A Magnetic Resonance Imaging-Compatible, Large-Scale Array for Trans-Skull Ultrasound Surgery and Therapy

TLDR

Advances in ultrasound transducer arrays and amplifiers have spurred proposals for noninvasive brain surgery, including a 500‑element hemispherical array that corrects skull‑induced wave distortion at ~0.8 MHz. This work seeks to develop a clinically viable hemispheric phased‑array system optimized for trans‑skull focusing. The array was tested by phase‑adaptive focusing through ex vivo human skulls into a brain phantom while being monitored with MRI guidance and thermometry. The 500‑element, 30‑cm‑diameter, 0.81‑MHz array achieved electronically steerable focus over ~30 × 30 × 26 mm, coagulated target tissue without excessive skull heating, and produced high‑intensity lesions without generating significant MRI noise or artifacts.

Abstract

Objective Advances in ultrasound transducer array and amplifier technologies have prompted many intriguing scientific proposals for ultrasound therapy. These include both mildly invasive and noninvasive techniques to be used in ultrasound brain surgery through the skull. In previous work, it was shown how a 500-element hemisphere-shaped transducer could correct the wave distortion caused by the skull with a transducer that operates at a frequency near 0.8 MHz. Because the objective for trans-skull focusing is its ultimate use in a clinical context, a new hemispheric phased-array system has now been developed with acoustic parameters that are optimized to match the values determined in preliminary studies. Methods The transducer was tested by focusing ultrasound through ex vivo human skulls and into a brain phantom by means of a phase-adaptive focusing technique. Simultaneously, the procedure was monitored by the use of magnetic resonance guidance and thermometry. Results The ultrasound focus of a 500-element 30-cm-diameter, 0.81-MHz array could be steered electronically through the skull over a volume of approximately 30 × 30 × 26 mm. Furthermore, temperature monitoring of the inner and outer surfaces of the skull showed that the array could coagulate targeted brain tissue without causing excessive skull heating. Conclusions The successful outcome of these experiments indicates that intensities high enough to destroy brain tissue can be produced without excessive heating of the surrounding areas and without producing large magnetic resonance noise and artifacts.