Shen Che-Chou, Chen You-An, Ku Hsin-Yu
Department of Electrical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan.
Department of Electrical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan.
Ultrasonics. 2023 Dec;135:107125. doi: 10.1016/j.ultras.2023.107125. Epub 2023 Jul 27.
High-intensity focused ultrasound (HIFU) is a promising non-invasive treatment method whose applications include tissue ablation, hemostasis, thrombolysis and blood-brain barrier opening etc. Its therapeutic effects come from the thermal necrosis and the mechanical destruction associated with acoustic cavitation. Passive acoustic mapping (PAM) is capable of simultaneous monitoring of HIFU-induced cavitation events using only receive beamforming. Nonetheless, conventional time exposure acoustics (TEA) algorithm has poor spatial resolution and suffers from the X-shaped artifacts. These factors lead to difficulties in precise localization of cavitation source. In this study, we proposed a novel adaptive PAM method which combines Delay-Multiply-and-Sum (DMAS) beamforming with virtual augmented aperture (VA) to overcome the problem. In DMAS-VA beamforming, the magnitude of each channel waveform is scaled by p-th root while the phase is multiplied by L. The p and L correspond respectively to the degree of signal coherence in DMAS beamforming and the augmentation factor of aperture size. After channel sum, p-th power is applied to restore the dimensionality of source strength and then the PAM image is reconstructed by accumulating the signal power over the observation time. Based on simulation and experimental results, the proposed DMAS-VA has better image resolution and image contrast compared with the conventional TEA. Moreover, since the VA method may introduce grating lobes into PAM because of the virtually augmented pitch size, DMAS coherent factor (DCF) is further developed to alleviate these image artifacts. Results indicate that, with DCF weighting, the PAM image of DMAS-VA beamforming could be constructed without detectable image artifacts from grating lobes and false main lobes.
高强度聚焦超声（HIFU）是一种很有前景的非侵入性治疗方法，其应用包括组织消融、止血、溶栓和血脑屏障开放等。其治疗效果源于热坏死以及与声空化相关的机械破坏。被动声学映射（PAM）仅使用接收波束形成就能同时监测HIFU诱导的空化事件。然而，传统的时间曝光声学（TEA）算法空间分辨率较差，且存在X形伪像。这些因素导致空化源的精确定位困难。在本研究中，我们提出了一种新颖的自适应PAM方法，该方法将延迟相乘求和（DMAS）波束形成与虚拟增强孔径（VA）相结合以克服该问题。在DMAS-VA波束形成中，每个通道波形的幅度按p次方根缩放，而相位乘以L。p和L分别对应于DMAS波束形成中的信号相干程度和孔径大小的增强因子。通道求和后，应用p次方来恢复源强度的维度，然后通过在观察时间内累积信号功率来重建PAM图像。基于仿真和实验结果，所提出的DMAS-VA与传统TEA相比具有更好的图像分辨率和图像对比度。此外，由于VA方法可能会因虚拟增大的间距尺寸而在PAM中引入栅瓣，因此进一步开发了DMAS相干因子（DCF）以减轻这些图像伪像。结果表明，通过DCF加权，可以构建DMAS-VA波束形成的PAM图像，而不会出现来自栅瓣和虚假主瓣的可检测图像伪像。
