Advertisement

Safety of transcranial focused ultrasound stimulation: A systematic review of the state of knowledge from both human and animal studies

  • Cristina Pasquinelli
    Affiliations
    Danish Research Centre for Magnetic Resonance, Centre for Functional and Diagnostic Imaging and Research, Copenhagen University Hospital Hvidovre, Denmark

    Center for Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Kgs, Lyngby, Denmark
    Search for articles by this author
  • Lars G. Hanson
    Affiliations
    Danish Research Centre for Magnetic Resonance, Centre for Functional and Diagnostic Imaging and Research, Copenhagen University Hospital Hvidovre, Denmark

    Center for Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Kgs, Lyngby, Denmark
    Search for articles by this author
  • Hartwig R. Siebner
    Affiliations
    Danish Research Centre for Magnetic Resonance, Centre for Functional and Diagnostic Imaging and Research, Copenhagen University Hospital Hvidovre, Denmark

    Department of Neurology, Copenhagen University Hospital Bispebjerg, Copenhagen, Denmark

    Institute of Clinical Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
    Search for articles by this author
  • Hyunjoo J. Lee
    Affiliations
    School of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, South Korea
    Search for articles by this author
  • Axel Thielscher
    Correspondence
    Corresponding author. Danish Research Centre for Magnetic Resonance, Centre for Functional and Diagnostic Imaging and Research, Copenhagen University Hospital Hvidovre, Section 714, Kettegaard Allé 30, 2650, Hvidovre, Denmark.
    Affiliations
    Danish Research Centre for Magnetic Resonance, Centre for Functional and Diagnostic Imaging and Research, Copenhagen University Hospital Hvidovre, Denmark

    Center for Magnetic Resonance, Department of Health Technology, Technical University of Denmark, Kgs, Lyngby, Denmark
    Search for articles by this author
Open AccessPublished:July 31, 2019DOI:https://doi.org/10.1016/j.brs.2019.07.024

      Highlights

      • TFUS is an emerging non-invasive stimulation method with excellent focality
      • We summarize the safety-related data from the available literature.
      • Adverse effects were absent in 30 studies, and were reported in 3 studies.
      • Many studies used parameters outside safety limits for diagnostic ultrasound.
      • Further studies are warranted to establish the safety margin for TFUS.

      Abstract

      Background

      Low-intensity transcranial focused ultrasound stimulation (TFUS) holds great promise as a highly focal technique for transcranial stimulation even for deep brain areas. Yet, knowledge about the safety of this novel technique is still limited.

      Objective

      To systematically review safety related aspects of TFUS. The review covers the mechanisms-of-action by which TFUS may cause adverse effects and the available data on the possible occurrence of such effects in animal and human studies.

      Methods

      Initial screening used key term searches in PubMed and bioRxiv, and a review of the literature lists of relevant papers. We included only studies where safety assessment was performed, and this results in 33 studies, both in humans and animals.

      Results

      Adverse effects of TFUS were very rare. At high stimulation intensity and/or rate, TFUS may cause haemorrhage, cell death or damage, and unintentional blood-brain barrier (BBB) opening. TFUS may also unintentionally affect long-term neural activity and behaviour. A variety of methods was used mainly in rodents to evaluate these adverse effects, including tissue staining, magnetic resonance imaging, temperature measurements and monitoring of neural activity and behaviour. In 30 studies, adverse effects were absent, even though at least one Food and Drug Administration (FDA) safety index was frequently exceeded. Two studies reported microhaemorrhages after long or relatively intense stimulation above safety limits. Another study reported BBB opening and neuronal damage in a control condition, which intentionally and substantially exceeded the safety limits.

      Conclusion

      Most studies point towards a favourable safety profile of TFUS. Further investigations are warranted to establish a solid safety framework for the therapeutic window of TFUS to reliably avoid adverse effects while ensuring neural effectiveness. The comparability across studies should be improved by a more standardized reporting of TFUS parameters.

      Keywords

      Introduction

      Weak Transcranial Focused Ultrasound Stimulation (TFUS) aims to modulate neural activity by delivering a focused ultrasonic beam to a small target area in the brain. Currently, interest in TFUS is strongly increasing as it holds the promise of a far better spatial resolution than established non-invasive stimulation techniques and of the ability to reach deep brain areas [
      • Bystritsky A.
      • Korb A.S.
      • Douglas P.K.
      • Cohen M.S.
      • Melega W.P.
      • Mulgaonkar A.P.
      • et al.
      A review of low-intensity focused ultrasound pulsation.
      ]. This might open up intriguing new applications such as epilepsy treatment or pre-surgical diagnostics prior to electrode implantation for deep-brain stimulation [
      • Min B.K.
      • Bystritsky A.
      • Jung K.I.
      • Fischer K.
      • Zhang Y.
      • Maeng L.S.
      • Park S.I.
      • Chung Y.A.
      • Jolesz F.A.
      • Yoo S.S.
      Focused ultrasound-mediated suppression of chemically-induced acute epileptic EEG activity.
      ,
      • Hakimova H.
      • Kim S.
      • Chu K.
      • Lee S.K.
      • Jeong B.
      • Jeon D.
      Ultrasound stimulation inhibits recurrent seizures and improves behavioral outcome in an experimental model of mesial temporal lobe epilepsy.
      ]. TFUS is also attractive because it can be readily combined with neuroimaging modalities such as functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) without interfering with the recordings, as it applies acoustic waves rather than electric or magnetic fields.
      Firmly establishing its safety profile is a central requirement when aiming to move TFUS from initial pilot studies towards broader testing in humans in-vivo. Reviews on safety and bio-effects of ultrasound (US) in diagnostics [
      • Fowlkes J.B.
      American Institute of Ultrasound in Medicine consensus report on potential bioeffects of diagnostic ultrasound: executive summary.
      ] and therapy [
      • Miller D.L.
      • Smith N.B.
      • Bailey M.R.
      • Czarnota G.J.
      • Hynynen K.
      • Makin I.R.S.
      Overview of therapeutic ultrasound applications and safety considerations.
      ] as well as guidelines for the clearance of commercial diagnostic and therapeutic US systems as medical devices [
      • US FDA
      Guidance for Industry and FDA Staff - information for manufacturers seeking marketing clearance of diagnostic ultrasound systems and transducers.
      ] are available and constitute a benchmark to avoid harmful effects also for TFUS. Relating the TFUS parameters to these guidelines, as done in many of the published studies, might be considered a conservative choice. However, several aspects put TFUS in a special position. TFUS usually employs lower frequency compared to diagnostic ultrasound (usually upper kHz range vs. MHz) and longer pulse bursts. TFUS has a static focus so that the total energy delivered at the focal point can be higher than the maximal local energy deposit for diagnostic US, as the latter uses scanning approaches. The mechanism-of-action of TFUS is still poorly understood, rendering it more difficult to principally exclude harmful effects. In addition, current findings about the dose-response curve of TFUS [
      • King R.L.
      • Brown J.R.
      • Newsome W.T.
      • Pauly K.B.
      Effective parameters for ultrasound-induced in vivo neurostimulation.
      ] suggest that future therapeutic applications might aim to use intensities above the safety limits for diagnostic US in order to increase the robustness of the neural effects. Such a choice requires solid knowledge about the safety margin of TFUS. Along similar lines, accurate dose control for human TFUS is complicated by the presence of the skull, which strongly attenuates the beam. The attenuation depends on the individual skull thickness and composition [
      • Pichardo S.
      • Sin V.W.
      • Hynynen K.
      Multi-frequency characterization of the speed of sound and attenuation coefficient for longitudinal transmission of freshly excised human skulls.
      ], which are difficult to account for and lead to conservative intensity choices with an increased risk of underdosing. If the safety margin of TFUS is not well established, the use of more lenient dosing strategies to mitigate this problem is not feasible.
      There is a pressing need to establish specific safety guidelines for TFUS. Yet, the current knowledge about the risk-benefit ratio and the therapeutic window of TFUS is still rudimentary because TFUS is at an early stage of development. Indeed, no dedicated phase I safety human study has been performed so far, but the safety profile needs to be systematically investigated and monitored to ensure the patients’ safety. However, relevant information is already available today, because some of the published studies on TFUS in animals or humans included safety-relevant tests. Here, we systematically summarize these findings to give an overview of the current state of knowledge about TFUS safety. We start by describing the relevant physical parameters used to characterize the TFUS stimulus. We then shortly describe the known physical mechanisms by which ultrasound can cause tissue damage and we introduce the established safety indices, based on the beam parameters. Finally, we introduce the methods that have so far been applied to test for adverse effects of TFUS, and list the corresponding results. In the discussion, we summarize the implications of the available findings for in-vivo human TFUS applications.

      Material and methods

      Literature review on the safety of TFUS

      For this systematic review, we followed the PRISMA guidelines [
      • Liberati A.
      • Altman D.G.
      • Tetzlaff J.
      • Mulrow C.
      • Gøtzsche P.C.
      • Ioannidis J.P.
      • Clarke M.
      • Devereaux P.J.
      • Kleijnen J.
      • Moher D.
      The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration.
      ,
      • Moher D.
      • Liberati A.
      • Tetzlaff J.
      • Altman D.G.
      Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement.
      ]. Details on the implementation of the PRISMA requirements in our review are stated in the Supplementary material (Table S1). Our review was based on searches in PubMed (www.ncbi.nlm.nih.gov/pubmed) and bioRxiv (https://www.biorxiv.org/) for published and pre-published studies, using the keywords ‘tFUS’, ‘LIFUP’, ‘noninvasive brain stimulation focused ultrasound’, ‘neuromodulation brain transcranial ultrasound’, ‘focused ultrasound transcranial brain stimulation’ and ‘pulsed ultrasound brain stimulation’. The eligibility criteria were low intensity, low frequency TFUS in the brain of animals or humans with safety assessment, without use of microbubbles. Additional sources were reviews of the literature lists of relevant papers, and papers pointed out by the reviewers during the peer-review process. Fig. 1 shows details of the literature search. The last complete search was performed in January 2019 by one of the authors, and the last update was done in June 2019. From each paper, the sonication parameters and the methods used to assess safety and adverse effects were extracted as shown in Table 2 and Table 3 and categorized as described further below. Often, only some of the safety indices were reported. In that case, we give estimated values when possible.
      Fig. 1
      Fig. 1Selection process for the studies included in this review. The scheme is from Refs. [
      • Liberati A.
      • Altman D.G.
      • Tetzlaff J.
      • Mulrow C.
      • Gøtzsche P.C.
      • Ioannidis J.P.
      • Clarke M.
      • Devereaux P.J.
      • Kleijnen J.
      • Moher D.
      The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration.
      ,
      • Moher D.
      • Liberati A.
      • Tetzlaff J.
      • Altman D.G.
      Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement.
      ].
      Table 1Allowed limits for MI, TI, Ispta and Isppa according to the FDA guidelines for diagnostic ultrasound. The limit for TI also applies to TIC when bone is close by.
      Ispta (mW/cm2)Isppa (W/cm2)MITI
      7201901.96
      Table 2Overview of the parameters used in the reviewed studies. Ispta values very often exceeded the limits for diagnostic US. Cases where the Ispta values were higher than 3 W/cm2, corresponding to the limit for physiotherapeutic US, are highlighted in bold. Also one case in which MI exceeded the limit of 1.9 is marked in bold. If needed, we calculated missing parameters from the available data stated in the paper, which we indicate by “*” in the table. When the peak pressure was reported, MI was calculated using its definition (eq. (2)), and Isppa in water as indicated in Fig. 1 (ρ = 1000 kg/m3, c = 1500 m/s). Ispta was finally determined as Isppa x DC. 1) For [
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ], only the parameters employed in the safety tests of that study are listed here. 2) For [
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ], only the parameters for the main experiment are reported. 3) In Ref. [
      • Deffieux T.
      • Younan Y.
      • Wattiez N.
      • Tanter M.
      • Pouget P.
      • Aubry J.F.
      Low-intensity focused ultrasound modulates monkey visuomotor behavior.
      ], Ispta was determined by using ISI instead of PRP as the total pulse duration; this strongly reduces the value. 4) For [
      • Legon W.
      • Bansal P.
      • Tyshynsky R.
      • Ai L.
      • Mueller J.K.
      Transcranial focused ultrasound neuromodulation of the human primary motor cortex.
      ], MI, Isppa and Ispta are not stated, but authors asserted that they used the same waveform as [
      • Legon W.
      • Sato T.F.
      • Opitz A.
      • Mueller J.
      • Barbour A.
      • Williams A.
      • Tyler W.J.
      Transcranial focused ultrasound modulates the activity of primary somatosensory cortex in humans.
      ]. 5) Not clear if it is in water or after cranial transmission. 6) A spatial average of intensity of 25–30 W/cm2 is used. 6) Modulated focused ultrasound means that two transducer, one driven at 2.25 MHz and the other at 1.75 MHz, producing a difference frequency at 500 kHz at the focus, and a carrier frequency of 2 MHz.
      StudyTargetParametersObserved neural effect and adverse effect (if any)
      fc [kHz]TBDPRFSDNumber of sonica-tionsISIMIIsppaIspta
      Legon et al. preprint [
      • Legon W.
      • Bansal P.
      • Ai L.
      • Mueller J.K.
      • Meekins G.
      • Gillick B.
      Safety of transcranial focused ultrasound for human neuromodulation.
      ]
      Human thalamus or M1Follow-up questionnaire of 7 experiments, only 3 are published so far ([
      • Legon W.
      • Sato T.F.
      • Opitz A.
      • Mueller J.
      • Barbour A.
      • Williams A.
      • Tyler W.J.
      Transcranial focused ultrasound modulates the activity of primary somatosensory cortex in humans.
      ,
      • Ai L.
      • Bansal P.
      • Mueller J.K.
      • Legon W.
      Effects of transcranial focused ultrasound on human primary motor cortex using 7T fMRI.
      ] and one preprint [
      • Legon W.
      • Bansal P.
      • Tyshynsky R.
      • Ai L.
      • Mueller J.K.
      Transcranial focused ultrasound neuromodulation of the human primary motor cortex.
      ])
      This work presents results on safety assessment.
      Verhagen et al. [
      • Verhagen L.
      • Gallea C.
      • Folloni D.
      • Constans C.
      • Jensen D.E.
      • Ahnine H.
      • Roumazeilles L.
      • Santin M.
      • Ahmed B.
      • Lehericy S.
      • Klein-Flügge M.C.
      Offline impact of transcranial focused ultrasound on cortical activation in primates.
      ]
      Non-human primate SMA, FPC and pre-SMA25030 ms10 Hz40 s12.4 in water *

      1.68 after cranial transmission (estimated from pressure peak) *
      48 W/cm2 in water *

      23.52 W/cm2 after cranial tx *
      14.4 W/cm2 in water *

      7.056 W/cm2 after cranial tx *
      Reversible change in brain connectivity, that last up to 2 h after treatment.
      Fisher et al., 2018 [
      • Fisher J.A.
      • Gumenchuk I.
      Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
      ]
      Mice primary somatosensory cortex510500 μs1 kHz1 s10.24 in water *0.69 W/cm2 in water345 mW/cm2 in water *Early sensory-evoked cortical responses (3.0 ± 0.7 ms earlier) and alteration of Ca2+ responses.
      510continuous????280 W/cm2 in waterParameter tested as control.

