Noninvasive Focused Ultrasound Spinal Cord Stimulation in Humans: A Computational Feasibility Study

TL;DR

This computational study demonstrates the feasibility of trans-spinal focused ultrasound in humans, highlighting the importance of transducer positioning, vertebral level, and anatomical pathways for effective spinal cord stimulation.

Contribution

First anatomically realistic computational model assessing trans-spinal focused ultrasound feasibility in humans, guiding future device design and procedural strategies.

Findings

Transducer positioning critically affects target intensity.

Thoracic spine yields higher intensity than cervical.

Vertebral lamina sonication increases intraspinal intensity.

Abstract

Background: Trans-spinal FUS (tsFUS) has recently been shown promise in modulating spinal reflexes in rodents, opening new avenues for spinal cord interventions in motor control and pain management. However, anatomical differences between rodents and human spinal cords require careful targeting strategies and transducer design adaptations for human applications. Aim: This study aims to computationally explore the feasibility of tsFUS in the human spinal cord by leveraging the intervertebral acoustic window and vertebral lamina. Method: Acoustic simulations were performed using an anatomically detailed human spinal cord model with an adapted single-element focusing transducer (SEFT) to investigate the focality and intensity of the acoustic quantities generated within the spinal cord. Results: Sonication through the intervertebral acoustic window using an adapted transducer achieved approximately 2-fold higher intensity and up to 20% greater beam overlap compared to commercial SEFT. Precise transducer positioning was critical; a 10 mm vertical shift resulted in a reduction of target intensity by approximately 7-fold. The vertebral level also substantially influenced the sonication outcomes, with the thoracic spine achieving 6-fold higher intensity than the cervical level. Sonication through the vertebral lamina resulted in approximately 2.5-fold higher intraspinal intensity in the cervical spine. Conclusion and Significance: This study presents the first systematic, anatomically realistic computational model for the feasibility of tsFUS in humans. Quantifying the trade-offs between acoustic path, vertebral level, and transducer geometry offers foundational design and procedural guidelines for the design and safety of spinal FUS protocols, supporting future studies of trans-spinal FUS in humans.