Auditory Mondrian masks the airborne-auditory artifact of focused ultrasound stimulation in humans
2023; Elsevier BV; Volume: 16; Issue: 2 Linguagem: Inglês
10.1016/j.brs.2023.03.002
ISSN1935-861X
AutoresWilliam Liang, Hongsun Guo, David R. Mittelstein, Mikhail G. Shapiro, Shinsuke Shimojo, Mohammad Shehata,
Tópico(s)Ultrasound and Cavitation Phenomena
ResumoLow-intensity Focused Ultrasound Stimulation (FUS) can modulate neural activity in cortical, subcortical, and deep brain regions, achieving millimeter precision through transcranial ultrasound stimulation (TUS) and can affect behavior [[1]Rabut C. Yoo S. Hurt R.C. Jin Z. Li H. Guo H. et al.Ultrasound technologies for imaging and modulating neural activity.Neuron. 2020; 108: 93-110Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar,[2]Rezayat E. Toostani I.G. A review on brain stimulation using low intensity focused ultrasound.Basic Clin Neurosci. 2016; 7: 187-194PubMed Google Scholar]. However, there is concern that FUS may induce an auditory effect, cortically activating the subject's auditory sensation, thus confounding behavioral and electrophysiological responses during TUS research [3Guo H. Hamilton Ii M. Offutt S.J. Gloeckner C.D. Li T. Kim Y. et al.Ultrasound produces extensive brain activation via a cochlear pathway.Neuron. 2018; 99: 866Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar, 4Sato T. Shapiro M.G. Tsao D.Y. Ultrasonic neuromodulation causes widespread cortical activation via an indirect auditory mechanism.Neuron. 2018; 98: 1031-10341 e5Abstract Full Text Full Text PDF PubMed Scopus (169) Google Scholar, 5Johnstone A. Nandi T. Martin E. Bestmann S. Stagg C. Treeby B. A range of pulses commonly used for human transcranial ultrasound stimulation are clearly audible.Brain Stimul. 2021; 14: 1353-1355Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar, 6Braun V. Blackmore J. Cleveland R.O. Butler C.R. Transcranial ultrasound stimulation in humans is associated with an auditory confound that can be effectively masked.Brain Stimul. 2020; 13: 1527-1534Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar, 7Park C. Chen M. Kim T. Implication of auditory confounding in interpreting somatosensory and motor responses in low-intensity focused transcranial ultrasound stimulation.J Neurophysiol. 2021; 125: 2356-2360Crossref PubMed Scopus (5) Google Scholar]. Several studies have mitigated this auditory artifact through ramping [[8]Mohammadjavadi M. Ye P.P. Xia A. Brown J. Popelka G. Pauly K.B. Elimination of peripheral auditory pathway activation does not affect motor responses from ultrasound neuromodulation.Brain Stimul. 2019; 12: 901-910Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar] or audio masking [[5]Johnstone A. Nandi T. Martin E. Bestmann S. Stagg C. Treeby B. A range of pulses commonly used for human transcranial ultrasound stimulation are clearly audible.Brain Stimul. 2021; 14: 1353-1355Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar,[6]Braun V. Blackmore J. Cleveland R.O. Butler C.R. Transcranial ultrasound stimulation in humans is associated with an auditory confound that can be effectively masked.Brain Stimul. 2020; 13: 1527-1534Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar], but prior research has had two main limitations. First, previous studies addressed a limited range of the total FUS parameter space: fundamental frequency (f0), pulse repetition frequency (PRF), duty cycle (DC), sonication duration (SD), and intensity (I) [[1]Rabut C. Yoo S. Hurt R.C. Jin Z. Li H. Guo H. et al.Ultrasound technologies for imaging and modulating neural activity.Neuron. 2020; 108: 93-110Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar,[7]Park C. Chen M. Kim T. Implication of auditory confounding in interpreting somatosensory and motor responses in low-intensity focused transcranial ultrasound stimulation.J Neurophysiol. 