      BBB intentionally opened. An increased number of astrocytes was found.
      Tufail et al., 2010 [
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ] 1)
      Mouse motor cortex5000.45 ms1.5 kHz67* (53) ms180*10 s0.13 after cranial tx211.72 mW/cm2 after cranial tx *142.2 mW/cm2 after cranial txNeuron's spike frequency and c-fos+ cell density increase and the activity of endogenous brain-derived neurotrophic factor (BDNF) were stimulated.

      Low frequency (250 KHz) and low intensities (up to around Ispta = 80 mW/cm2) result in more robust EMG response.

      The EMG failure probability increased with shorter ISI (200 ms), but decrease with multiple stimuli.

      BBB intentionally opened with the use of microbubbles.
      Kim et al., 2012 [
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ] 2)
      Rat abduncens nerve3500.36 ms1.5 kHz200 ms101 s0.9 after cranial tx (estimated)8.6 W/cm2 after cranial tx (estimated)4.6 W/cm2 after cranial tx (estimated)fc = 650 kHz and Isppa in the range 0.5–20 W/cm2 did not elicit eye movement in any animals. Movements observed when fc = 350 KHz for an Isppa of 8.6 W/cm2.
      Lee et al., 2015 [
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ]
      Sheep SM1 and V12501 ms500 Hz300 ms100 (groups of sonications repeated up to 8 times per animal)5 s (motor cortex)

      or 1 s (visual cortex)
      in the range 0.5–1.4 after cranial txUp to 11.8 W/cm2 after cranial tx –SM1

      Up to 14.3 W/cm2 after cranial tx -V1
      Up to 5.9 W/cm2 * after cranial tx –SM1

      Up to 7.15 W/cm2 * after cranial tx -V1
      MEP or VEPs were detected over a certain intensity threshold, which varied across sheep and was always above diagnostic limits, and in some cases also above the physiotherapy limit. In both cases, higher Isppa result in stronger response amplitude.

      Four animals which underwent 600 sonications at Isppa = 6.6–10.5 W/cm2 showed micro-hemorrhages in the primary visual cortex.
      Yoo et al., 2011 [
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ]
      Rabbit (after craniotomy), SM and visual area

      (the bottom line is only for temperature increase study)
      6900.05, 0.5, 10 and 50 ms10, 20, 100 and 1000 Hz0.5, 1, 1.5, 2, 9 s1<0.5 in water (for an Isppa = 3.3 W/cm2, resulting in clear BOLD activity)3.3, 6.4, 9,5, 12.6 W/cm2 in water1.6 W/cm2 in water (for Isppa = 3.3 W/cm2)The BOLD activation was observed at a much lower acoustic intensity (Isppa = 3.3 W/cm2, Ispta = 1.6 W/cm2) compared to the intensity that resulted in forepaw movement (Isppa = 12.6 W/cm2, Ispta = 6.3 W/cm2)
      6900.5 ms100 Hz27 s1?23 W/cm21.15 W/cm2Parameter tested as control for temperature increase
      Lee et al., 2015 [
      • Lee W.
      • Kim H.
      • Jung Y.
      • Song I.U.
      • Chung Y.A.
      • Yoo S.S.
      Image-guided transcranial focused ultrasound stimulates human primary somatosensory cortex.
      ]
      Human S12501 ms500 Hz*300 msAround 2003 s0.62 after cranial tx (maximal simulated value across N = 12 subjects)3 W/cm2 in water

      2.5 W/cm2 after cranial tx (maximal simulated value)
      1.5 W/cm2 in water

      1.25 W/cm2 * after cranial tx

      (maximal simulated value)
      Tactile sensations were not the same among subjects, but mostly at the hand area contralateral to the sonicated hemisphere. 1 out of 12 subjects did not report any sensation. Different peak amplitudes of EEG recording of SEP with and without stimulation.
      Lee et al., 2016 [
      • Lee W.
      • Chung Y.A.
      • Jung Y.
      • Song I.U.
      • Yoo S.S.
      Simultaneous acoustic stimulation of human primary and secondary somatosensory cortices using transcranial focused ultrasound.
      ]
      Human S1+S22101 ms500 Hz500 ms207 s?35 W/cm2 in water

      <8.8 W/cm2 after cranial tx (estimated)
      17.5 W/cm2 in water

      < 4.4 W/cm2 after cranial tx (estimated)
      Response rates of elicited sensations during the FUS procedures were different among subjects (68 ± 28% S1, 59 ± 22% S2, 61 ± 26% S1+S2, average ± sd across subjects).
      Kim et al., 2014 [
      • Kim H.
      • Chiu A.
      • Lee S.D.
      • Fischer K.
      • Yoo S.S.
      Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
      ]
      Rats somatomotor area350 and 6500.25, 0.5, 1, 2, 3 or 5 msin the range [0.06, 2.8] kHz

      and continous wave
      150, 200, 300 or 400 ms?2 or 3 s1.38

      (value for animal with signs of bleeding)
      22.4 W/cm2 after cranial tx

      (max value reported, corresponding to animal with signs of bleeding)
      11.2 W/cm2 after cranial tx

      (max value reported)
      Motor responses were observed at minimum threshold (Isppa = 4.9–5.6 W/cm2, Ispta = 2.5–2.8 W/cm2) in a limited range of sonication parameters (TBS = 1–5 ms, 50% of duty cycle, and SD = 300 ms, at fc = 350 kHz). Pulsed sonication elicited motor responses at lower acoustic intensities than its equivalent continuous sonication (Isppa = 7.73 W/cm2).

      One animal which underwent a sonication of Ispta = 11.2 W/cm2 for a short period of time (<9 s using 1 ms TBD, 50% duty cycle and 300 ms SD) showed signs of local bleeding.
      Kim et al., 2013 [
      • Kim H.
      • Park M.
      • Wang S.
      • Chiu A.
      • Fischer K.
      • Yoo S.S.
      PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
      ]
      Rats3500.5 ms1 kHz300 ms1200 *2 s0.74 after cranial tx6 W/cm2 * after cranial tx3 W/cm2 after cranial txChanges in glucose metabolism for up to more than 1 h after sonication.
      Lee et al., 2016 [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ]
      Human V12701 ms500 Hz300 ms5013 s (fMRI) or 2.5 s (EEG)2.8 * in water

      1.2 after cranial tx (maximal simulated value across N = 19 subjects)
      16.6 W/cm2 in water

      11.6 W/cm2 after cranial tx (maximal simulated value)
      8.3 W/cm2 in water *

      5.8 W/cm2 * after cranial tx (maximal simulated value)
      fMRI: 11 out of 19 participants reported the perception of phosphenes, and a clear fMRI response.

      EEG: 10/10 subjects reported phosphene sensation. Changes in VEP EEG peak.
      Yoo et al., 2011 [
      • Yoo S.S.
      • Kim H.
      • Min B.K.
      • Franck S.P.E.
      Transcranial focused ultrasound to the thalamus alters anesthesia time in rats.
      ]
      Rats thalamus6500.5 ms100 Hz20 min10.61 after cranial tx6 W/cm2 after cranial tx300 mW/cm2 after cranial txThe sonication reduced the time to emergence of voluntary movement from intraperitoneal ketamine-xylazine anesthesia. A preliminary test showed that a Isppa = 3.3 W/cm2 failed to decrease the duration of the anesthetic state.
      Deffieux et al., 2013 [
      • Deffieux T.
      • Younan Y.
      • Wattiez N.
      • Tanter M.
      • Pouget P.
      • Aubry J.F.
      Low-intensity focused ultrasound modulates monkey visuomotor behavior.
      ]
      Monkey frontal eye field3201 ms1 kHz100 ms40≥30 s1.06 in water *

      0.6 after cranial tx (average across several skull positions)
      12 W/cm2 in water*

      4 W/cm2 after cranial tx
      6 W/cm2 in water*

      13.5 mW/cm2 after cranial tx 3)

      2 W/cm2 after cranial tx (using standard formula) *
      Ultrasound increased antisaccade latencies in two monkeys.
      Mueller et al., 2014 [
      • Mueller J.
      • Legon W.
      • Opitz A.
      • Sato T.F.
      • Tyler W.J.
      Transcranial focused ultrasound modulates intrinsic and evoked EEG dynamics.
      ]
      Human somatosensory cortex5000.36 ms1 kHz500 ms1206 s1.13 in water23.9 W/cm2 in water8.6 W/cm2 in water*The phase distribution of beta frequencies was altered, together with a change in phase rate of beta and gamma frequencies.
      Legon et al., 2014 [
      • Legon W.
      • Sato T.F.
      • Opitz A.
      • Mueller J.
      • Barbour A.
      • Williams A.
      • Tyler W.J.
      Transcranial focused ultrasound modulates the activity of primary somatosensory cortex in humans.
      ]
      Human S15000.36 ms1 kHz500 ms??1.13 in water23.9 W/cm2 in water8.6 W/cm2 in water*Amplitudes of SEPs (recorded by EEG) elicited by median nerve stimulation were significantly attenuated. The spectral content of sensory-evoked brain oscillations were significantly modulated by tFUS.
      Legon et al. preprint [
      • Legon W.
      • Bansal P.
      • Tyshynsky R.
      • Ai L.
      • Mueller J.K.
      Transcranial focused ultrasound neuromodulation of the human primary motor cortex.
      ]
      Human M15000.36 ms1 kHz500 ms14)4)4)The amplitude of single-pulse TMS MEPs was decreased; the intracortical facilitation was attenuated; no effect on intracortical inhibition. Ultrasound reduces reaction time on a simple stimulus response task
      Lee et al., 2018 [
      • Lee W.
      • Croce P.
      • Margolin R.W.
      • Cammalleri A.
      • Yoon K.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
      ]
      Rats (anesthetized and awake) motor cortex6001 ms500 Hz300 ms105–10 s1.38Minimum value: 2.1 W/cm2; incremented by 1 W/cm2; maximum value: 14.9 W/cm2

      5)
      7.5 W/cm2

      5)
      Different thresholds to evoke observed motor response: Isppa = 3.4 ± 1.8 W/cm2 for the awake condition (grand mean response rate 76.2%)

      Isppa = 10.2 ± 2.4 W/cm2 (grand mean response rate 68.6%) or 12.4 ± 2.8 W/cm2 (grand mean response rate 38.6%) for 2 different types of anesthetics5)
      Yoo et al., 2017 [
      • Yoo S.S.
      • Yoon K.
      • Croce P.
      • Cammalleri A.
      • Margolin R.W.
      • Lee W.
      Focused ultrasound brain stimulation to anesthetized rats induces long-term changes in somatosensory evoked potentials.
      ]
      Rats somatosensory cortex6500.5 ms100 Hz10 min1?4.2 W/cm2

      5)
      210 mW/cm2

      5)
      Different SEP features compared to controls were evident and persisted beyond 35 min after the administration of FUS.
      Yang et al., 2018 [
      • Yang P.F.
      • Phipps M.A.
      • Newton A.T.
      • Chaplin V.
      • Gore J.C.
      • Caskey C.F.
      • Chen L.M.
      Neuromodulation of sensory networks in monkey brain by focused ultrasound with MRI guidance and detection.
      ]
      Monkey S12500.252 ms2 kHz300 ms10 *3 s1.87 in water

      1.08 after cranial tx (estimated after measurement on skull attenuation)
      29.5 W/cm2 in water

      9.9 W/cm2 after cranial tx
      1.34 W/cm2 in water

      0.452 W/cm2 after cranial tx
      Excitation effects with BOLD fMRI not only at the target but also off-target somatosensory and associated brain regions as a cause of modulation

      in downstream brain regions.
      Daniels et al., 2018 [
      • Daniels D.
      • Sharabi S.
      • Last D.
      • Guez D.
      • Salomon S.
      • Zivli Z.
      • Castel D.
      • Volovick A.
      • Grinfeld J.
      • Rachmilevich I.
      • Amar T.
      • Liraz-Zaltsman S.
      • Sargsyan N.
      • Yael Mardor Y.
      • Harnof S.
      Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
      ]
      Pigs (after craniotomy) auditory and rats inferior colliculus230 (1000 element transducer)100 ms0.333 Hz *52 s1Rats: 0.08 *

      Pigs and rats: 0.17 *

      5)
      Rats: 2.3 W/cm2

      Pigs and rats: 4.6 W/cm2

      5)
      Rats: 765.9 mW/cm2 *

      Pigs and rats: 1.53 W/cm2 *

      5)
      AEP decreases by 59.8 ± 3.3% (with Isppa = 2.3 W/cm2) and by 36.9 ± 7.5% (with Isppa = 4.6 W/cm2) of the baseline value in rats.

      AEP amplitudes decreased to an average of 27.7 ± 5.9% of baseline in pigs.

