Robertson, James; (2017) Accurate Simulation of Low-Intensity Transcranial Ultrasound Propagation for Neurostimulation. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text James Robertson PhD 18.10.17 Electronic Copyright Removed Version.pdf - Accepted Version Download (115MB) | Preview

Abstract

Neural stimulation with low-intensity ultrasound is a potentially transformative technology with applications in therapy and research. To develop, it will require ultrasound to be tightly focused on brain structures with accurate spatial targeting and fine control over the ultrasound amplitude at the target. However, the skull is an impediment to the effective focusing of ultrasound. Simulations of ultrasound propagation through acoustic property maps derived from medical images can be used to derive focusing drive signals for multi-element arrays. Focusing effectiveness is dependent on the fidelity of the numerical simulations. In combination with MRI based treatment verification, model based focusing has been used to focus high-intensity ultrasound onto the brain for ablation. This thesis presents a thorough and systematic study of the simulation parameters required to achieve effective transcranial focusing. The literature on ultrasonic neurostimulation, transcranial ultrasonic focusing, and the derivation of property maps from medical images is reviewed. The sampling criteria required to ensure numerical accuracy for the k-space pseudospectral time domain simulation scheme is established through testing of individual sources of numerical error, and convergence testing of a simulated time-reversal protocol. With numerical accuracy assured, the importance of acoustic property maps is examined through simulations to determine the sensitivity of intracranial fields to the properties of the skull layer. These results are corroborated by matching experimental measurements of ultrasound propagation through skull bone phantoms with spatially registered simulations. Finally, the impact of image related homogenisation and loss of internal bone structure is determined using simulations through co-registered clinical CT and micro-CT data of the skull.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item