2021; 125: 2356-2360Crossref PubMed Scopus (5) Google Scholar,9Younan Y. Deffieux T. Larrat B. Fink M. Tanter M. Aubry J.F. Influence of the pressure field distribution in transcranial ultrasonic neurostimulation.Med Phys. 2013; 40082902Crossref PubMed Scopus (131) Google Scholar, 10King R.L. Brown J.R. Newsome W.T. Pauly K.B. Effective parameters for ultrasound-induced in vivo neurostimulation.Ultrasound Med Biol. 2013; 39: 312-331Abstract Full Text Full Text PDF PubMed Scopus (311) Google Scholar, 11Fomenko A. Neudorfer C. Dallapiazza R.F. Kalia S.K. Lozano A.M. Low-intensity ultrasound neuromodulation: an overview of mechanisms and emerging human applications.Brain Stimul. 2018; 11: 1209-1217Abstract Full Text Full Text PDF PubMed Scopus (138) Google Scholar]. Second, as these studies used direct stimulation of subjects, they did not distinguish between airborne or tissue conduction effects [5Johnstone A. Nandi T. Martin E. Bestmann S. Stagg C. Treeby B. A range of pulses commonly used for human transcranial ultrasound stimulation are clearly audible.Brain Stimul. 2021; 14: 1353-1355Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar, 6Braun V. Blackmore J. Cleveland R.O. Butler C.R. Transcranial ultrasound stimulation in humans is associated with an auditory confound that can be effectively masked.Brain Stimul. 2020; 13: 1527-1534Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar, 7Park C. Chen M. Kim T. Implication of auditory confounding in interpreting somatosensory and motor responses in low-intensity focused transcranial ultrasound stimulation.J Neurophysiol. 2021; 125: 2356-2360Crossref PubMed Scopus (5) Google Scholar]. Here, we evaluate the airborne auditory artifact over a range of FUS parameters through sonographic characterization, the human subject's response to recorded audio clips, and a two-interval forced choice (2IFC) task to test the effectiveness of three mask types: square [[5]Johnstone A. Nandi T. Martin E. Bestmann S. Stagg C. Treeby B. A range of pulses commonly used for human transcranial ultrasound stimulation are clearly audible.Brain Stimul. 2021; 14: 1353-1355Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar,[6]Braun V. Blackmore J. Cleveland R.O. Butler C.R. Transcranial ultrasound stimulation in humans is associated with an auditory confound that can be effectively masked.Brain Stimul. 2020; 13: 1527-1534Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar], pulsed sine, and random multitone. The multitone random mask, or Auditory Mondrian, is inspired by the visual Mondrian used in the continuous flash suppression to mask visual targets [[12]Tsuchiya N. Koch C. Continuous flash suppression reduces negative afterimages.Nat Neurosci. 2005; 8: 1096-1101Crossref PubMed Scopus (656) Google Scholar]. We recruited 228 healthy participants for the three online auditory psychophysical experiments (See Supplementary Methods for details). In experiment 1, participants performed a detection task in which they were asked whether they detected a distinct sound while listening to audio recordings of FUS sham and stimulation trials. In experiments 2 and 3, participants performed a two-interval forced choice (2IFC) task in which they chose which interval of a pair contained the FUS stimulation embedded in an auditory mask. Audio clips from the microphone were used without volume (loudness) manipulation and confirmed by experimenters to match the sound produced from the FUS setup. In an artificial environment, we found that the ultrasound transducer is a primary source of airborne auditory artifacts (Fig. S1 A and B). Short-time Fourier transforms (STFT) of the audio recordings of FUS revealed clear frequency bands at the PRF and harmonics, along with additional frequency bands in the