A numerical analysis of anti-focus driving to improve focal quality in transvertebral focused ultrasound.

Ultrasound-mediated drug delivery holds promise for treating spinal cord diseases, but the spinal canal is a challenging focused ultrasound target. This in silico study explores Tikhonov-regularized anti-focus (AF) driving to improve transspinal focal quality when focusing with a 256-element, 400-kHz spine-specific array. Three focusing methods were simulated for 80 targets in an ex vivo human spine: (1) uncorrected (water-case) phases and amplitudes, (2) simulation-derived phases and inverse amplitude correction, and (3) AF-derived phases and amplitude shading. Simulation-derived correction reduced mean targeting error from 4.6 ± 2.1 mm to 3.2 ± 3.1 mm relative to the uncorrected case. AF-derived driving further reduced mean targeting error to 1.4 ± 0.9 mm but decreased driving efficiency (maximum pressure normalized to driving amplitude) by 51%. Simulations were then extended to 304 targets in 4 human spines from an open-source CT dataset. Targeting error was reduced from 5.2 ± 2.8 mm in the uncorrected case to 4.0 ± 3.2 mm when using phase and inverse amplitude correction (23% decrease), whereas AF-derived driving reduced mean targeting error to 1.6 ± 1.4 mm (69% decrease). In conclusion, AF-based driving greatly reduces targeting error in transspinal focused ultrasound simulations at a cost of reduced driving efficiency.