      This effect lasted between 30 min and 1 month in most treated animals.
      Kim et al., 2018 [
      • Kim H.
      • Kim S.
      • Sim N.S.
      • Pasquinelli C.
      • Thielscher A.
      • Lee J.H.
      • Lee H.J.
      Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
      ]
      Mice motor cortex183 (CMUT 32 elements array)4.5 ms200 Hz200 ms25Around 9.6 s0.12 *

      before cranial tx
      Up to 61.5 mW/cm2 before cranial txUp to 55.4 mW/cm2

      before cranial tx
      At an intensity of Ispta = 34.1 mW/cm2, the average stimulation success rate of four mice was over 70%.
      Dallapiazza et al., 2018 [
      • Dallapiazza R.F.
      • Timbie K.F.
      • Holmberg S.
      • Gatesman J.
      • Lopes M.B.
      • Price R.J.
      • Miller G.W.
      • Elias W.J.
      Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound.
      ]
      Swine thalamic regions1.145 MHz (single element)

      650 and 220 kHz (multi element phased array transducer)
      43.7 ms10 Hz40 s10.53

      5)
      ?

      6)
      ?Suppression of SSEP amplitude
      Kim et al., 2015 [
      • Kim H.
      • Park M.Y.
      • Lee S.D.
      • Lee W.
      • Chiu A.
      • Yoo S.S.
      Suppression of EEG visual-evoked potentials in rats via neuromodulatory focused ultrasound.
      ]
      Rats visual cortex3500.5 ms20, 100, 166 Hz150 s1Max 0.75

      5)
      1, 3, 5 W/cm2

      5)
      Max 250 mW/cm2

      5)
      Isppa = 1 W/cm2, TBD = 0.5 at PRF = 100 Hz and Isppa = 3W/cm2, TBD = 0.5 ms, PRF = 20 Hz, corresponding to 50 and 30 mW/cm2 Ispta did not change VEP.

      Isppa = 3 W/cm2 with TBD = 0.5 ms and PRF = 100 Hz (5% duty cycle) successfully suppressed the VEP.

      Higher duty cycle (8.3%) increased the VEP. The same effect was observed at Isppa = 5 W/cm2 and 5% duty cycle.
      Min et al., 2011 [
      • Min B.K.
      • Bystritsky A.
      • Jung K.I.
      • Fischer K.
      • Zhang Y.
      • Maeng L.S.
      • Park S.I.
      • Chung Y.A.
      • Jolesz F.A.
      • Yoo S.S.
      Focused ultrasound-mediated suppression of chemically-induced acute epileptic EEG activity.
      ]
      Rats thalamus6900.5 ms100 Hz180 s10.33 after cranial transmission2.6 W/cm2

      after cranial transmission
      130 mW/cm2

      after cranial transmission
      Suppression of the number of epileptic signal bursts. Average among all 9 rats that underwent treatment.
      Baek et al., 2018 [
      • Baek H.
      • Pahk K.J.
      • Kim M.J.
      • Youn I.
      • Kim H.
      Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
      ]
      Mice lateral cerebellar nucleus (LCN)3500.5 ms1 kHz300 ms6002 s0.54 in water2.5 W/cm2 in water1.25 W/cm2 in waterEnhancement of sensorimotor recovery after stroke.

      Decreased level of brain edema and tissue swelling in the affected hemisphere 3 days after the stroke.
      Folloni et al. [
      • Folloni D.
      • Verhagen L.
      • Mars R.B.
      • Fouragnan E.
      • Constans C.
      • Aubry J.F.
      • Rushworth M.F.
      • Sallet J.
      Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation.
      ]
      Monkey amygdala and anterior cingulate cortex (ACC)25030 ms10 Hz40 s1Maximum 2.64 in amygdala and 1.64 in ACC * (from estimation after cranial transmission)Maximum 51 W/cm2 in amygdala and 17 W/cm2 in ACC (estimation after cranial transmission)Maximum 15.3 W/cm2 in amygdala and 5.3 W/cm2 in ACC (estimation after cranial transmission)After TFUS, the functional coupling of the stimulated areas, but not of control areas, was selectively reduced. This effect was measured by fMRI and lasted for more than 1 h after stimulation.
      Li et al., 2016 [
      • Li G.F.
      • Zhao H.X.
      • Zhou H.
      • Yan F.
      • Wang J.Y.
      • Xu C.X.
      • Wang C.Z.
      • Niu L.L.
      • Meng L.
      • Wu S.
      • Zhang H.L.
      Improved anatomical specificity of non-invasive neuro-stimulation by high frequency (5 MHz) ultrasound.
      ]
      Mice motor cortex1 MHz (and high frequency, 5 MHz)0.5 ms1 kHz300 ms20 *

      600 * (for safety assessment)
      3 s?260–460 mW/cm2 after cranial transmission *From 130 to 230 mW/cm2 after cranial transmissionThe peak EEG amplitude increased with increasing Ispta.
      Yang et al., 2012 [
      • Yang P.S.
      • Kim H.
      • Lee W.
      • Bohlke M.
      • Park S.
      • Maher T.J.
      • Yoo S.S.
      Transcranial focused ultrasound to the thalamus is associated with reduced extracellular GABA levels in rats.
      ]
      Rats thalamus6500.5 ms100 Hz20 min10.23.5 W/cm2 after cranial transmission175 mW/cm2 after cranial transmissionExtracellular GABA level started to decrease upon sonication and remained reduced compared to control group up to 100 min after the end of sonication. The same effect was not observed for the extracellular glutamate level.
      Han et al., 2017 [
      • Han S.
      • Kim M.
      • Kim H.
      • Shin H.
      • Youn I.
      Ketamine inhibits ultrasound stimulation-induced neuromodulation by blocking cortical neuron activity.
      ]
      Mice motor cortex3500.23 ms1.5 kHz66.67 ms600 * (for safety assessment)2 s0.1–1.16 *3.38–39.5 W/cm2 *

      10 W/cm2 * (for safety assessment)

      -all after cranial transmission-
      1.16–13.55 W/cm2

      3.46 W/cm2 (for safety assessment)

      -all after cranial transmission-
      The robustness of the visual observed responses increased and the latency of the response decreased with increasing Ispta. Ispta = 3.46 w/cm2 was sufficient to induce strong motor response; no response was observed for Ispta<1.16 W/cm2. Ultrasound-induced motor responses were inhibited more than 20 min after ketamine injection. This was confirmed in in vitro cortical neuron sample by fluorescence calcium imaging, showing a dose-dependent effect.
      Gulick et al., 2017 [
      • Gulick D.W.
      • Li T.
      • Kleim J.A.
      • Towe B.C.
      Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
      ]
      Rat motor cortex (after craniotomy)2000.5 ms1 kHz300 ms or 3 ms?2 s or 10 sMax 3.19 W/cm2 * or

      30 W/cm2 *
      4.5 W/cm2 or 9 mW/cm2US directly evokes hindlimb movement, even at short burst (3 ms) and had short latency (10 ms) and long refractory (3 s) periods. US modulation significantly suppressed forelimb and hindlimb responses following ECS for several minutes after the stimulation, but shows no short-term effect.
      Younan et al., 2012 [
      • Younan Y.
      • Deffieux T.
      • Larrat B.
      • Fink M.
      • Tanter M.
      • Aubry J.F.
      Influence of the pressure field distribution in transcranial ultrasonic neurostimulation.
      ]
      Rat cortex (target to elicit motor response, not corresponding to motor cortex)3200.23 ms2 kHz250 ms?10 sFrom 0.7 to 1.77 *Isppa of 7.5 W/cm2 (to have 50% response) in water. Via computer stimulation, it corresponds to 17.5 W/cm2 after cranial transmission due to reverberation3.75 W/cm2 in water and 8.75 W/cm2 after cranial transmission * (to have 50% response)A pressure threshold of 0.79 and 0.59 MPa was required to reach 50% of responsiveness, for deep or light anesthesia stage, respectively, and the sigmoid respond was less sharp in the light anesthesia stage. These pressures corresponded to an average Isppa of 7.5 W/cm2.
      Mehić et al., 2014 [
      • Mehić E.
      • Xu J.M.
      • Caler C.J.
      • Coulson N.K.
      • Moritz C.T.
      • Mourad P.D.
      Increased anatomical specificity of neuromodulation via modulated focused ultrasound.
      ]
      Different locations in mice cortex500 (from unfocused ultrasound or modulated focused ultrasound 6), mFUS)0.2 ms1.5 kHz10 s1?0.45–16 W/cm2 for unfocused US *

      3–33 W/cm2 for mFUS *
      0.155.25 W/cm2 for unfocused US

      1–10 W/cm2 for mFUS
      Increasing the Ispta increase the motor movement robustness, assessed by visual assessment with unfocused US and mFUS, and the normalized success rate in mFUS.
      Table 3Overview of the safety assessments included in the reviewed studies. The involved methods are: fluorescein isothiocyanate-dextran (FITC–Dextran), trypan blue dye (T.b.), Evans blue dye (E.B.), magnetic resonance contrast agent (MR c.a.) to assess the BBB opening; antibodies to Caspase-3, quantitative transmission electron microscopy (e.m.), hematoxylin and eosin (H&E), terminal deoxynucleotidyl transferase dUTP Nick-End Labeling (TUNEL) assay, cresyl violet (c.v.), GFAP (glial fibrillary acidic protein), VAF (Vanadium acid fuchsin) and luxol fast blue dye (LFB) to monitor cell death, damage, brain ultrastructure and hemorrhage; Sensors (thermo-couple [
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ,
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ,
      • Kim H.
      • Kim S.
      • Sim N.S.
      • Pasquinelli C.
      • Thielscher A.
      • Lee J.H.
      • Lee H.J.
      Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
      ,
      • Gulick D.W.
      • Li T.
      • Kleim J.A.
      • Towe B.C.
      Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
      ,
      • Yoo S.S.
      • Yoon K.
      • Croce P.
      • Cammalleri A.
      • Margolin R.W.
      • Lee W.
      Focused ultrasound brain stimulation to anesthetized rats induces long-term changes in somatosensory evoked potentials.
      ] or optical fiber based thermal sensor [
      • Li G.F.
      • Zhao H.X.
      • Zhou H.
      • Yan F.
      • Wang J.Y.
      • Xu C.X.
      • Wang C.Z.
      • Niu L.L.
      • Meng L.
      • Wu S.
      • Zhang H.L.
      Improved anatomical specificity of non-invasive neuro-stimulation by high frequency (5 MHz) ultrasound.
      ]), maximum temperature increase, (equation (7) to estimate ΔTmax), magnetic resonance thermometry (MR th.) and the bioheat equation for the temperature increase; motor task (m.t.) or other for behavioral assessments. 1) safety assessed in rats that did not undergo pentylenetetrazol (PTZ) injection to induce epileptic activity [
      • Min B.K.
      • Bystritsky A.
      • Jung K.I.
      • Fischer K.
      • Zhang Y.
      • Maeng L.S.
      • Park S.I.
      • Chung Y.A.
      • Jolesz F.A.
      • Yoo S.S.
      Focused ultrasound-mediated suppression of chemically-induced acute epileptic EEG activity.
      ] or photothrombosis procedure to induce ischemic stroke [
      • Baek H.
      • Pahk K.J.
      • Kim M.J.
      • Youn I.
      • Kim H.
      Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
      ].
      TargetBBB integrityCell death, damage, brain ultrastructure and hemorrhageThermal effectBehaviour
      FITC–DextranT.b.E.B.MR c.a.Caspase-3E.m.H&ETUNELC.v.GFAPVAFMRILFBSensorsΔTmaxMR th.Bioheat eq.M.t.Other
      Legon et al. preprint [
      • Legon W.
      • Bansal P.
      • Ai L.
      • Mueller J.K.
      • Meekins G.
      • Gillick B.
      Safety of transcranial focused ultrasound for human neuromodulation.
      ]
      Human thalamus or M1X
      Verhagen et al. [
      • Verhagen L.
      • Gallea C.
      • Folloni D.
      • Constans C.
      • Jensen D.E.
      • Ahnine H.
      • Roumazeilles L.
      • Santin M.
      • Ahmed B.
      • Lehericy S.
      • Klein-Flügge M.C.
      Offline impact of transcranial focused ultrasound on cortical activation in primates.
      ]
      Non-human primate SMA, FPC and pre-SMAXXX
      Fisher et al., 2018 [
      • Fisher J.A.
      • Gumenchuk I.
      Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
      ]
      Mice primary somatosensory cortexXX
      Tufail et al., 2010 [
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ]
      Mouse motor cortexXXXXXX
      Kim et al., 2012 [
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ]
      Rat abduncens nerveXXXX
      Lee et al., 2015 [
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ]
      Sheep SM1 and V1XXX
      Yoo et al., 2011 [
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ]
      Rabbit (after craniotomy), SM and visual areaXXXXXX
      Lee et al., 2015 [
      • Lee W.
      • Kim H.
      • Jung Y.
      • Song I.U.
      • Chung Y.A.
      • Yoo S.S.
      Image-guided transcranial focused ultrasound stimulates human primary somatosensory cortex.
      ]
      Human S1XX
      Lee et al., 2016 [
      • Lee W.
      • Chung Y.A.
      • Jung Y.
      • Song I.U.
      • Yoo S.S.
      Simultaneous acoustic stimulation of human primary and secondary somatosensory cortices using transcranial focused ultrasound.
      ]
      Human S1+S2X
      Kim et al., 2014 [
      • Kim H.
      • Chiu A.
      • Lee S.D.
      • Fischer K.
      • Yoo S.S.
      Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
      ]
      Rats somatomotor areaXX
      Kim et al., 2013 [
      • Kim H.
      • Park M.
      • Wang S.
      • Chiu A.
      • Fischer K.
      • Yoo S.S.
      PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
      ]
      RatsXX
      Lee et al., 2016 [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ]
      Human V1XX
      Lee et al., 2018 [
      • Lee W.
      • Croce P.
      • Margolin R.W.
      • Cammalleri A.
      • Yoon K.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
      ]
      Rats (anesthetized and awake) motor cortexXXXXXXX
      Yoo et al., 2017 [
      • Yoo S.S.
      • Yoon K.
      • Croce P.
      • Cammalleri A.
      • Margolin R.W.
      • Lee W.
      Focused ultrasound brain stimulation to anesthetized rats induces long-term changes in somatosensory evoked potentials.
      ]
      Rats somatosensory cortexXX
      Yang et al., 2018 [
      • Yang P.F.
      • Phipps M.A.
      • Newton A.T.
      • Chaplin V.
      • Gore J.C.
      • Caskey C.F.
      • Chen L.M.
      Neuromodulation of sensory networks in monkey brain by focused ultrasound with MRI guidance and detection.
      ]
      Monkey S1X
      Daniels et al., 2018 [
      • Daniels D.
      • Sharabi S.
      • Last D.
      • Guez D.
      • Salomon S.
      • Zivli Z.
      • Castel D.
      • Volovick A.
      • Grinfeld J.
      • Rachmilevich I.
      • Amar T.
      • Liraz-Zaltsman S.
      • Sargsyan N.
      • Yael Mardor Y.
      • Harnof S.
      Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
      ]
      Pigs (after craniotomy) auditory and rats inferior colliculusXXXX
      Kim et al., 2018 [
      • Kim H.
      • Kim S.
      • Sim N.S.
      • Pasquinelli C.
      • Thielscher A.
      • Lee J.H.
      • Lee H.J.
      Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
      ]
      MiceXXX
      Dallapiazza et al. [
      • Dallapiazza R.F.
      • Timbie K.F.
      • Holmberg S.
      • Gatesman J.
      • Lopes M.B.
      • Price R.J.
      • Miller G.W.
      • Elias W.J.
      Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound.
      ]
      Swine thalamic regionsXXXX
      Min et al., 2011 [
      • Min B.K.
      • Bystritsky A.
      • Jung K.I.
      • Fischer K.
      • Zhang Y.
      • Maeng L.S.
      • Park S.I.
      • Chung Y.A.
      • Jolesz F.A.
      • Yoo S.S.
      Focused ultrasound-mediated suppression of chemically-induced acute epileptic EEG activity.
      ]
      Rats thalamus 1)XX
      Baek et al., 2018 [
      • Baek H.
      • Pahk K.J.
      • Kim M.J.
      • Youn I.
      • Kim H.
      Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
      ]
      Mice lateral cerebellar nucleus (LCN)X 1)XX
      Li et al., 2016 [
      • Li G.F.
      • Zhao H.X.
      • Zhou H.
      • Yan F.
      • Wang J.Y.
      • Xu C.X.
      • Wang C.Z.
      • Niu L.L.
      • Meng L.
      • Wu S.
      • Zhang H.L.
      Improved anatomical specificity of non-invasive neuro-stimulation by high frequency (5 MHz) ultrasound.
      ]
      Mice motor cortexXX
      Yang et al., 2012 [
      • Yang P.S.
      • Kim H.
      • Lee W.
      • Bohlke M.
      • Park S.
      • Maher T.J.
      • Yoo S.S.
      Transcranial focused ultrasound to the thalamus is associated with reduced extracellular GABA levels in rats.
      ]
      Rats thalamusX
      Han et al., 2017 [
      • Han S.
      • Kim M.
      • Kim H.
      • Shin H.
      • Youn I.
      Ketamine inhibits ultrasound stimulation-induced neuromodulation by blocking cortical neuron activity.
      ]
      Mice motor cortexX
      Gulick et al., 2017 [
      • Gulick D.W.
      • Li T.
      • Kleim J.A.
      • Towe B.C.
      Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
      ]
      Rat motor cortex (after craniotomy)XX
      Younan et al., 2012 [
      • Younan Y.
      • Deffieux T.
      • Larrat B.
      • Fink M.
      • Tanter M.
      • Aubry J.F.
      Influence of the pressure field distribution in transcranial ultrasonic neurostimulation.
      ]
      Rat cortex (target to elicit motor response, not corresponding to motor cortex)X
      Mehić et al., 2014 [
      • Mehić E.
      • Xu J.M.
      • Caler C.J.
      • Coulson N.K.
      • Moritz C.T.
      • Mourad P.D.
      Increased anatomical specificity of neuromodulation via modulated focused ultrasound.
      ]
      Different locations in mice cortexX