human hearing range that did not fit with the corresponding PRF (Fig. 1 A). These additional frequency bands were consistent at approximately 8 and 12 kHz throughout all PRF and even seen with continuous wave US bursts (Fig. 1 A, yellow arrows) and appeared regardless of the coupling method or the cone and arm setup (Fig. 1 B). The electrical spectrum density showed no peaks at the human frequency range, so it likely did not contribute to the auditory artifact. Based on the above acoustic analyses, we concluded that the ultrasound transducer is a source of airborne auditory artifacts. In the detection task (experiment 1), participants significantly detected the auditory artifact (see supplementary audio tones 1 and 2 and their STFT in Figs. S2–3) with high confidence from the sham at nearly all tested parameters (Fig. S4, paired t-test from chance, Bonferroni corrected p < 0.001). For continuous US bursts (Fig. S4 A-C), an F0 of 375 kHz was significantly undetected compared to other F0, likely due to diminished transducer efficiency (one-way repeated measures ANOVA F(3, 147) = 100.672, p = 0.000, with Tukey post-hoc test p < 0.5). An SD of 125 ms was the most perceivable, with a gradual decline as the duration increased (one-way repeated measures ANOVA F(4, 196) = 13.966, p = 0.000, with Tukey post-hoc test p < 0.5 between 50 ms and 125 ms). Detection increased linearly with increasing pressure (one-way repeated measures ANOVA F(2, 98) = 119.327, p = 0.000, with Tukey post-hoc test p < 0.5 between P = 0.4 MPa and other pressures). For pulsed US bursts (Fig S4 D-I), there was a significant decline at PRF = 20 kHz, the upper range of the human hearing range, in both detection level and confidence (d prime, one-way repeated measures ANOVA F(7, 343) = 35.634, p = 0.000, with Tukey post-hoc test p < 0.5 between PRF = 20 kHz and other frequencies). In the 2IFC tasks (experiments 2 and 3), human participants testing confirmed that Mondrian outperformed Monotone audio masks at masking airborne auditory artifacts (Fig. 1C and D). We tested the effectiveness of a square mask (1 kHz, similar to previously reported [[5]Johnstone A. Nandi T. Martin E. Bestmann S. Stagg C. Treeby B. A range of pulses commonly used for human transcranial ultrasound stimulation are clearly audible.Brain Stimul. 2021; 14: 1353-1355Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar,[6]Braun V. Blackmore J. Cleveland R.O. Butler C.R. Transcranial ultrasound stimulation in humans is associated with an auditory confound that can be effectively masked.Brain Stimul. 2020; 13: 1527-1534Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar]), a pulsed sine mask (PF = 7 kHz, PRF = 1 kHz, and DC = 50%, simulating the FUS), and a multitone random mask (Auditory Mondrian) consisting of eight overlapping mini-pulsed-sine tones (SD = 500 ms, PF = 15 kHz, PRF = 1–15 kHz, DC = 50%) with random shifts and a total duration of 2 sec (see supplementary audio tones 3–12 and Fig. S5). The square and pulsed sine masks decreased participant confidence to chance level at all continuous and only low-pressure pulsed waveforms (Fig. S6, paired t-tests from chance with Bonferroni corrected p > 0.05). In contrast, the Auditory Mondrian mask decreased participants' confidence to chance level at pulsed FUS at all tested pressures (Fig. 1 E and F, paired t-tests from chance with Bonferroni corrected p > 0.05). In conclusion, we identified a source of airborne auditory artifacts in FUS arising from the ultrasound transducer, with some contribution from the holding apparatus, generating tones within the auditory spectrum that are both dependent and independent of PRF. Human detection of the auditory artifact depended on the US parameters of I and F0 for the continuous waveform and PRF for the pulsed waveform. Finally, an Auditory Mondrian effectively masked these airborne auditory artifacts. This study has several limitations. First, we only focused on the auditory artifact's airborne component rather than the tissue-conduction component [[6]Braun V. Blackmore J. Cleveland R.O. Butler C.R. Transcranial ultrasound stimulation in humans is associated with an auditory confound that can be effectively masked.Brain Stimul. 