Masoumi Mohammad Hadi, Kaddoura Tarek, Zemp Roger
IEEE Trans Ultrason Ferroelectr Freq Control. 2024 Jun;71(6):648-658. doi: 10.1109/TUFFC.2024.3400229. Epub 2024 Jun 10.
Two-dimensional sparse arrays and row-column arrays are both alternatives to 2-D fully addressed arrays with lower channel counts. Row-column arrays have recently demonstrated fast 3-D structural and flow imaging but commonly suffer from high grating lobes or require multiplexing to achieve better quality. Two-dimensional sparse arrays enable full-volume acquisitions for each transmit event, but plane-wave transmissions with them usually lack quality in terms of uniformity of wavefronts. Here, we propose a novel architecture that combines both types of these arrays in one aperture, enabling imaging using row-column or sparse arrays alone or a hybrid imaging scheme where the row-column array is used in transmission and a 2-D sparse array in reception. This hybrid imaging scheme can potentially solve the shortcomings of each of these approaches. The sparse array layout chosen is a Costas array, characterized by having only one element per row and column, facilitating its integration with row-column arrays. We simulate images acquired with TOBE-Costas arrays using the hybrid imaging scheme and compare them to row-column and sparse spiral arrays of equivalent aperture size (128λ × 128λ at 7.5 MHz) in ultrafast plane-wave imaging of point targets and 3-D power Doppler imaging of synthetic flow phantoms. Our simulation results show that TOBE-Costas arrays exhibit superior resolution and lower sidelobe levels compared with plane-wave compounding with row-column arrays. Compared with density-tapered spiral arrays, they provide a larger field of view and finer resolution.
二维稀疏阵列和行列阵列都是通道数较少的二维全寻址阵列的替代方案。行列阵列最近已展示出快速的三维结构和血流成像,但通常会受到高旁瓣的影响,或者需要复用才能获得更好的质量。二维稀疏阵列能够在每次发射事件时进行全容积采集,但使用它们进行平面波传输时,波前均匀性方面通常较差。在此,我们提出一种新颖的架构,将这两种类型的阵列组合在一个孔径中,能够单独使用行列阵列或稀疏阵列进行成像,或者采用一种混合成像方案,即发射时使用行列阵列,接收时使用二维稀疏阵列。这种混合成像方案有可能解决这些方法各自的缺点。所选择的稀疏阵列布局是一种科斯塔斯阵列,其特点是每行每列只有一个元件,便于与行列阵列集成。我们使用混合成像方案模拟了用TOBE - 科斯塔斯阵列采集的图像,并将其与等效孔径尺寸(7.5 MHz时为128λ×128λ)的行列阵列和稀疏螺旋阵列在点目标的超快平面波成像以及合成血流模型的三维功率多普勒成像中进行比较。我们的模拟结果表明,与使用行列阵列的平面波复合成像相比,TOBE - 科斯塔斯阵列具有更高的分辨率和更低的旁瓣水平。与密度渐变螺旋阵列相比,它们提供了更大的视野和更精细的分辨率。