Cross-Image Dependency Modeling for Breast Ultrasound Segmentation
IEEE Transactions on Medical Imaging · 2023-06 · 42 (6) · 1619-1631
DOI: 10.1109/tmi.2022.3233648 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2023-06. Verfasst von Huisi Wu und 4 weiteren Autor:innen. Hauptthemen laut MEDLINE: Artifacts, Ultrasonography, Mammary. Die Arbeit wurde bisher 48-mal zitiert (Crossref) und nennt 49 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Guangdong Province, China, COVID-19 Prevention Project of Guangdong Province, China.
Abstract (Originalsprache)
We present a novel deep network (namely BUSSeg) equipped with both within- and cross-image long-range dependency modeling for automated lesions segmentation from breast ultrasound images, which is a quite daunting task due to (1) the large variation of breast lesions, (2) the ambiguous lesion boundaries, and (3) the existence of speckle noise and artifacts in ultrasound images. Our work is motivated by the fact that most existing methods only focus on modeling the within-image dependencies while neglecting the cross-image dependencies, which are essential for this task under limited training data and noise. We first propose a novel cross-image dependency module (CDM) with a cross-image contextual modeling scheme and a cross-image dependency loss (CDL) to capture more consistent feature expression and alleviate noise interference. Compared with existing cross-image methods, the proposed CDM has two merits. First, we utilize more complete spatial features instead of commonly used discrete pixel vectors to capture the semantic dependencies between images, mitigating the negative effects of speckle noise and making the acquired features more representative. Second, the proposed CDM includes both intra- and inter-class contextual modeling rather than just extracting homogeneous contextual dependencies. Furthermore, we develop a parallel bi-encoder architecture (PBA) to tame a Transformer and a convolutional neural network to enhance BUSSeg's capability in capturing within-image long-range dependencies and hence offer richer features for CDM. We conducted extensive experiments on two representative public breast ultrasound datasets, and the results demonstrate that the proposed BUSSeg consistently outperforms state-of-the-art approaches in most metrics.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Guangdong Province, China
- COVID-19 Prevention Project of Guangdong Province, China
- Innovation and Technology Fund under Guangdong-Hong Kong Technology Cooperation Funding Scheme
- Hong Kong Research Grant Council under General Research Fund Scheme
Zitieren
APA
Wu, H., Huang, X., Guo, X., Wen, Z., & Qin, J. (2023). Cross-Image Dependency Modeling for Breast Ultrasound Segmentation. IEEE Transactions on Medical Imaging, 42(6), 1619-1631. https://doi.org/10.1109/tmi.2022.3233648
Vancouver
Wu H, Huang X, Guo X, Wen Z, Qin J. Cross-Image Dependency Modeling for Breast Ultrasound Segmentation. IEEE Transactions on Medical Imaging. 2023;42(6):1619-1631. doi:10.1109/tmi.2022.3233648
BibTeX
@article{Wu2023,
author = {Wu, Huisi and Huang, Xiaoting and Guo, Xinrong and Wen, Zhenkun and Qin, Jing},
title = {Cross-Image Dependency Modeling for Breast Ultrasound Segmentation},
journal = {IEEE Transactions on Medical Imaging},
year = {2023},
volume = {42},
number = {6},
pages = {1619--1631},
doi = {10.1109/tmi.2022.3233648}
}
EUVIMED