      Mechanism of ultrasound neuromodulation

      Despite many hypotheses, the exact underlying mechanism of neuromodulation using low-intensity ultrasound is yet to be understood [
      • Fomenko A.
      • Neudorfer C.
      • Dallapiazza R.F.
      • Kalia S.K.
      • Lozano A.M.
      Low-intensity ultrasound neuromodulation: an overview of mechanisms and emerging human applications.
      ]. The initial hypotheses for the ultrasound neuromodulation were thermal effects and acoustic cavitation. While an increase in the tissue temperature could perturb neuronal activity levels, the temperature increase due to low-intensity ultrasound is often less than 0.1 °C. Thus, the thermal effects of low-intensity ultrasound are most likely negligible. The second hypothesis is based on acoustic cavitation. This hypothesis postulates that the ultrasound generates nanobubbles in the lipophilic zone of the plasma membrane, which then vibrates according to the pressure variations, alters the local curvature of the bilayer, and changes overall neuronal excitability [
      • Sassaroli E.
      • Vykhodtseva N.
      Acoustic neuromodulation from a basic science prospective.
      ]. However, since nanobubbles are formed at an intensity larger than 100 mW/cm2, generation of micro or nanobubbles at the intensity used in standard neuromodulation protocols must be confirmed. The recent hypotheses now focus more on the effects of acoustic radiation forces on the permeability of the ion channels, such as mechanosensitive channels [
      • Brohawn S.G.
      • Su Z.
      • MacKinnon R.
      Mechanosensitivity is mediated directly by the lipid membrane in TRAAK and TREK1 K+ channels.
      ] and voltage-gated calcium, sodium, and potassium channels [
      • Tyler W.J.
      Noninvasive neuromodulation with ultrasound? A continuum mechanics hypothesis.
      ]. Another kind of hypotheses includes plasma deformation, which postulates that vibration of surrounding extra- and intracellular environment evokes mechanical changes in either the plasma membrane tension or the lipid bilayer and modulates neuronal activities [
      • Tyler W.J.
      Noninvasive neuromodulation with ultrasound? A continuum mechanics hypothesis.
      ].
      Contrary to these works on the mechanisms involved with direct modulation of ion channels and membranes, an indirect in vivo ultrasound neuromodulation through auditory or cochlear pathways has been also recently proposed [
      • Sato T.
      • Shapiro M.G.
      • Tsao D.Y.
      Ultrasonic neuromodulation causes widespread cortical activation via an indirect auditory mechanism.
      ,
      • Guo H.
      • Hamilton II, M.
      • Offutt S.J.
      • Gloeckner C.D.
      • Li T.
      • Kim Y.
      • Legon W.
      • Alford J.K.
      • Lim H.H.
      Ultrasound produces extensive brain activation via a cochlear pathway.
      ]. These studies demonstrated that ultrasound-induced activities were eliminated or reduced upon transection of the auditory nerves or removal of cochlear fluids. These results raised an important question of whether direct activation of neurons in the intact brain is possible. While more in-depth studies on the experimental protocols such as sharpness of the pulse, pulse repetition frequency, and bone transduction must be performed, these studies underscore the need for a solid understanding of the underlying mechanism of ultrasound neuromodulation [
      • Guo H.
      • Hamilton II, M.
      • Offutt S.J.
      • Gloeckner C.D.
      • Li T.
      • Kim Y.
      • Legon W.
      • Alford J.K.
      • Lim H.H.
      Ultrasound produces extensive brain activation via a cochlear pathway.
      ].

      Physical parameters and safety indices of US waves

      A sketch of an experimental setup for TFUS is shown in Fig. 2A, using the stimulation of a rat as example. The main indices used to assess safety are:
      • Ispta (spatial peak temporal average intensity) is the temporal average intensity, calculated at the position of the spatial maximum
      • Isppa (spatial peak pulse average intensity) is the pulse average intensity, calculated at the position of the spatial maximum
      • MI (mechanical index) gives an estimation of the likelihood of inertial cavitation
      • TI (thermal index) is the steady-state temperature increase in soft tissue during ultrasound sonication
      • TIC (thermal index for cranial bone) is a modification of TI, when the skull is close to the transducer face
      Fig. 2
      Fig. 2Overview of the TFUS setup and parameters. A) The ultrasound pressure wave is generated by a transducer and delivered to the target through a guide filled with acoustic gel. B) The pressure stimulus over time is shown to indicate the main parameters. C) The main intensity values are shown for a fixed space position, together with their relationship with the pressure signal.
      Ispta, TI and TIC are related to the risk of thermal bio-effects, while Isppa and MI are related to the risk of cavitation. The upper limits for these five indices allowed for diagnostic ultrasound are shown in Table 1. It should be noted that another guideline, IEC standard 60601-2-5 for physiotherapy US equipment, sets an upper limit for the “effective intensity’’, defined as the ratio of acoustic output power to effective radiating area, of 3 W/cm2. The standard also states that this value should only be reached for short times to prevent substantial heating. The “effective intensity’’ of 3 W/cm2 is usually interpreted as the upper limit for Ispta [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ,
      • Szabo T.L.
      Diagnostic ultrasound imaging:inside out.
      ,
      • Baek H.
      • Pahk K.J.
      • Kim M.J.
      • Youn I.
      • Kim H.
      Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
      ]. Lee and colleagues [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ] compare the intensities used in their study against this limit rather than using the FDA guidelines for diagnostic US. Complementary to TI, the temperature increase at the target can be calculated as ΔTmax (equation 7 and 8 in Supplementary material) or through the bio-heat equation [
      • Pennes H.H.
      Analysis of tissue and arterial blood temperatures in the resting human forearm.
      ,
      • Dallapiazza R.F.
      • Timbie K.F.
      • Holmberg S.
      • Gatesman J.
      • Lopes M.B.
      • Price R.J.
      • Miller G.W.
      • Elias W.J.
      Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound.
      ,
      • Verhagen L.
      • Gallea C.
      • Folloni D.
      • Constans C.
      • Jensen D.E.
      • Ahnine H.
      • Roumazeilles L.
      • Santin M.
      • Ahmed B.
      • Lehericy S.
      • Klein-Flügge M.C.
      Offline impact of transcranial focused ultrasound on cortical activation in primates.
      ]. A more detailed explanation of these indices and formulae can be found in the Supplementary Material.

      Mechanisms underlying tissue damage by US

      Ultrasound waves may cause harmful effects on tissues via two physical mechanisms, mechanical and thermal. The main mechanical effect is cavitation, in which vapor cavities (or “bubbles”) form in the soft tissues during the periods of low pressure (i.e. the minima) of the acoustic wave cycles. Depending on intensity and center frequency, this can result in a stable oscillation (stable or non-inertial cavitation) or can result in violent bubble collapses (inertial cavitation) that create large forces in their neighborhood. The air bubbles can have an endogenous origin (for example in the lungs or intestine), or they can be created by the mechanical wave itself, if the peak rarefaction pressure (i.e. the pressure during the minima) is small enough to allow the liquid to reach vaporization. Alternatively, ultrasound contrast agents (UCA), which contain microbubbles, can be injected for, e.g. clinical purposes [
      • Cosgrove D.
      Ultrasound contrast agents: an overview.
      ] or gene and drug delivery [
      • Ferrara K.
      • Pollard R.
      • Borden M.
      Ultrasound microbubble contrast agents: fundamentals and application to gene and drug delivery.
      ].
      When a mechanical wave propagates linearly in a medium, its amplitude decreases exponentially starting from the source. The attenuation is caused by both scattering, i.e. the change in the direction of wave propagation due to the presence of microscopic obstacles along the beam, and absorption. Absorption is the process by which the wave energy is converted into heat, and therefore the medium is heated. Several ways to model or monitor the resulting temperature increase in the medium exist, and they will be further discussed below.

