A Spine-Specific Phased Array for Transvertebral Ultrasound Therapy: Design and Simulation

Objective: To design and simulate the performance of two spine-specific phased arrays in sonicating targets spanning the thoracic spine, with the objective of efficiently producing controlled foci in the spinal canal. Methods: Two arrays (256 elements each, 500 kHz) were designed using multi-layered ray acoustics simulation: a four-component array with dedicated components for sonicating via the paravertebral and transvertebral paths, and a two-component array with spine-specific adaptive focusing. Mean array efficiency (canal focus pressure/water focus pressure) was evaluated using forward simulation in neutral and flexed spines to investigate methods that reduce spine-induced insertion loss. Target-specific four-component array reconfiguration and lower frequency sonication (250 kHz) were tested to determine their effects on array efficiency and focal dimensions. Results: When neutral, two- and four-component efficiencies were 32 ± 11% and 29 ± 13%, respectively, spine flexion significantly increased four-component efficiency (36 ± 18%), but not two-component efficiency (33 ± 15%). Target-specific four-component re-configuration significantly improved efficiency (36 ± 8%). Both arrays produced controlled foci centered within the canal with similar 50% pressure contour dimensions: 10.8-11.9 mm (axial), 4.2-5.6 mm (lateral), and 5.9-6.2 mm (vertical). Simulation at 250 kHz also improved two- and four-component efficiency (43 ± 17% and 36 ± 13%, respectively), but doubled the lateral focal dimensions. Conclusion: Simulation shows that the spine-specific arrays are capable of producing controlled foci in the thoracic spinal canal. Significance: The complex geometry of the human spine presents geometrical and acoustical challenges for transspine ultrasound focusing, and the design of these spine-specific ultrasound arrays is crucial to the clinical translation of focused ultrasound for the treatment of spinal cord disease.