Design and Optimization of Ultrasound Phased Arrays for Large-Scale Ultrasound Neuromodulation

Transcranial ultrasound stimulation applied in ischemic stroke rehabilitation: A review

“…Ultrasound stimulation at different parameters induces different therapeutic effects. TFUS can utilize ultrasound phased array technology to electronically drive an ultrasound transducer array to direct focused ultrasound beams to different neural targets, enabling large-scale ultrasound neuromodulation within a given tissue volume (Monteith et al, 2013;Ilham et al, 2021), it can also transmit acoustic energy to a target area in the brain through single-element transducers, acting on a focal point (Park et al, 2022). LIFU (frequency:<1 MHz; intensity: 0.5-100 W/cm 2 ) (Tyler et al, 2018) stimulates nerve tissue mainly through the pressure generated by ultrasonic radiation, improving the blood supply around the brain lesion by means of neural regulation without causing tissue damage (Bystritsky et al, 2011;Fomenko et al, 2018;Wang et al, 2022a).…”

32 Element Piezoelectric Micromachined Ultrasound Transducer (PMUT) Phased Array for Neuromodulation

“…Based on the design procedure presented in [30], the geometries, i.e., element length (L), element pitch (d), element width (a), and the number of elements (N) of an US phased array were optimized using k-Wave, which is an open-source MATLAB toolbox. Since the element thickness is not considered in k-Wave, the optimization process for a bulk array in [30] can still be used for the 2D geometry optimization of the PMUT array. The array was designed for tFUS of a rat's brain with high FoM while having the least off target stimulation effects.…”

“…For this design, the following assumptions were made: 1) Array aperture D and element length L were limited by the nominal dimensions of a rat's head (Dmax = L max = 25 mm) ; 2) Focal distance, F, was set to 20 mm so that it can cover the standard depth of a rat's brain [31]; 3) the resonant frequency, f, was set to ~ 1 MHz considering high attenuation encountered by higher frequency US penetrating the scalp, skull, and brain tissue [32]; 4) maximum steering angle (θ s,max ) was ±60°; 5) the minimum kerf of 96.5 μm (λ/16, where λ is the wavelength) was set as the grid resolution of the k-Wave simulations; 6) H(y) < 0.7, where H(y) is an equivalent to the directivity function and is defined as the ratio of the peak output pressure that occurred on a line parallel to the x axis (corresponds to different y values with z = 0 in Fig. 1) to the peak output pressure at the focal spot (corresponds to a single point in the whole xy plane) [30]. Since the US beam area is defined by half of the maximum power, which corresponds to ~ 0.7 of peak pressure, a threshold of 0.7 is reasonable for H(y).…”

“…Fig. 2 demonstrates the results acquired by implementing the optimization method described in [30]. Setting the initial values of the interelement spacing, d = λ/2 = 0.78 mm, and the element width, a = d/2 = 0.385 mm, first the element length, L was optimized for the maximum FoM.…”

“…A design methodology for geometry optimization for large-scale US neuromodulation has been proposed in [30]. The design method maximizes a figure of merit (FoM) that simultaneously considers the total input power consumed by the array, the peak pressure engendered at the focal spot, and the overall focal volume defined by the half power beam width.…”