      Types of adverse and side effects caused by TFUS

      In this section, we summarize the potential adverse and side effects, which have so far been tested in TFUS studies, and briefly outline the employed techniques to assess the occurrence of these effects. The majority of results were obtained in animal studies, which tested for the following effects:
      • Blood-brain barrier (BBB) opening: The BBB is a semi-permeable membrane formed by endothelial cells which separates the vessels and the central nervous system (CNS) [
        • Abbott N.J.
        • Patabendige A.A.
        • Dolman D.E.
        • Yusof S.R.
        • Begley D.J.
        Structure and function of the blood–brain barrier.
        ]. Air bubbles subjected to cavitation can break the BBB. Exploiting this effect, TFUS combined with US contrast agents is tested as a method for targeted drug delivery [
        • Aryal M.
        • Arvanitis C.D.
        • Alexander P.M.
        • McDannold N.
        Ultrasound-mediated blood–brain barrier disruption for targeted drug delivery in the central nervous system.
        ]. However, BBB opening is undesired for normal TFUS. Assessing BBB integrity is usually based on the intravenous injection of a substance, which cannot cross the barrier under normal conditions, prior to sonication. It is then tested whether TFUS causes the substance to diffuse into brain tissue. The dyes fluorescein isothiocyanate-dextran (FITC-dextran) [
        • Tufail Y.
        • Matyushov A.
        • Baldwin N.
        • Tauchmann M.L.
        • Georges J.
        • Yoshihiro A.
        • et al.
        Transcranial pulsed ultrasound stimulates intact brain circuits.
        ], trypan blue [
        • Kim H.
        • Taghados S.J.
        • Fischer K.
        • Maeng L.S.
        • Park S.
        • Yoo S.S.
        Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
        ,
        • Yoo S.S.
        • Bystritsky A.
        • Lee J.H.
        • Zhang Y.
        • Fischer K.
        • Min B.K.
        • McDannold N.J.
        • Pascual-Leone A.
        • Jolesz F.A.
        Focused ultrasound modulates region-specific brain activity.
        ,
        • Lee W.
        • Croce P.
        • Margolin R.W.
        • Cammalleri A.
        • Yoon K.
        • Yoo S.S.
        Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
        ] or Evans blue [
        • Fisher J.A.
        • Gumenchuk I.
        Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
        ] have been used for this purpose, and their presence inside the brain was investigated in post-mortem microscopy analyses of brain slices. Alternatively, an MRI contrast agent (a gadolinium chelate) was injected before the stimulation and its penetration into brain tissue was tested by assessing the MRI signal change due to the contrast agent [
        • Yoo S.S.
        • Bystritsky A.
        • Lee J.H.
        • Zhang Y.
        • Fischer K.
        • Min B.K.
        • McDannold N.J.
        • Pascual-Leone A.
        • Jolesz F.A.
        Focused ultrasound modulates region-specific brain activity.
        ].
      • Bleeding: The occurrence of bleeding has been investigated using tissue staining, in particular hematoxylin and eosin (H&E) staining [
        • Min B.K.
        • Bystritsky A.
        • Jung K.I.
        • Fischer K.
        • Zhang Y.
        • Maeng L.S.
        • Park S.I.
        • Chung Y.A.
        • Jolesz F.A.
        • Yoo S.S.
        Focused ultrasound-mediated suppression of chemically-induced acute epileptic EEG activity.
        ,
        • Baek H.
        • Pahk K.J.
        • Kim M.J.
        • Youn I.
        • Kim H.
        Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
        ,
        • Verhagen L.
        • Gallea C.
        • Folloni D.
        • Constans C.
        • Jensen D.E.
        • Ahnine H.
        • Roumazeilles L.
        • Santin M.
        • Ahmed B.
        • Lehericy S.
        • Klein-Flügge M.C.
        Offline impact of transcranial focused ultrasound on cortical activation in primates.
        ,
        • Kim H.
        • Taghados S.J.
        • Fischer K.
        • Maeng L.S.
        • Park S.
        • Yoo S.S.
        Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
        ,
        • Kim H.
        • Park M.
        • Wang S.
        • Chiu A.
        • Fischer K.
        • Yoo S.S.
        PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
        ,
        • Kim H.
        • Kim S.
        • Sim N.S.
        • Pasquinelli C.
        • Thielscher A.
        • Lee J.H.
        • Lee H.J.
        Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
        ,
        • Lee W.
        • Lee S.D.
        • Park M.Y.
        • Foley L.
        • Purcell-Estabrook E.
        • Kim H.
        • Fischer K.
        • Maeng L.S.
        • Yoo S.S.
        Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
        ,
        • Kim H.
        • Chiu A.
        • Lee S.D.
        • Fischer K.
        • Yoo S.S.
        Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
        ,
        • Li G.F.
        • Zhao H.X.
        • Zhou H.
        • Yan F.
        • Wang J.Y.
        • Xu C.X.
        • Wang C.Z.
        • Niu L.L.
        • Meng L.
        • Wu S.
        • Zhang H.L.
        Improved anatomical specificity of non-invasive neuro-stimulation by high frequency (5 MHz) ultrasound.
        ,
        • Yang P.S.
        • Kim H.
        • Lee W.
        • Bohlke M.
        • Park S.
        • Maher T.J.
        • Yoo S.S.
        Transcranial focused ultrasound to the thalamus is associated with reduced extracellular GABA levels in rats.
        ,
        • Han S.
        • Kim M.
        • Kim H.
        • Shin H.
        • Youn I.
        Ketamine inhibits ultrasound stimulation-induced neuromodulation by blocking cortical neuron activity.
        ], which reliably stain blood cells. Yet, H&E staining is not specific to blood cells and thus requires experience to correctly interpret the results.
      • Cell death and damage: A general approach to qualitatively analyze the presence of cell death and damage is through H&E staining [
        • Min B.K.
        • Bystritsky A.
        • Jung K.I.
        • Fischer K.
        • Zhang Y.
        • Maeng L.S.
        • Park S.I.
        • Chung Y.A.
        • Jolesz F.A.
        • Yoo S.S.
        Focused ultrasound-mediated suppression of chemically-induced acute epileptic EEG activity.
        ,
        • Baek H.
        • Pahk K.J.
        • Kim M.J.
        • Youn I.
        • Kim H.
        Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
        ,
        • Dallapiazza R.F.
        • Timbie K.F.
        • Holmberg S.
        • Gatesman J.
        • Lopes M.B.
        • Price R.J.
        • Miller G.W.
        • Elias W.J.
        Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound.
        ,
        • Verhagen L.
        • Gallea C.
        • Folloni D.
        • Constans C.
        • Jensen D.E.
        • Ahnine H.
        • Roumazeilles L.
        • Santin M.
        • Ahmed B.
        • Lehericy S.
        • Klein-Flügge M.C.
        Offline impact of transcranial focused ultrasound on cortical activation in primates.
        ,
        • Kim H.
        • Taghados S.J.
        • Fischer K.
        • Maeng L.S.
        • Park S.
        • Yoo S.S.
        Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
        ,
        • Yoo S.S.
        • Bystritsky A.
        • Lee J.H.
        • Zhang Y.
        • Fischer K.
        • Min B.K.
        • McDannold N.J.
        • Pascual-Leone A.
        • Jolesz F.A.
        Focused ultrasound modulates region-specific brain activity.
        ,
        • Lee W.
        • Croce P.
        • Margolin R.W.
        • Cammalleri A.
        • Yoon K.
        • Yoo S.S.
        Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
        ,
        • Kim H.
        • Park M.
        • Wang S.
        • Chiu A.
        • Fischer K.
        • Yoo S.S.
        PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
        ,
        • Kim H.
        • Kim S.
        • Sim N.S.
        • Pasquinelli C.
        • Thielscher A.
        • Lee J.H.
        • Lee H.J.
        Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
        ,
        • Lee W.
        • Lee S.D.
        • Park M.Y.
        • Foley L.
        • Purcell-Estabrook E.
        • Kim H.
        • Fischer K.
        • Maeng L.S.
        • Yoo S.S.
        Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
        ,
        • Kim H.
        • Chiu A.
        • Lee S.D.
        • Fischer K.
        • Yoo S.S.
        Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
        ,
        • Li G.F.
        • Zhao H.X.
        • Zhou H.
        • Yan F.
        • Wang J.Y.
        • Xu C.X.
        • Wang C.Z.
        • Niu L.L.
        • Meng L.
        • Wu S.
        • Zhang H.L.
        Improved anatomical specificity of non-invasive neuro-stimulation by high frequency (5 MHz) ultrasound.
        ,
        • Yang P.S.
        • Kim H.
        • Lee W.
        • Bohlke M.
        • Park S.
        • Maher T.J.
        • Yoo S.S.
        Transcranial focused ultrasound to the thalamus is associated with reduced extracellular GABA levels in rats.
        ,
        • Han S.
        • Kim M.
        • Kim H.
        • Shin H.
        • Youn I.
        Ketamine inhibits ultrasound stimulation-induced neuromodulation by blocking cortical neuron activity.
        ,
        • Daniels D.
        • Sharabi S.
        • Last D.
        • Guez D.
        • Salomon S.
        • Zivli Z.
        • Castel D.
        • Volovick A.
        • Grinfeld J.
        • Rachmilevich I.
        • Amar T.
        • Liraz-Zaltsman S.
        • Sargsyan N.
        • Yael Mardor Y.
        • Harnof S.
        Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
        ,
        • Mehić E.
        • Xu J.M.
        • Caler C.J.
        • Coulson N.K.
        • Moritz C.T.
        • Mourad P.D.
        Increased anatomical specificity of neuromodulation via modulated focused ultrasound.
        ], as described above, cresyl violet Nissl staining [
        • Verhagen L.
        • Gallea C.
        • Folloni D.
        • Constans C.
        • Jensen D.E.
        • Ahnine H.
        • Roumazeilles L.
        • Santin M.
        • Ahmed B.
        • Lehericy S.
        • Klein-Flügge M.C.
        Offline impact of transcranial focused ultrasound on cortical activation in primates.
        ], or luxol fast blue dye (LFB) [
        • Dallapiazza R.F.
        • Timbie K.F.
        • Holmberg S.
        • Gatesman J.
        • Lopes M.B.
        • Price R.J.
        • Miller G.W.
        • Elias W.J.
        Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound.
        ], used to identify myelin in nervous tissue. Cell death can be of two types, apoptosis and necrosis. While apoptosis is part of the normal life cycle of the cells, necrosis is harmful and triggered by external factors or disease. It is possible to differentiate between both types of cell death based on morphological criteria, but this requires experience [
        • Elmore S.A.
        • Dixon D.
        • Hailey J.R.
        • Harada T.
        • Herbert R.A.
        • Maronpot R.R.
        • et al.
        Recommendations from the INHAND apoptosis/necrosis working group.
        ]. Additional techniques specifically label apoptotic cells and have therefore been used to distinguish between apoptosis and necrosis. For example, the presence of fragmented DNA is a sign of apoptosis, and not necrosis, and it can be labeled by terminal deoxynucleotidyl transferase (TdT) dUTP Nick-End Labeling (TUNEL) assay [
        • Min B.K.
        • Bystritsky A.
        • Jung K.I.
        • Fischer K.
        • Zhang Y.
        • Maeng L.S.
        • Park S.I.
        • Chung Y.A.
        • Jolesz F.A.
        • Yoo S.S.
        Focused ultrasound-mediated suppression of chemically-induced acute epileptic EEG activity.
        ,
        • Yoo S.S.
        • Bystritsky A.
        • Lee J.H.
        • Zhang Y.
        • Fischer K.
        • Min B.K.
        • McDannold N.J.
        • Pascual-Leone A.
        • Jolesz F.A.
        Focused ultrasound modulates region-specific brain activity.
        ]. Alternatively, since apoptosis is mediated by caspase [
        • Porter A.G.
        • Jänicke R.U.
        Emerging roles of caspase-3 in apoptosis.
        ], standard immunocytochemistry techniques with antibodies against cleaved caspase-3 can be used [
        • Tufail Y.
        • Matyushov A.
        • Baldwin N.
        • Tauchmann M.L.
        • Georges J.
        • Yoshihiro A.
        • et al.
        Transcranial pulsed ultrasound stimulates intact brain circuits.
        ,
        • Lee W.
        • Croce P.
        • Margolin R.W.
        • Cammalleri A.
        • Yoon K.
        • Yoo S.S.
        Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
        ].
        An alternative approach to detect cell damage is staining for acidophilic cells, for example with VAF (vanadium acid fuchsin) [
        • Lee W.
        • Croce P.
        • Margolin R.W.
        • Cammalleri A.
        • Yoon K.
        • Yoo S.S.
        Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
        ,
        • Victorov I.V.
        • Prass K.
        • Dirnagl U.
        Improved selective, simple, and contrast staining of acidophilic neurons with vanadium acid fuchsin.
        ]. Acidophilia refers to the property of cells of staining readily with an acid dye and occurs after acute neuronal damage and death in brain ischemia.
        Finally, also transmission electron microscopy has been used to quantitatively observe the effect of ultrasound on brain ultrastructure (postsynaptic density, docked vesicles, etc.) [
        • Tufail Y.
        • Matyushov A.
        • Baldwin N.
        • Tauchmann M.L.
        • Georges J.
        • Yoshihiro A.
        • et al.
        Transcranial pulsed ultrasound stimulates intact brain circuits.
        ]. It has been shown that neural trauma causes an abnormal increase in the number of astrocytes [
        • Chen Y.
        • Swanson R.A.
        Astrocytes and brain injury.
        ] that can be detected by the expression of GFAP (glial fibrillary acidic protein) [
        • Lee W.
        • Croce P.
        • Margolin R.W.
        • Cammalleri A.
        • Yoon K.
        • Yoo S.S.
        Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
        ,
        • Fisher J.A.
        • Gumenchuk I.
        Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
        ,
        • Daniels D.
        • Sharabi S.
        • Last D.
        • Guez D.
        • Salomon S.
        • Zivli Z.
        • Castel D.
        • Volovick A.
        • Grinfeld J.
        • Rachmilevich I.
        • Amar T.
        • Liraz-Zaltsman S.
        • Sargsyan N.
        • Yael Mardor Y.
        • Harnof S.
        Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
        ]. One study assessed possible permanent tissue damage after sonication in rats using MRI [
        • Daniels D.
        • Sharabi S.
        • Last D.
        • Guez D.
        • Salomon S.
        • Zivli Z.
        • Castel D.
        • Volovick A.
        • Grinfeld J.
        • Rachmilevich I.
        • Amar T.
        • Liraz-Zaltsman S.
        • Sargsyan N.
        • Yael Mardor Y.
        • Harnof S.
        Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
        ].
      • Irreversible changes of neural activity: Recordings after the sonication can determine whether changes in neural activity are reversible and characterize the duration of recovery. The effects on local neural activity in the TFUS target region can be detected directly via invasive recordings or voltage sensitive dyes [
        • Fisher J.A.
        • Gumenchuk I.
        Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
        ] or Ca2+ imaging in transgenic mice that express the green fluorescent calcium indicator [
        • Fisher J.A.
        • Gumenchuk I.
        Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
        ]. TFUS-related changes in extracellular concentrations of excitatory and inhibitory neurotransmitters such as glutamate and λ-aminobutyric acid (GABA) can be measured via microdialysis techniques [
        • Yang P.S.
        • Kim H.
        • Lee W.
        • Bohlke M.
        • Park S.
        • Maher T.J.
        • Yoo S.S.
        Transcranial focused ultrasound to the thalamus is associated with reduced extracellular GABA levels in rats.
        ]. TFUS has also been combined with measurements of the forelimb and hindlimb responses to epidural cortical stimulation (ECS) to assess the cortical excitability changes after sonication [
        • Gulick D.W.
        • Li T.
        • Kleim J.A.
        • Towe B.C.
        Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
        ]. Alternatively, electroencephalography (EEG) [
        • Yoo S.S.
        • Bystritsky A.
        • Lee J.H.
        • Zhang Y.
        • Fischer K.
        • Min B.K.
        • McDannold N.J.
        • Pascual-Leone A.
        • Jolesz F.A.
        Focused ultrasound modulates region-specific brain activity.
        ], functional MRI [
        • Verhagen L.
        • Gallea C.
        • Folloni D.
        • Constans C.
        • Jensen D.E.
        • Ahnine H.
        • Roumazeilles L.
        • Santin M.
        • Ahmed B.
        • Lehericy S.
        • Klein-Flügge M.C.
        Offline impact of transcranial focused ultrasound on cortical activation in primates.
        ,
        • Folloni D.
        • Verhagen L.
        • Mars R.B.
        • Fouragnan E.
        • Constans C.
        • Aubry J.F.
        • Rushworth M.F.
        • Sallet J.
        Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation.
        ], PET [
        • Kim H.
        • Park M.
        • Wang S.
        • Chiu A.
        • Fischer K.
        • Yoo S.S.
        PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
        ], measurements of peripheral muscle evoked potentials (MEP) [
        • Legon W.
        • Bansal P.
        • Tyshynsky R.
        • Ai L.
        • Mueller J.K.
        Transcranial focused ultrasound neuromodulation of the human primary motor cortex.
        ], sensory evoked potentials (SEP) [
        • Yoo S.S.
        • Yoon K.
        • Croce P.
        • Cammalleri A.
        • Margolin R.W.
        • Lee W.
        Focused ultrasound brain stimulation to anesthetized rats induces long-term changes in somatosensory evoked potentials.
        ], somatosensory evoked potentials (SSEP) [
        • Dallapiazza R.F.
        • Timbie K.F.
        • Holmberg S.
        • Gatesman J.
        • Lopes M.B.
        • Price R.J.
        • Miller G.W.
        • Elias W.J.
        Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound.
        ], visual evoked potentials (VEP) [
        • Kim H.
        • Park M.Y.
        • Lee S.D.
        • Lee W.
        • Chiu A.
        • Yoo S.S.
        Suppression of EEG visual-evoked potentials in rats via neuromodulatory focused ultrasound.
        ] or auditory evoked potentials (AEP) [
        • Daniels D.
        • Sharabi S.
        • Last D.
        • Guez D.
        • Salomon S.
        • Zivli Z.
        • Castel D.
        • Volovick A.
        • Grinfeld J.
        • Rachmilevich I.
        • Amar T.
        • Liraz-Zaltsman S.
        • Sargsyan N.
        • Yael Mardor Y.
        • Harnof S.
        Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
        ] give non-invasive but less specific measurements of neural activity changes.
      • Undesired changes in animal behavior: TFUS may affect normal behavior in unintended ways. In animals, this is controlled by monitoring of every day behavior, like food uptake, defecation and movement behavior and checking for signals of pain and distress or change in weight [
        • Kim H.
        • Taghados S.J.
        • Fischer K.
        • Maeng L.S.
        • Park S.
        • Yoo S.S.
        Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
        ,
        • Yoo S.S.
        • Bystritsky A.
        • Lee J.H.
        • Zhang Y.
        • Fischer K.
        • Min B.K.
        • McDannold N.J.
        • Pascual-Leone A.
        • Jolesz F.A.
        Focused ultrasound modulates region-specific brain activity.
        ,
        • Lee W.
        • Croce P.
        • Margolin R.W.
        • Cammalleri A.
        • Yoon K.
        • Yoo S.S.
        Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
        ,
        • Kim H.
        • Park M.
        • Wang S.
        • Chiu A.
        • Fischer K.
        • Yoo S.S.
        PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
        ,
        • Kim H.
        • Kim S.
        • Sim N.S.
        • Pasquinelli C.
        • Thielscher A.
        • Lee J.H.
        • Lee H.J.
        Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
        ,
        • Lee W.
        • Lee S.D.
        • Park M.Y.
        • Foley L.
        • Purcell-Estabrook E.
        • Kim H.
        • Fischer K.
        • Maeng L.S.
        • Yoo S.S.
        Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
        ,
        • Gulick D.W.
        • Li T.
        • Kleim J.A.
        • Towe B.C.
        Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
        ,
        • Younan Y.
        • Deffieux T.
        • Larrat B.
        • Fink M.
        • Tanter M.
        • Aubry J.F.
        Influence of the pressure field distribution in transcranial ultrasonic neurostimulation.
        ]. In addition, tasks such as the rotorod task and wire-hanging task allow for quantitatively assessing the impact of TFUS on specific aspects of behavior [
        • Tufail Y.
        • Matyushov A.
        • Baldwin N.
        • Tauchmann M.L.
        • Georges J.
        • Yoshihiro A.
        • et al.
        Transcranial pulsed ultrasound stimulates intact brain circuits.
        ]. One study induced ischemic stroke in mice and compared the behavioral changes of the mice which were treated with TFUS via a balance test and an adhesive removal test [
        • Baek H.
        • Pahk K.J.
        • Kim M.J.
        • Youn I.
        • Kim H.
        Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
        ].
      Adverse effects can be caused by cavitation or tissue heating. As outlined above, cavitation is prevented by controlling the pressure levels. While the temperature increase in the brain can be roughly estimated using Equations (7) and (8) in the Supplemetary Material [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ,
      • Baek H.
      • Pahk K.J.
      • Kim M.J.
      • Youn I.
      • Kim H.
      Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
      ,
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ,
      • Lee W.
      • Croce P.
      • Margolin R.W.
      • Cammalleri A.
      • Yoon K.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
      ,
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ,
      • Kim H.
      • Chiu A.
      • Lee S.D.
      • Fischer K.
      • Yoo S.S.
      Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
      ,
      • Yoo S.S.
      • Yoon K.
      • Croce P.
      • Cammalleri A.
      • Margolin R.W.
      • Lee W.
      Focused ultrasound brain stimulation to anesthetized rats induces long-term changes in somatosensory evoked potentials.
      ,
      • Lee W.
      • Kim H.
      • Jung Y.
      • Song I.U.
      • Chung Y.A.
      • Yoo S.S.
      Image-guided transcranial focused ultrasound stimulates human primary somatosensory cortex.
      ], some studies inserted a thermocouple [
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ,
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ,
      • Kim H.
      • Kim S.
      • Sim N.S.
      • Pasquinelli C.
      • Thielscher A.
      • Lee J.H.
      • Lee H.J.
      Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
      ,
      • Gulick D.W.
      • Li T.
      • Kleim J.A.
      • Towe B.C.
      Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
      ,
      • Yoo S.S.
      • Yoon K.
      • Croce P.
      • Cammalleri A.
      • Margolin R.W.
      • Lee W.
      Focused ultrasound brain stimulation to anesthetized rats induces long-term changes in somatosensory evoked potentials.
      ] or an optical fiber based thermal sensor [
      • Li G.F.
      • Zhao H.X.
      • Zhou H.
      • Yan F.
      • Wang J.Y.
      • Xu C.X.
      • Wang C.Z.
      • Niu L.L.
      • Meng L.
      • Wu S.
      • Zhang H.L.
      Improved anatomical specificity of non-invasive neuro-stimulation by high frequency (5 MHz) ultrasound.
      ] in the brain of the animal after craniotomy to track the temperature change in real time during sonication. A non-invasive alternative to this approach is measuring the temperature increase with thermocouples in a phantom [
      • Kim H.
      • Kim S.
      • Sim N.S.
      • Pasquinelli C.
      • Thielscher A.
      • Lee J.H.
      • Lee H.J.
      Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
      ], or MR thermometry [
      • Dallapiazza R.F.
      • Timbie K.F.
      • Holmberg S.
      • Gatesman J.
      • Lopes M.B.
      • Price R.J.
      • Miller G.W.
      • Elias W.J.
      Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound.
      ,
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ,
      • Daniels D.
      • Sharabi S.
      • Last D.
      • Guez D.
      • Salomon S.
      • Zivli Z.
      • Castel D.
      • Volovick A.
      • Grinfeld J.
      • Rachmilevich I.
      • Amar T.
      • Liraz-Zaltsman S.
      • Sargsyan N.
      • Yael Mardor Y.
      • Harnof S.
      Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
      ,
      • Yang P.F.
      • Phipps M.A.
      • Newton A.T.
      • Chaplin V.
      • Gore J.C.
      • Caskey C.F.
      • Chen L.M.
      Neuromodulation of sensory networks in monkey brain by focused ultrasound with MRI guidance and detection.
      ], which exploits temperature sensitive MR parameters such as the water proton resonance frequency, or T1 and T2 relaxation times [
      • Rieke V.
      • Pauly K.B.
      MR thermometry.
      ].
      So far, most TFUS studies used animal models. Tests for adverse effects in the few human studies were based on neurological examinations and/or structural MR imaging before and at one or several time points after the experiment [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ,
      • Lee W.
      • Kim H.
      • Jung Y.
      • Song I.U.
      • Chung Y.A.
      • Yoo S.S.
      Image-guided transcranial focused ultrasound stimulates human primary somatosensory cortex.
      ,
      • Lee W.
      • Chung Y.A.
      • Jung Y.
      • Song I.U.
      • Yoo S.S.
      Simultaneous acoustic stimulation of human primary and secondary somatosensory cortices using transcranial focused ultrasound.
      ]. In some of the studies, the participants were additionally contacted by telephone 2 months after the experiment and interviewed about any changes in their mental and physical health status, including experiences of any discomfort [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ,
      • Lee W.
      • Kim H.
      • Jung Y.
      • Song I.U.
      • Chung Y.A.
      • Yoo S.S.
      Image-guided transcranial focused ultrasound stimulates human primary somatosensory cortex.
      ]. A pre-print manuscript [
      • Legon W.
      • Bansal P.
      • Ai L.
      • Mueller J.K.
      • Meekins G.
      • Gillick B.
      Safety of transcranial focused ultrasound for human neuromodulation.
      ] presents results of phone interviews based on a ‘Participant report of symptoms questionnaire’ of 64 participants who had participated in one or more of seven human TFUS experiments before.

