跳至主導覽 跳至搜尋 跳過主要內容

UA-MAE: An Uncertainty-Aware Masked Autoencoder for Breast Lesion Segmentation in Ultrasound Images

  • Xiangyu Xiong
  • , Yue Sun
  • , Jiaju Huang
  • , Da Huang
  • , Shaobin Chen
  • , Zhuoneng Zhang
  • , Tao Tan
  • Macao Polytechnic University

研究成果: Conference contribution同行評審

摘要

Accurate segmentation of breast lesions is vital for diagnosing breast diseases. Masked image modeling (MIM) with random masking performs well in self-supervised learning but struggles in breast ultrasound segmentation due to (1) ambiguous representations from similar intensities near lesion boundaries and (2) a bias toward irrelevant regions. We propose UA-MAE, an uncertainty-aware masked autoencoder that uses pixel-wise uncertainty maps to dynamically select masking patches, prioritizing boundaries and morphologically relevant lesion areas. Experiments on two public datasets for pre-training and three for fine-tuning show UA-MAE outperforming four state-of-theart SSL methods and two supervised approaches in segmentation accuracy across diverse breast ultrasound images. The code is available at https://github.com/yXiangXiong/UA-MAE.

原文English
主出版物標題Proceedings - 2025 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2025
編輯Juan Liu, Jingshan Huang, Xiaowo Wang, Fa Zhang, Xiufen Zou, Tian Tian, Xiaohua Hu, Bin Hu, Yi Xiong
發行者Institute of Electrical and Electronics Engineers Inc.
頁面4295-4298
頁數4
ISBN(電子)9798331515577
DOIs
出版狀態Published - 2025
事件2025 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2025 - Wuhan, China
持續時間: 15 12月 202518 12月 2025

出版系列

名字Proceedings - 2025 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2025

Conference

Conference2025 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2025
國家/地區China
城市Wuhan
期間15/12/2518/12/25

指紋

深入研究「UA-MAE: An Uncertainty-Aware Masked Autoencoder for Breast Lesion Segmentation in Ultrasound Images」主題。共同形成了獨特的指紋。

引用此