Suppressing HIFU interference in ultrasound images using 1D U-Net-based neural networks

作者全名："Yang, Kun; Li, Qiang; Liu, Hengxin; Zeng, Qingxuan; Cai, Dejia; Xu, Jiahong; Zhou, Yingying; Tsui, Po-Hsiang; Zhou, Xiaowei"

作者地址："[Yang, Kun; Li, Qiang; Liu, Hengxin; Zeng, Qingxuan] Tianjin Univ, Sch Microelect, Tianjin, Peoples R China; [Cai, Dejia; Xu, Jiahong; Zhou, Yingying; Zhou, Xiaowei] Chongqing Med Univ, Coll Biomed Engn, State Key Lab Ultrasound Engn Med, Chongqing, Peoples R China; [Tsui, Po-Hsiang] Chang Gung Univ, Coll Med, Dept Med Imaging & Radiol Sci, Taoyuan, Taiwan; [Tsui, Po-Hsiang] Chang Gung Mem Hosp Linkou, Dept Pediat, Div Pediat Gastroenterol, Taoyuan, Taiwan; [Tsui, Po-Hsiang] Chang Gung Univ, Res Ctr Radiat Med, Taoyuan, Taiwan"

通信作者："Zhou, XW (通讯作者)，Chongqing Med Univ, Coll Biomed Engn, State Key Lab Ultrasound Engn Med, Chongqing, Peoples R China.; Tsui, PH (通讯作者)，Chang Gung Univ, Coll Med, Dept Med Imaging & Radiol Sci, Taoyuan, Taiwan.; Tsui, PH (通讯作者)，Chang Gung Mem Hosp Linkou, Dept Pediat, Div Pediat Gastroenterol, Taoyuan, Taiwan.; Tsui, PH (通讯作者)，Chang Gung Univ, Res Ctr Radiat Med, Taoyuan, Taiwan."

来源：PHYSICS IN MEDICINE AND BIOLOGY

ESI学科分类：MOLECULAR BIOLOGY & GENETICS

WOS号：WOS:001184858700001

JCR分区：Q2

影响因子：3.5

年份：2024

卷号：69

期号：7

开始页：　

结束页：　

文献类型：Article

关键词：high-intensity focused ultrasound (HIFU) interference; 1D U-Net; HIFU therapy; US-guide HIFU; deep learning

摘要："Objective. One big challenge with high-intensity focused ultrasound (HIFU) is that the intense acoustic interference generated by HIFU irradiation overwhelms the B-mode monitoring images, compromising monitoring effectiveness. This study aims to overcome this problem using a one-dimensional (1D) deep convolutional neural network. Approach. U-Net-based networks have been proven to be effective in image reconstruction and denoising, and the two-dimensional (2D) U-Net has already been investigated for suppressing HIFU interference in ultrasound monitoring images. In this study, we propose that the one-dimensional (1D) convolution in U-Net-based networks is more suitable for removing HIFU artifacts and can better recover the contaminated B-mode images compared to 2D convolution. Ex vivo and in vivo HIFU experiments were performed on a clinically equivalent ultrasound-guided HIFU platform to collect image data, and the 1D convolution in U-Net, Attention U-Net, U-Net++, and FUS-Net was applied to verify our proposal. Main results. All 1D U-Net-based networks were more effective in suppressing HIFU interference than their 2D counterparts, with over 30% improvement in terms of structural similarity (SSIM) to the uncontaminated B-mode images. Additionally, 1D U-Nets trained using ex vivo datasets demonstrated better generalization performance in in vivo experiments. Significance. These findings indicate that the utilization of 1D convolution in U-Net-based networks offers great potential in addressing the challenges of monitoring in ultrasound-guided HIFU systems."

基金机构："Natural Science Foundation of Tianjin Municipality https://doi.org/10.13039/501100006606 [2021YJSB118]; Postgraduate Research Innovation Project of Tianjin, China [22JCZDJC00220]; Tianjin Natural Science Foundation [2022KFKT004]; Foundation of State Key Laboratory of Ultrasound in Medicine and Engineering [62071323]; National Natural Science Foundation of China [CMRPD1N0011, CMRPD1M0401]; Chang Gung Memorial Hospital, Taiwan"

基金资助正文："This work was supported by the Postgraduate Research Innovation Project of Tianjin, China (Grant No. 2021YJSB118), the Tianjin Natural Science Foundation (Grant No. 22JCZDJC00220), the Foundation of State Key Laboratory of Ultrasound in Medicine and Engineering (Grant No. 2022KFKT004) and the National Natural Science Foundation of China (Grant No. 62071323). This work was supported in part by Chang Gung Memorial Hospital, Taiwan under Grant Nos. CMRPD1N0011 and CMRPD1M0401."