Computational optimization of transcranial focused ultrasound stimulation: Toward noninvasive, selective stimulation of deep brain structures

Low-intensity focused ultrasound is emerging as a high-resolution highly selective alternative to standard noninvasive transcranial brain stimulation techniques. A major challenge in using ultrasound devices is designing a stimulator capable of efficiently focusing the acoustic wave to selectively target a specific brain region by compensating for the wavefront distortions induced by the intact skull. Single-element transducers are efficient in stimulating cortical areas in both non-human and human primates. However, reaching deeper brain structures with millimeter resolution and high specificity requires the use of ad hoc multi-element devices characterized by a specific number of piezoelectric elements that optimize the energy deposition in the focal region while simultaneously minimizing the off-focus dispersion. The high cost and complexity of adequately controlling the thousands of elements used generally for such stimulators have limited their use in neuromodulation applications. This study defines the optimal configuration of a multi-element stimulator for low-intensity focused ultrasound through a full-wave realistic numerical model that includes both the stimulator geometry and a detailed anatomical head model. The performance of the device was evaluated. We investigated the influence of the number of piezoelectric elements in the stimulator on its transcranial focusing capabilities. Our results confirm that the focusing optimization improves as the number of elements increased (from 16 to 256). With only 96 point-sources, there was a good trade-off between cost and focusing efficiency. Our study provides a cost-effective stimulator design that enables a standard focusing procedure and a steering technique enacted without prior knowledge about the skull's local acoustic impedance.