2020; 13: 1527-1534Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar] of the auditory artifact. Nevertheless, this work will aid further researchers in isolating and excluding the airborne part. A second limitation is our use of online psychophysics experiments. While in-person would be ideal, modern online auditory psychophysics experiments have been proven to match in-person results [[13]Woods K.J.P. Siegel M.H. Traer J. McDermott J.H. Headphone screening to facilitate web-based auditory experiments.Atten Percept Psychophys. 2017; 79: 2064-2072Crossref PubMed Scopus (183) Google Scholar,[14]Kell A.J.E. Yamins D.L.K. Shook E.N. Norman-Haignere S.V. McDermott J.H. A task-optimized neural network replicates human auditory behavior, predicts brain responses, and reveals a cortical processing hierarchy.Neuron. 2018; 98: 630-644 e16Abstract Full Text Full Text PDF PubMed Scopus (221) Google Scholar]. This study introduces the Auditory Mondrian as a new masking design for acoustic psychophysics [[15]Gelfand S.A. Hearing : an introduction to psychological and physiological acoustics. sixth ed. CRC Press, Boca Raton2018: 1Google Scholar]. As a new design, further investigation and parameter adjustment with in-person direct FUS stimulation studies are needed to maximize its efficiency while minimizing unpleasant auditory perceptions. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. This work was supported by the National Institutes of Health (Grant Number: RF1MH117080), Canon Medical Systems, and the Japan Society For Promotion of Science (JSPS) (Grants-in-Aid for Scientific Research-Fostering Joint International Research(B), Grant Number 18KK0280). The following is the Supplementary data to this article. Download .docx (5.53 MB) Help with docx files Multimedia component 1/cms/asset/9a53a7a3-018d-4b96-9e39-35da15920fc7/mmc2.mp3Loading ... Download .mp3 (7.66 MB) Help with .mp3 files Multimedia component 2/cms/asset/bf45f5a9-c60c-4d5d-ab8e-82ce17e812ef/mmc3.mp3Loading ... Download .mp3 (3.37 MB) Help with .mp3 files Multimedia component 3/cms/asset/9356dd5c-dad4-453a-b161-dbbda3cfcfe2/mmc4.mp3Loading ... Download .mp3 (0.05 MB) Help with .mp3 files Multimedia component 4/cms/asset/ac2981b4-8f57-4687-8c85-07d7458e49d1/mmc5.mp3Loading ... Download .mp3 (0.05 MB) Help with .mp3 files Multimedia component 5/cms/asset/2cb0b213-8d96-48fb-b5b0-f62a5a737baa/mmc6.mp3Loading ... Download .mp3 (0.05 MB) Help with .mp3 files Multimedia component 6/cms/asset/6abdf3d1-1cfa-4880-865d-75a30fcb6f64/mmc7.mp3Loading ... Download .mp3 (0.05 MB) Help with .mp3 files Multimedia component 7/cms/asset/1d7ec937-5be5-4007-ae25-acb5e4be91c7/mmc8.mp3Loading ... Download .mp3 (0.05 MB) Help with .mp3 files Multimedia component 8/cms/asset/89c1fcae-00c4-484d-8158-86295297c979/mmc9.mp3Loading ... Download .mp3 (0.05 MB) Help with .mp3 files Multimedia component 9/cms/asset/893c6d24-a932-4094-8508-f253392ac70c/mmc10.mp3Loading ... Download .mp3 (0.05 MB) Help with .mp3 files Multimedia component 10/cms/asset/6627c623-24be-4d33-b018-99f58a87571d/mmc11.mp3Loading ... Download .mp3 (0.05 MB) Help with .mp3 files Multimedia component 11/cms/asset/26b784b4-2649-4286-afc2-e86f58f33c5f/mmc12.mp3Loading ... Download .mp3 (0.05 MB) Help with .mp3 files Multimedia component 12/cms/asset/0df79762-01a0-4306-9c8b-11dbfaffe8bc/mmc13.mp3Loading ... Download .mp3 (0.05 MB) Help with .mp3 files Multimedia component 13
Referência(s)