Ultrasound Produces Extensive Brain Activation via a Cochlear Pathway

Abstract

Ultrasound (US) can noninvasively activate intact brain circuits, making it a promising neuromodulation technique. However, little is known about the underlying mechanism. Here, we apply transcranial US and perform brain mapping studies in guinea pigs using extracellular electrophysiology. We find that US elicits extensive activation across cortical and subcortical brain regions. However, transection of the auditory nerves or removal of cochlear fluids eliminates the US-induced activity, revealing an indirect auditory mechanism for US neural activation. Our findings indicate that US activates the ascending auditory system through a cochlear pathway, which can activate other non-auditory regions through cross-modal projections. This cochlear pathway mechanism challenges the idea that US can directly activate neurons in the intact brain, suggesting that future US stimulation studies will need to control for this effect to reach reliable conclusions.

Abstract via europepmc.

Flags from extraction

• exposures[0].timing.pulse_duration_ms — Value combines the single-pulse experiments (0.1-10 ms PD) with A1; a separate general statement gives 0.1-2.5 ms PD for multi-pulse trials on A1 or SC1 combined, not separated per target.
• exposures[0].timing.pulse_repetition_frequency_hz — PRF range (10 Hz-16 kHz) is given in aggregate for A1 and SC1 combined sweep ('tested on A1 or SC1'), not separated per target in the main text; supplemental Tables S1-S3 (not available in provided text) likely give per-target values.
• exposures[0].timing.duty_cycle_pct — Duty cycle range (0.2-60%) is an aggregate figure for A1 and SC1 combined, not separated per target.
• exposures[0].timing.sonication_duration_s — 'TD' (trial duration) may represent the total trial/recording window rather than the active sonication train length implied by number-of-pulses x PRF in some multi-pulse conditions (e.g., 20 pulses at 1 kHz PRF spans ~20 ms, yet TD is reported as up to 6 s); relationship not reconcilable from stated parameters. Range also aggregates A1 and SC1 sweeps.
• exposures[0].free_field.pressure_kpa — Range (25 kPa-2 MPa) aggregates A1 and SC1 combined sweep from the Experiment Design paragraph, not separated per target.
• exposures[1].timing.duty_cycle_pct — Duty cycle range (0.2-60%) is an aggregate figure for A1 and SC1 combined, not separated per target.
• exposures[1].free_field.pressure_kpa — Range (25 kPa-2 MPa) aggregates A1 and SC1 combined sweep; explicit SC1 example values found in text were 50, 200 and 400 kPa.
• exposures[2].timing.pulse_duration_ms — Quote provided is the closest available complete sentence; the specific parameter set for the visual-cortex exposure (400 kPa, 1 kHz PRF, 2.5 ms PD, 20 pulses, 6 s TD) is stated in the Figure 6 legend, reproduced in protocol_description. Note the stated 2.5 ms pulse duration exceeds the 1 ms period implied by a 1 kHz PRF, an internal inconsistency in the source.
• n_subjects — Different sub-experiments in this study report different, likely overlapping, animal-group sizes (e.g., n=2 for combined A1/ICC recording, n=7 for overall SC1 activation percentage, n=4 for TD/PRF and visual-cortex experiments, n=3 for nerve transection, n=3 for cochlear fluid removal); the paper does not state a single total number of distinct animals used in the study.
• exposures[0].in_situ.method — Craniotomy exposed the brain directly (no intact skull) in this preparation; paper does not separately report in-situ (post-skull) values distinct from the free-field water-tank calibration, so left not_reported rather than assumed equal.
• exposures[1].in_situ.method — Craniotomy exposed the brain directly (no intact skull) in this preparation; paper does not separately report in-situ (post-skull) values distinct from the free-field water-tank calibration, so left not_reported rather than assumed equal.
• exposures[2].in_situ.method — Craniotomy exposed the brain directly (no intact skull) in this preparation; paper does not separately report in-situ (post-skull) values distinct from the free-field water-tank calibration, so left not_reported rather than assumed equal.