@inproceedings{6d597720894a4a6596370720e3ed495c,
title = "UA-MAE: An Uncertainty-Aware Masked Autoencoder for Breast Lesion Segmentation in Ultrasound Images",
abstract = "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.",
keywords = "Breast ultrasound, Masked autoencoders, Segmentation, Self-supervised learning, Uncertainty map",
author = "Xiangyu Xiong and Yue Sun and Jiaju Huang and Da Huang and Shaobin Chen and Zhuoneng Zhang and Tao Tan",
note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 2025 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2025 ; Conference date: 15-12-2025 Through 18-12-2025",
year = "2025",
doi = "10.1109/BIBM66473.2025.11356317",
language = "English",
series = "Proceedings - 2025 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2025",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "4295--4298",
editor = "Juan Liu and Jingshan Huang and Xiaowo Wang and Fa Zhang and Xiufen Zou and Tian Tian and Xiaohua Hu and Bin Hu and Yi Xiong",
booktitle = "Proceedings - 2025 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2025",
address = "United States",
}