计算机科学
快速傅里叶变换
成像体模
梁(结构)
算法
光学
反向传播
超声波传感器
材料科学
声学
物理
人工神经网络
人工智能
作者
Pavel B. Rosnitskiy,Tatiana D. Khokhlova,George R. Schade,Oleg A. Sapozhnikov,Vera A. Khokhlova
出处
期刊:IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control
[Institute of Electrical and Electronics Engineers]
日期:2024-01-17
卷期号:71 (3): 341-353
被引量:1
标识
DOI:10.1109/tuffc.2024.3355390
摘要
High-intensity focused ultrasound (HIFU) applications for thermal or mechanical ablation of renal tumors often encounter challenges due to significant beam aberration and refraction caused by oblique beam incidence, inhomogeneous tissue layers, and presence of gas and bones within the beam. These losses can be significantly mitigated through sonication geometry planning, patient positioning, and aberration correction using multielement phased arrays. Here, a sonication planning algorithm is introduced, which uses the simulations to select the optimal transducer position and evaluate the effect of aberrations and acoustic field quality at the target region after aberration correction. Optimization of transducer positioning is implemented using a graphical user interface (GUI) to visualize a segmented 3-D computed tomography (CT)-based acoustic model of the body and to select sonication geometry through a combination of manual and automated approaches. An HIFU array (1.5 MHz, 256 elements) and three renal cell carcinoma (RCC) cases with different tumor locations and patient body habitus were considered. After array positioning, the correction of aberrations was performed using a combination of backpropagation from the focus with an ordinary least squares (OLS) optimization of phases at the array elements. The forward propagation was simulated using a combination of the Rayleigh integral and k-space pseudospectral method (k-Wave toolbox). After correction, simulated HIFU fields showed tight focusing and up to threefold higher maximum pressure within the target region. The addition of OLS optimization to the aberration correction method yielded up to 30% higher maximum pressure compared to the conventional backpropagation and up to 250% higher maximum pressure compared to the ray-tracing method, particularly in strongly distorted cases.
科研通智能强力驱动
Strongly Powered by AbleSci AI