← Explore

Device for Multifocal Delivery of Ultrasound Into Deep Brain Regions in Humans

Thomas Riis, Daniel Feldman, Adam Losser, Brian Mickey, Jan Kubanek

IEEE Transactions on Biomedical Engineering 2024, 71, 660-668 · 10.1109/tbme.2023.3313987

human patientdepressionclinical scaleother mri

Abstract

Low-intensity focused ultrasound provides the means to noninvasively stimulate or release drugs in specified deep brain targets. However, successful clinical translations require hardware that maximizes acoustic transmission through the skull, enables flexible electronic steering, and provides accurate and reproducible targeting while minimizing the use of MRI. We have developed a device that addresses these practical requirements. The device delivers ultrasound through the temporal and parietal skull windows, which minimize the attenuation and distortions of the ultrasound by the skull. The device consists of 252 independently controlled elements, which provides the ability to modulate multiple deep brain targets at a high spatiotemporal resolution, without the need to move the device or the subject. And finally, the device uses a mechanical registration method that enables accurate deep brain targeting both inside and outside of the MRI. Using this method, a single MRI scan is necessary for accurate targeting; repeated subsequent treatments can be performed reproducibly in an MRI-free manner. We validated these functions by transiently modulating specific deep brain regions in two patients with treatment-resistant depression.

Abstract via europepmc.

Specieshuman
Subjects2 participants
Sessions per subject1
Randomisedyes
Blindingsingle
Sham / controlunfocused or detuned
Auditory controlnot reported
Readout timingoffline
Anaesthesianot applicable
Readoutsclinical scale, other mriPsychiatrist-rated 7-point scale for depression and anxiety mood states; Generic Assessment of Side Effects (GASE) questionnaire; structural T1-w and T2-w MRI
Direction of effectinhibitoryUltrasound was delivered at a low duty cycle intended to suppress (inhibit) neural activity in the hyperactive subgenual cingulate cortex; stimulation for 60 s or longer led to significant improvements (reductions) in depression and anxiety metrics, with no effect from sham.
Adverse eventsnone observedNo adverse effects were noted by either subject or the attending psychiatrist; structural MRI showed no apparent changes.

Exposures

Exposure 1: Subgenual cingulate cortex (SGC) stimulation

Target: subgenual anterior cingulate cortex — “subgenual cingulate cortex (SGC)
Device: custom-built

Pulse timing
Waveformpulsed
Fundamental frequency (kHz)650✓✓
Pulse duration (ms)30✓✓
Pulse repetition frequency (Hz)not reported
Duty cycle (%)0.75✓✓
Sonication duration (s)60, 180swept✓✓
Pressure and intensity, by domain
Free-field pressure (kPa)not reported
Free-field Isppa (W/cm²)not reported
Free-field Ispta (W/cm²)not reported
In-situ estimatemeasurementsingle value
In-situ pressure (kPa)1,000✓✓
In-situ Isppa (W/cm²)31✓✓
In-situ Ispta (W/cm²)0.233✓✓
Protocol, in the paper’s words

Ultrasound was delivered into the target in 30 ms ON periods followed by 4 s OFF periods; sonication duration varied from 60-180 s. Targets were presented randomly in a blinded manner, interleaved with an active sham (unfocused plane wave with the same energy and waveform) over a 1.5 hour neuromodulation session; a psychiatrist rated mood immediately after each of 27 total stimuli.

Exposure 2: Ventral striatum stimulation

Target: nucleus accumbens — “ventral striatum
Device: custom-built

Pulse timing
Waveformpulsed
Fundamental frequency (kHz)650✓✓
Pulse duration (ms)30✓✓
Pulse repetition frequency (Hz)not reported
Duty cycle (%)0.75✓✓
Sonication duration (s)60, 180swept✓✓
Pressure and intensity, by domain
Free-field pressure (kPa)not reported
Free-field Isppa (W/cm²)not reported
Free-field Ispta (W/cm²)not reported
In-situ estimatemeasurementsingle value
In-situ pressure (kPa)1,000✓✓
In-situ Isppa (W/cm²)31✓✓
In-situ Ispta (W/cm²)0.233✓✓
Protocol, in the paper’s words

Ultrasound was delivered into the target in 30 ms ON periods followed by 4 s OFF periods; sonication duration varied from 60-180 s. Targets were presented randomly in a blinded manner, interleaved with an active sham (unfocused plane wave with the same energy and waveform) over a 1.5 hour neuromodulation session.

Flags from extraction

  • exposures[0].timing.pulse_repetition_frequency_hzPaper gives 30 ms ON and 4 s OFF periods and a duty cycle (0.75%) but never labels a PRF value; computing PRF from the ON+OFF period would require summing two separately stated numbers, which is arithmetic this extraction avoids.
  • sham_typeActive sham delivered the same energy/waveform as an unfocused (defocused) plane wave rather than a powered-off transducer or a different anatomical site, which does not cleanly match a listed sham_type category.
  • direction_of_effectDirection is based on the authors' stated intent to suppress SGC activity via low duty cycle and the resulting mood improvement; no direct neural activity measurement (e.g., imaging) confirmed inhibition in this paper.
  • n_subjectsOnly 2 patients were studied; findings are proof-of-concept.