for ultrasound image segmentation of thyroid nodules [J ] . Journal of Taiyuan University of Technology , 2022 , 53 ( 06 ): 1134 - 1142 .

Jiang Xiaoyong , Li Zhongyi , Huang Langyue , et al . Review on Neural Network Pruning Techniques [J ] . Acta Scientia Applicatae Sinica , 2022 , 40 ( 05 ): 838 - 849 .

Alert me when the article has been cited

Submit

AI-powered Verification System for Medicinal and Edible Homologous Substances: Evidence-based Identification, Data Standardization and Web-based Practice

EAWA-UNet: Edge-Aware and Wavelet-Aggregated Network for Skin Lesion Segmentation with Blurry Boundaries

Interpretable GAN-based Breast Ultrasound Mass Augmentation for Improved Deep Learning BI-RADS Classification

Applications and Prospects of General Large Language Models in Medical Imaging

Applications of Deep Learning in Medical Imaging: A Systematic Review

Related Author

Ruiying Li Southwest University

Shuo Tian Southwest University

Yuxin Zhang Southwest University

Shenghua Li Southwest University

Jingjing Hua Southwest University

Guanxun Cui School of Computer Science and Technology, Chongqing University of Technology, Banan District

Zekan Li School of Computer Science and Technology, Chongqing University of Technology, Banan District

Jie Zhu Henan Medical University, School of Health Management;Henan Medical University, Xinxiang High Performance Computing Medical Engineering Technology Research Center

Related Institution

Southwest University

School of Computer Science and Technology, Chongqing University of Technology, Banan District

Henan Medical University, School of Health Management

Henan Medical University, Xinxiang High Performance Computing Medical Engineering Technology Research Center

School of Stomatology, Ningxia Medical University, Yinchuan, Ningxia, Hui Autonomous Region

⁰