      Results

      Studies screened in this review

      This systematic review follows PRISMA guidelines [
      • Liberati A.
      • Altman D.G.
      • Tetzlaff J.
      • Mulrow C.
      • Gøtzsche P.C.
      • Ioannidis J.P.
      • Clarke M.
      • Devereaux P.J.
      • Kleijnen J.
      • Moher D.
      The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration.
      ,
      • Moher D.
      • Liberati A.
      • Tetzlaff J.
      • Altman D.G.
      Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement.
      ], and the PRISMA checklist can be found in the Supplementary Methods (Table S1). The reviewing process shown in the PRISMA diagram Flow (Fig. 1) resulted in the selection of 31 peer-reviewed and 2 pre-published studies included in this review. From each of those papers (a complete list with citations is shown in Table 2), the sonication parameters (Table 2) and the methods used to assess safety and adverse effects (Table 3) were extracted and categorized as described further below. Often, only some of the safety indices were reported. In that case, we give estimated values when possible. The risk of bias was assessed and is reported in a separate section in the Supplementary Material.

      BBB opening

      BBB opening did not occur in any of the included studies [
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ,
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ,
      • Lee W.
      • Croce P.
      • Margolin R.W.
      • Cammalleri A.
      • Yoon K.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
      ], except for two cases where it was intentionally provoked in control conditions [
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ,
      • Fisher J.A.
      • Gumenchuk I.
      Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
      ], using a high Isppa of 280 W/cm2 [
      • Fisher J.A.
      • Gumenchuk I.
      Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
      ] or an ultrasound contrast agent [
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ].

      Bleeding

      Several studies [
      • Min B.K.
      • Bystritsky A.
      • Jung K.I.
      • Fischer K.
      • Zhang Y.
      • Maeng L.S.
      • Park S.I.
      • Chung Y.A.
      • Jolesz F.A.
      • Yoo S.S.
      Focused ultrasound-mediated suppression of chemically-induced acute epileptic EEG activity.
      ,
      • Baek H.
      • Pahk K.J.
      • Kim M.J.
      • Youn I.
      • Kim H.
      Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
      ,
      • Verhagen L.
      • Gallea C.
      • Folloni D.
      • Constans C.
      • Jensen D.E.
      • Ahnine H.
      • Roumazeilles L.
      • Santin M.
      • Ahmed B.
      • Lehericy S.
      • Klein-Flügge M.C.
      Offline impact of transcranial focused ultrasound on cortical activation in primates.
      ,
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ,
      • Kim H.
      • Park M.
      • Wang S.
      • Chiu A.
      • Fischer K.
      • Yoo S.S.
      PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
      ,
      • Kim H.
      • Kim S.
      • Sim N.S.
      • Pasquinelli C.
      • Thielscher A.
      • Lee J.H.
      • Lee H.J.
      Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
      ,
      • Li G.F.
      • Zhao H.X.
      • Zhou H.
      • Yan F.
      • Wang J.Y.
      • Xu C.X.
      • Wang C.Z.
      • Niu L.L.
      • Meng L.
      • Wu S.
      • Zhang H.L.
      Improved anatomical specificity of non-invasive neuro-stimulation by high frequency (5 MHz) ultrasound.
      ,
      • Yang P.S.
      • Kim H.
      • Lee W.
      • Bohlke M.
      • Park S.
      • Maher T.J.
      • Yoo S.S.
      Transcranial focused ultrasound to the thalamus is associated with reduced extracellular GABA levels in rats.
      ,
      • Han S.
      • Kim M.
      • Kim H.
      • Shin H.
      • Youn I.
      Ketamine inhibits ultrasound stimulation-induced neuromodulation by blocking cortical neuron activity.
      ] tested for bleeding, without finding evidence for it. A further study [
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ] tested different sonication parameters on eight sheep in total, and reported four animals with micro-hemorrhages in the primary visual cortex after undergoing 600 sonications at 6.6 W/cm2 Isppa (6 repetitions of 100 sonications, with 30 s gaps). While the reported value for Isppa is within FDA limits, our calculated value for Ispta of 3.3 W/cm2 is exceeding the diagnostic limit, and is also slightly higher than the limit for physiotherapeutic US of 3 W/cm2. Interestingly, a sheep undergoing a single sonication at an Isppa of 13.4 W/cm2 did not present micro-hemorrhages. In another study [
      • Kim H.
      • Chiu A.
      • Lee S.D.
      • Fischer K.
      • Yoo S.S.
      Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
      ], 1 of 37 rats was exposed to a high intensity (11.2 W/cm2 Ispta) for a short period of time (<9 s using 1 ms TBD, 50% duty cycle and 300 ms SD). It exhibited several areas containing hemosiderin, which indicate the potential of local bleeding, while none of the other animals showed any sign of bleeding.

