TY - GEN
T1 - 3D Deep Adaptively Efficient Attention Model to segment Automated Breast Ultrasound
AU - Jeong, Hyunsu
AU - Yoon, Chiho
AU - Lim, Hyunseok
AU - Won, Jongjun
AU - Luo, Gongning
AU - Xu, Mingwang
AU - Kim, Kiduk
AU - Kim, Namkug
AU - Kim, Chulhong
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - 3D automated breast ultrasound (ABUS) data includes a huge number of slices with lesion-related artifacts and size variants, causing false positive cases. In addition, applying transformer-based models with self-attention considering long-range dependency to 3D ABUS data is challenging because of the small number of data. We propose 3D deep adaptively efficient attention model with a multi-task learning to segment the small number of 3D ABUS datasets. Our network consists of adaptive hybrid-transformer blocks for computationally efficient self-attention and multi-task learning to minimize false positive cases. Additionally, a novel attention gate is also implanted to considering long-range dependency between channel-attentive features. Our model performed 58.27%, 23.59 mm, and 0.054% on Dice, HD95, and FPR, respectively.
AB - 3D automated breast ultrasound (ABUS) data includes a huge number of slices with lesion-related artifacts and size variants, causing false positive cases. In addition, applying transformer-based models with self-attention considering long-range dependency to 3D ABUS data is challenging because of the small number of data. We propose 3D deep adaptively efficient attention model with a multi-task learning to segment the small number of 3D ABUS datasets. Our network consists of adaptive hybrid-transformer blocks for computationally efficient self-attention and multi-task learning to minimize false positive cases. Additionally, a novel attention gate is also implanted to considering long-range dependency between channel-attentive features. Our model performed 58.27%, 23.59 mm, and 0.054% on Dice, HD95, and FPR, respectively.
KW - 3D automated breast ultrasound (ABUS) segmentation
KW - Channel-attentive model
KW - Hybrid transformer
KW - Multi-task learning
UR - https://www.scopus.com/pages/publications/85216492821
U2 - 10.1109/UFFC-JS60046.2024.10793736
DO - 10.1109/UFFC-JS60046.2024.10793736
M3 - 会议稿件
AN - SCOPUS:85216492821
T3 - IEEE Ultrasonics, Ferroelectrics, and Frequency Control Joint Symposium, UFFC-JS 2024 - Proceedings
BT - IEEE Ultrasonics, Ferroelectrics, and Frequency Control Joint Symposium, UFFC-JS 2024 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2024 IEEE Ultrasonics, Ferroelectrics, and Frequency Control Joint Symposium, UFFC-JS 2024
Y2 - 22 September 2024 through 26 September 2024
ER -