      Cell damage or death

      Most of the studies testing for cell damage or death [
      • Min B.K.
      • Bystritsky A.
      • Jung K.I.
      • Fischer K.
      • Zhang Y.
      • Maeng L.S.
      • Park S.I.
      • Chung Y.A.
      • Jolesz F.A.
      • Yoo S.S.
      Focused ultrasound-mediated suppression of chemically-induced acute epileptic EEG activity.
      ,
      • Baek H.
      • Pahk K.J.
      • Kim M.J.
      • Youn I.
      • Kim H.
      Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
      ,
      • Dallapiazza R.F.
      • Timbie K.F.
      • Holmberg S.
      • Gatesman J.
      • Lopes M.B.
      • Price R.J.
      • Miller G.W.
      • Elias W.J.
      Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound.
      ,
      • Verhagen L.
      • Gallea C.
      • Folloni D.
      • Constans C.
      • Jensen D.E.
      • Ahnine H.
      • Roumazeilles L.
      • Santin M.
      • Ahmed B.
      • Lehericy S.
      • Klein-Flügge M.C.
      Offline impact of transcranial focused ultrasound on cortical activation in primates.
      ,
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ,
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ,
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ,
      • Lee W.
      • Croce P.
      • Margolin R.W.
      • Cammalleri A.
      • Yoon K.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
      ,
      • Kim H.
      • Park M.
      • Wang S.
      • Chiu A.
      • Fischer K.
      • Yoo S.S.
      PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
      ,
      • Kim H.
      • Kim S.
      • Sim N.S.
      • Pasquinelli C.
      • Thielscher A.
      • Lee J.H.
      • Lee H.J.
      Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
      ,
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ,
      • Kim H.
      • Chiu A.
      • Lee S.D.
      • Fischer K.
      • Yoo S.S.
      Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
      ,
      • Li G.F.
      • Zhao H.X.
      • Zhou H.
      • Yan F.
      • Wang J.Y.
      • Xu C.X.
      • Wang C.Z.
      • Niu L.L.
      • Meng L.
      • Wu S.
      • Zhang H.L.
      Improved anatomical specificity of non-invasive neuro-stimulation by high frequency (5 MHz) ultrasound.
      ,
      • Yang P.S.
      • Kim H.
      • Lee W.
      • Bohlke M.
      • Park S.
      • Maher T.J.
      • Yoo S.S.
      Transcranial focused ultrasound to the thalamus is associated with reduced extracellular GABA levels in rats.
      ,
      • Han S.
      • Kim M.
      • Kim H.
      • Shin H.
      • Youn I.
      Ketamine inhibits ultrasound stimulation-induced neuromodulation by blocking cortical neuron activity.
      ,
      • Daniels D.
      • Sharabi S.
      • Last D.
      • Guez D.
      • Salomon S.
      • Zivli Z.
      • Castel D.
      • Volovick A.
      • Grinfeld J.
      • Rachmilevich I.
      • Amar T.
      • Liraz-Zaltsman S.
      • Sargsyan N.
      • Yael Mardor Y.
      • Harnof S.
      Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
      ,
      • Mehić E.
      • Xu J.M.
      • Caler C.J.
      • Coulson N.K.
      • Moritz C.T.
      • Mourad P.D.
      Increased anatomical specificity of neuromodulation via modulated focused ultrasound.
      ] did not observe harmful effects of TFUS. One recent study [
      • Fisher J.A.
      • Gumenchuk I.
      Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
      ] observed no differential GFAP expression between the control and sonicated hemisphere for Isppa = 0.69 W/cm2, suggesting the absence of neural trauma. However, an increased number of astrocytes was observed for a control condition with Isppa = 280 W/cm2 (∼1.5 times above the FDA limit). Interestingly, no damage was observed even when AEP was not fully recovered after one month [
      • Daniels D.
      • Sharabi S.
      • Last D.
      • Guez D.
      • Salomon S.
      • Zivli Z.
      • Castel D.
      • Volovick A.
      • Grinfeld J.
      • Rachmilevich I.
      • Amar T.
      • Liraz-Zaltsman S.
      • Sargsyan N.
      • Yael Mardor Y.
      • Harnof S.
      Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
      ] in rats.

      Long-term change of neural activity

      Yoo et al. [
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ] tested parameter ranges for excitatory and inhibitory TFUS effects in craniotomized rats. While excitatory effects were very short-termed, suppression effects lasted several minutes. A reduction in the EEG response of up to 80% and a corresponding reduction of the BOLD signal that both lasted up to 10 min were reported for a long sonication duration of 9 s. Dallapiazza et al. [
      • Dallapiazza R.F.
      • Timbie K.F.
      • Holmberg S.
      • Gatesman J.
      • Lopes M.B.
      • Price R.J.
      • Miller G.W.
      • Elias W.J.
      Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound.
      ] observed peak electrophysiological suppression in SSEP 5 min post-treatment, and the values returned to near baseline within 20 min. A further study [
      • Fisher J.A.
      • Gumenchuk I.
      Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
      ] tested the facilitatory effects of ultrasound on somatosensory evoked potentials by measuring the changes in fractional fluorescence in the brains of mice dyed with voltage sensitive dyes. The TFUS-related changes disappeared within 20 min after ultrasound stimulation. Yang et al. [
      • Yang P.S.
      • Kim H.
      • Lee W.
      • Bohlke M.
      • Park S.
      • Maher T.J.
      • Yoo S.S.
      Transcranial focused ultrasound to the thalamus is associated with reduced extracellular GABA levels in rats.
      ] observed a decreased extracellular GABA level (approximately 20% below baseline) compared to a control group that lasted up to 100 min after the sonication ended. The same effect was not observed for glutamate. Gulick et al. [
      • Gulick D.W.
      • Li T.
      • Kleim J.A.
      • Towe B.C.
      Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
      ] showed that TFUS significantly suppressed forelimb and hindlimb responses to ECS for several minutes after the stimulation blocks, even though effects immediately after single, short TFUS trials were absent. Kim et al. [
      • Kim H.
      • Park M.
      • Wang S.
      • Chiu A.
      • Fischer K.
      • Yoo S.S.
      PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
      ] observed a local increase in glucose metabolism induced by FUS to rat brain. This effect was demonstrated via PET imaging, which was started 20 min after the sonication and performed for 1 h. After that time, the metabolism had still not returned to baseline. In a work [
      • Verhagen L.
      • Gallea C.
      • Folloni D.
      • Constans C.
      • Jensen D.E.
      • Ahnine H.
      • Roumazeilles L.
      • Santin M.
      • Ahmed B.
      • Lehericy S.
      • Klein-Flügge M.C.
      Offline impact of transcranial focused ultrasound on cortical activation in primates.
      ] on primates, the authors observed change in functional connectivity after a long sonication of 40 s at Ispta = 7 W/cm2. The change lasted for more than 1 h after sonication. A similar effect was observed in a related work [
      • Folloni D.
      • Verhagen L.
      • Mars R.B.
      • Fouragnan E.
      • Constans C.
      • Aubry J.F.
      • Rushworth M.F.
      • Sallet J.
      Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation.
      ], where they used a sonication of 40 s at a maximum Ispta of 15.3 W/cm2. Yoo et al. [
      • Yoo S.S.
      • Yoon K.
      • Croce P.
      • Cammalleri A.
      • Margolin R.W.
      • Lee W.
      Focused ultrasound brain stimulation to anesthetized rats induces long-term changes in somatosensory evoked potentials.
      ] observed that the SEP signals after 10 min sonication were distinctively different compared to the control condition, even 35 min after the sonication. Daniels et al. [
      • Daniels D.
      • Sharabi S.
      • Last D.
      • Guez D.
      • Salomon S.
      • Zivli Z.
      • Castel D.
      • Volovick A.
      • Grinfeld J.
      • Rachmilevich I.
      • Amar T.
      • Liraz-Zaltsman S.
      • Sargsyan N.
      • Yael Mardor Y.
      • Harnof S.
      Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
      ] observed a full recovery of AEP amplitudes in rats within maximum 1 week post-treatment with Isppa = 2.3 W/cm2, while the signal from 5 out of 10 rats recovered up to one month post-treatment for an Isppa = 4.6 W/cm2. In the same study, 1 out of 5 pigs showed a fully recovered signal 1 h post-treatment while the other did not show any recovery 3 h post-treatment (in all 5 cases, Isppa = 4.6 W/cm2). Kim et al. [
      • Kim H.
      • Park M.Y.
      • Lee S.D.
      • Lee W.
      • Chiu A.
      • Yoo S.S.
      Suppression of EEG visual-evoked potentials in rats via neuromodulatory focused ultrasound.
      ] observed an increase in VEP in rats up to 5 min post-treatment with Isppa = 5 W/cm2 and a slight increase of VEP 150 s after treatment when Isppa = 3W/cm2. One study [
      • Baek H.
      • Pahk K.J.
      • Kim M.J.
      • Youn I.
      • Kim H.
      Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
      ] induced ischemic stroke in mice and found a better sensorimotor performance in mice that underwent 20 min TFUS session via a balance test and an adhesive removal test. These improvements lasted for 4 weeks after treatment, suggesting an enhancement in brain plasticity. In the same study, the TFUS treatment in cerebellar LCN significantly lowered the percentage change in increased water content and tissue swelling in the ipsilateral hemisphere to the stroke.

      Animal behavior

      Several studies tested for changes from normal daily behavior after the sonication studies [
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ,
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ,
      • Lee W.
      • Croce P.
      • Margolin R.W.
      • Cammalleri A.
      • Yoon K.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
      ,
      • Kim H.
      • Park M.
      • Wang S.
      • Chiu A.
      • Fischer K.
      • Yoo S.S.
      PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
      ,
      • Kim H.
      • Kim S.
      • Sim N.S.
      • Pasquinelli C.
      • Thielscher A.
      • Lee J.H.
      • Lee H.J.
      Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
      ,
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ,
      • Gulick D.W.
      • Li T.
      • Kleim J.A.
      • Towe B.C.
      Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
      ,
      • Younan Y.
      • Deffieux T.
      • Larrat B.
      • Fink M.
      • Tanter M.
      • Aubry J.F.
      Influence of the pressure field distribution in transcranial ultrasonic neurostimulation.
      ], but did not find any abnormalities. A single study also employed behavioral tasks [
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ] (rotorod running task and wire-hanging task), without revealing differences in motor performance.

      Temperature

      Theoretical calculations based on Equations (7) and (8) in the Supplementary Methods suggest that “typical” TFUS parameters used so far in most studies cause negligible temperature increases in brain tissue [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ,
      • Baek H.
      • Pahk K.J.
      • Kim M.J.
      • Youn I.
      • Kim H.
      Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
      ,
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ,
      • Lee W.
      • Croce P.
      • Margolin R.W.
      • Cammalleri A.
      • Yoon K.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
      ,
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ,
      • Kim H.
      • Chiu A.
      • Lee S.D.
      • Fischer K.
      • Yoo S.S.
      Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
      ,
      • Yoo S.S.
      • Yoon K.
      • Croce P.
      • Cammalleri A.
      • Margolin R.W.
      • Lee W.
      Focused ultrasound brain stimulation to anesthetized rats induces long-term changes in somatosensory evoked potentials.
      ,
      • Lee W.
      • Kim H.
      • Jung Y.
      • Song I.U.
      • Chung Y.A.
      • Yoo S.S.
      Image-guided transcranial focused ultrasound stimulates human primary somatosensory cortex.
      ]. In a recent study [
      • Verhagen L.
      • Gallea C.
      • Folloni D.
      • Constans C.
      • Jensen D.E.
      • Ahnine H.
      • Roumazeilles L.
      • Santin M.
      • Ahmed B.
      • Lehericy S.
      • Klein-Flügge M.C.
      Offline impact of transcranial focused ultrasound on cortical activation in primates.
      ], this was partly confirmed using the more realistic bio-heat equation to estimate the temperature increase after 40 s of TFUS through a 3 mm thick skull, with an Ispta = 7 W/cm2 in the brain. The maximal increase in the brain was less than 0.2 °C. Interestingly, however, they found rather strong increases in the skull (2.8 °C). Also experimental results show mostly only small temperature increases due to sonication [
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ,
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ,
      • Kim H.
      • Kim S.
      • Sim N.S.
      • Pasquinelli C.
      • Thielscher A.
      • Lee J.H.
      • Lee H.J.
      Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
      ,
      • Li G.F.
      • Zhao H.X.
      • Zhou H.
      • Yan F.
      • Wang J.Y.
      • Xu C.X.
      • Wang C.Z.
      • Niu L.L.
      • Meng L.
      • Wu S.
      • Zhang H.L.
      Improved anatomical specificity of non-invasive neuro-stimulation by high frequency (5 MHz) ultrasound.
      ,
      • Yoo S.S.
      • Yoon K.
      • Croce P.
      • Cammalleri A.
      • Margolin R.W.
      • Lee W.
      Focused ultrasound brain stimulation to anesthetized rats induces long-term changes in somatosensory evoked potentials.
      ]. However, it is important to note that the overall temperature increase depends on the combination of several TFUS parameters. For example, one study [
      • Li G.F.
      • Zhao H.X.
      • Zhou H.
      • Yan F.
      • Wang J.Y.
      • Xu C.X.
      • Wang C.Z.
      • Niu L.L.
      • Meng L.
      • Wu S.
      • Zhang H.L.
      Improved anatomical specificity of non-invasive neuro-stimulation by high frequency (5 MHz) ultrasound.
      ] reported a measured peak temperature increase of 0.2 °C for an extended stimulation (∼30 min) at a low Ispta≤230 mW/cm2 at 1 MHz (1.6 °C at 5 MHz for otherwise same parameters). In contrast, another study [
      • Gulick D.W.
      • Li T.
      • Kleim J.A.
      • Towe B.C.
      Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
      ] reported a temperature increase up to 3 °C after two blocks of 5 min stimulation at 200 kHz, separated by a 2 min break, at Ispta = 4.5 W/cm2 and a MI = 3.1 (higher than the allowed limit). Both studies applied longer durations than used in most other TFUS studies so far, but the combination with the higher Ispta caused noticeable temperature rises in the second study.
      A temperature increase of 0.5 °C was reported through MR thermometry after 30 s sonication at Isppa = 9.9 W/cm2 [
      • Yang P.F.
      • Phipps M.A.
      • Newton A.T.
      • Chaplin V.
      • Gore J.C.
      • Caskey C.F.
      • Chen L.M.
      Neuromodulation of sensory networks in monkey brain by focused ultrasound with MRI guidance and detection.
      ]. Another study [
      • Daniels D.
      • Sharabi S.
      • Last D.
      • Guez D.
      • Salomon S.
      • Zivli Z.
      • Castel D.
      • Volovick A.
      • Grinfeld J.
      • Rachmilevich I.
      • Amar T.
      • Liraz-Zaltsman S.
      • Sargsyan N.
      • Yael Mardor Y.
      • Harnof S.
      Focused ultrasound-induced suppression of auditory evoked potentials in vivo.
      ] reported temperature variation within the measurement noise level of the baseline temperatures (±2 °C) with MR thermography. The strongest effect was reported by a study using MR thermometry (sensitivity 0.3 ± 0.06 °C [
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ]), demonstrating an increase of ∼0.7 °C in the sonicated area [
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ], using an Isppa = 23 W/cm2 for 27 s. Dallapiazza et al. [
      • Dallapiazza R.F.
      • Timbie K.F.
      • Holmberg S.
      • Gatesman J.
      • Lopes M.B.
      • Price R.J.
      • Miller G.W.
      • Elias W.J.
      Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound.
      ] showed a negligible temperature increase during treatment using both MR thermometry and estimations based on the bio-heat equation.

      Findings from human studies

      In a recent preprint work [
      • Legon W.
      • Bansal P.
      • Tyshynsky R.
      • Ai L.
      • Mueller J.K.
      Transcranial focused ultrasound neuromodulation of the human primary motor cortex.
      ], the authors tested the effects of ultrasound stimulation on motor cortex excitability measured by single-pulse transcranial magnetic stimulation (TMS). They report significant changes in the recorded muscle responses to TMS only when it was applied during, but not after, sonication. Follow-up neurological exams and anatomical MRIs after the TFUS experiment did not reveal any abnormalities or changes in the mental or physical status, nor any discomfort associated with the procedure [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ,
      • Lee W.
      • Kim H.
      • Jung Y.
      • Song I.U.
      • Chung Y.A.
      • Yoo S.S.
      Image-guided transcranial focused ultrasound stimulates human primary somatosensory cortex.
      ,
      • Lee W.
      • Chung Y.A.
      • Jung Y.
      • Song I.U.
      • Yoo S.S.
      Simultaneous acoustic stimulation of human primary and secondary somatosensory cortices using transcranial focused ultrasound.
      ]. Follow-up interviews at later time points confirmed those observations. A recent study published as preprint [
      • Legon W.
      • Bansal P.
      • Ai L.
      • Mueller J.K.
      • Meekins G.
      • Gillick B.
      Safety of transcranial focused ultrasound for human neuromodulation.
      ] presents results from a follow-up questionnaire after TFUS that could be obtained from 64 out of in total 120 participants. Seven subjects reported mild or moderate symptoms (mild neck pain, scalp tingling, headache, difficulty paying attention, muscle twitches and anxiety) that they felt were possibly or probably related to the experiment. These initial symptoms disappeared upon follow-up. The authors found a linear correlation (r = 0.797, p = 0.0319) between Isppa and the occurrence of observed symptoms among the 7 subjects who reported mild to moderate symptoms that were perceived as ‘possibly’ or ‘probably’ related to participation in TFUS experiments.

      Discussion and conclusions

      Harmful effects of TFUS were absent in the majority of the 33 studies reviewed here. In two cases, microhemorrhages occurred in a subset of the tested animals when using a high Ispta of 11.2 W/cm2 for a short duration [
      • Kim H.
      • Chiu A.
      • Lee S.D.
      • Fischer K.
      • Yoo S.S.
      Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
      ] or an Ispta of ≥3.3 W/cm2 for a high number of sonications (≥500) given at a relatively short ISI of 1s [
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ]. Both doses are clearly above the safety limits of the FDA guidelines for diagnostic US and above the IEC standard 60601-2-5 for physiotherapy US equipment. However, this also holds for several other included studies, where no adverse effects were reported. While the parameters chosen in one of the studies [
      • Kim H.
      • Chiu A.
      • Lee S.D.
      • Fischer K.
      • Yoo S.S.
      Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
      ] did not result in substantial heating, as also pointed out by the authors, the high Ispta of 11.2 W/cm2 differentiates it from many other TFUS studies. That might indicate that mechanical effects caused the microhemorrhages, even though the limits for MI and Isppa were not exceeded. However, as this was observed in only one of the tested animals, this conclusion remains very speculative and a replication including sham controls would be favorable to ensure that the microhemorrhages were indeed related to TFUS. In the second study [
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ], the chosen parameter combination might have led to a high total energy deposit, opening the possibility that a thermal mechanism underlay the adverse effects that occurred in four animals. For example, Gulick et al. [
      • Gulick D.W.
      • Li T.
      • Kleim J.A.
      • Towe B.C.
      Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
      ] observed a temperature increase of 3 °C for a less intense protocol using an Ispta = 4.5 W/cm2 and in total 180 sonications in a time period of 13 min. It seems reasonable to assume that heating might have been even higher in the four animals that showed microhemorrhages in Ref. [
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ] and indicates that calculating the temperature increase for a single sonication, as done in Ref. [
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ], can strongly underestimate the real increase.
      While Isppa stayed below the safety limit in all studies, MI exceeded the limit in two studies [
      • Gulick D.W.
      • Li T.
      • Kleim J.A.
      • Towe B.C.
      Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
      ,
      • Folloni D.
      • Verhagen L.
      • Mars R.B.
      • Fouragnan E.
      • Constans C.
      • Aubry J.F.
      • Rushworth M.F.
      • Sallet J.
      Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation.
      ] and Ispta exceeded the FDA limits for diagnostic US in soft tissue in 14 out of the 20 studies in which Ispta was reported or could be calculated post hoc (values after cranial transmission or for craniotomized animals). Ispta was also above the physiotherapeutic limit in 11 of the 20 studies. This suggests that Ispta is the most sensitive safety index in case of TFUS and, unlike current practice, should be reported so that it can be followed up by a more detailed estimation of the thermal effects when its limits are exceeded. We consider this relevant as the current studies indicate that TFUS parameters within the FDA limits for diagnostic US might often lack neural stimulation effectiveness. For example, an Ispta of around 2 W/cm2 for pulsed waves and 4 W/cm2 for continuous waves was necessary to reach a 50% success rate for stimulation at 500 kHz [
      • King R.L.
      • Brown J.R.
      • Newsome W.T.
      • Pauly K.B.
      Effective parameters for ultrasound-induced in vivo neurostimulation.
      ]. Similarly, while many studies included in this review reported neural effects for parameters within the safety limits [
      • Min B.K.
      • Bystritsky A.
      • Jung K.I.
      • Fischer K.
      • Zhang Y.
      • Maeng L.S.
      • Park S.I.
      • Chung Y.A.
      • Jolesz F.A.
      • Yoo S.S.
      Focused ultrasound-mediated suppression of chemically-induced acute epileptic EEG activity.
      ,
      • Baek H.
      • Pahk K.J.
      • Kim M.J.
      • Youn I.
      • Kim H.
      Modulation of cerebellar cortical plasticity using low-intensity focused ultrasound for poststroke sensorimotor function recovery.
      ,
      • Tufail Y.
      • Matyushov A.
      • Baldwin N.
      • Tauchmann M.L.
      • Georges J.
      • Yoshihiro A.
      • et al.
      Transcranial pulsed ultrasound stimulates intact brain circuits.
      ,
      • Fisher J.A.
      • Gumenchuk I.
      Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo.
      ,
      • Kim H.
      • Kim S.
      • Sim N.S.
      • Pasquinelli C.
      • Thielscher A.
      • Lee J.H.
      • Lee H.J.
      Miniature ultrasound ring array transducers for transcranial ultrasound neuromodulation of freely-moving small animals.
      ,
      • Yoo S.S.
      • Yoon K.
      • Croce P.
      • Cammalleri A.
      • Margolin R.W.
      • Lee W.
      Focused ultrasound brain stimulation to anesthetized rats induces long-term changes in somatosensory evoked potentials.
      ,
      • Kim H.
      • Park M.Y.
      • Lee S.D.
      • Lee W.
      • Chiu A.
      • Yoo S.S.
      Suppression of EEG visual-evoked potentials in rats via neuromodulatory focused ultrasound.
      ,
      • Yang P.F.
      • Phipps M.A.
      • Newton A.T.
      • Chaplin V.
      • Gore J.C.
      • Caskey C.F.
      • Chen L.M.
      Neuromodulation of sensory networks in monkey brain by focused ultrasound with MRI guidance and detection.
      ,
      • Yoo S.S.
      • Kim H.
      • Min B.K.
      • Franck S.P.E.
      Transcranial focused ultrasound to the thalamus alters anesthesia time in rats.
      ], several studies found stable effects only when exceeding at least one of the safety indices ([
      • Kim H.
      • Taghados S.J.
      • Fischer K.
      • Maeng L.S.
      • Park S.
      • Yoo S.S.
      Noninvasive transcranial stimulation of rat abducens nerve by focused ultrasound.
      ,
      • Yoo S.S.
      • Bystritsky A.
      • Lee J.H.
      • Zhang Y.
      • Fischer K.
      • Min B.K.
      • McDannold N.J.
      • Pascual-Leone A.
      • Jolesz F.A.
      Focused ultrasound modulates region-specific brain activity.
      ,
      • Lee W.
      • Croce P.
      • Margolin R.W.
      • Cammalleri A.
      • Yoon K.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of motor cortical areas in freely-moving awake rats.
      ,
      • Lee W.
      • Lee S.D.
      • Park M.Y.
      • Foley L.
      • Purcell-Estabrook E.
      • Kim H.
      • Fischer K.
      • Maeng L.S.
      • Yoo S.S.
      Image-Guided focused ultrasound-mediated regional brain stimulation in sheep.
      ,
      • Kim H.
      • Chiu A.
      • Lee S.D.
      • Fischer K.
      • Yoo S.S.
      Focused ultrasound-mediated non-invasive brain stimulation: examination of sonication parameters.
      ] and Table 2). In addition, recent studies show that heating of the skull (potentially causing indirect heating of soft tissue) and/or brain tissue can reach several degrees for more intense and long protocols [
      • Verhagen L.
      • Gallea C.
      • Folloni D.
      • Constans C.
      • Jensen D.E.
      • Ahnine H.
      • Roumazeilles L.
      • Santin M.
      • Ahmed B.
      • Lehericy S.
      • Klein-Flügge M.C.
      Offline impact of transcranial focused ultrasound on cortical activation in primates.
      ,
      • Gulick D.W.
      • Li T.
      • Kleim J.A.
      • Towe B.C.
      Comparison of electrical and ultrasound neurostimulation in rat motor cortex.
      ]. The systematic assessment of heating will thus be relevant in future studies that might aim at extending the parameter envelope of TFUS and should be part of any safety test of new sonication regimes in particular for human TFUS.
      It is worth noting that the safety limit of 720 mW/cm2 for Ispta, which was generally used in TFUS studies so far and which we also applied here, was introduced to limit the heating in soft tissue. In case of transcranial US, the FDA limits for diagnostic US actually apply an even stricter limit of 94 mW/cm2 for Ispta to prevent excessive heating of the skull, which absorbs most of the beam energy. It seems that almost none of the studies published so far reported neural effects for intensities below this threshold. However, it is important to stress that both limits are based on worst-case scenarios and exceeding them does not necessarily mean that strong heating occurs. Rather, the FDA standard for diagnostic US requires a case-by-case estimation of the maximum temperature rise in soft tissue and skull once they are exceeded, specific for the used ultrasound parameters and setup. Simulations of the propagations of the TFUS beam through the skull, combined with evaluations of the bio-heat equation for TFUS [
      • Verhagen L.
      • Gallea C.
      • Folloni D.
      • Constans C.
      • Jensen D.E.
      • Ahnine H.
      • Roumazeilles L.
      • Santin M.
      • Ahmed B.
      • Lehericy S.
      • Klein-Flügge M.C.
      Offline impact of transcranial focused ultrasound on cortical activation in primates.
      ], might be valuable tools that allow realistic estimates of the amount of heating for new sonication regimes on a more standard basis.
      The neural aftereffects can exceed 1 h [
      • Verhagen L.
      • Gallea C.
      • Folloni D.
      • Constans C.
      • Jensen D.E.
      • Ahnine H.
      • Roumazeilles L.
      • Santin M.
      • Ahmed B.
      • Lehericy S.
      • Klein-Flügge M.C.
      Offline impact of transcranial focused ultrasound on cortical activation in primates.
      ,
      • Kim H.
      • Park M.
      • Wang S.
      • Chiu A.
      • Fischer K.
      • Yoo S.S.
      PET/CT imaging evidence of FUS-mediated (18)F-FDG uptake changes in rat brain.
      ,
      • Folloni D.
      • Verhagen L.
      • Mars R.B.
      • Fouragnan E.
      • Constans C.
      • Aubry J.F.
      • Rushworth M.F.
      • Sallet J.
      Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation.
      ], making TFUS a potent neuromodulation modality. This is encouraging for therapeutic applications. In contrast to diagnostic US, future TFUS applications might resort to repeated sessions over extended time periods to achieve and maintain therapeutic efficacy. As such, a safety framework will also need to cover these more intense settings (see, e.g. Refs. [
      • Elmar F.
      • Wilfurth S.
      • Landgrebe M.
      • Eichhammer P.
      • Hajak G.
      • Langguth B.
      Anodal skin lesions after treatment with transcranial direct current stimulation.
      ,
      • Palm U.
      • Keeser D.
      • Schiller C.
      • Fintescu Z.
      • Reisinger E.
      • Padberg F.
      • Nitsche M.
      Skin lesions after treatment with transcranial direct current stimulation (tDCS).
      ] for a related example of adverse effects that only occurred after repeated applications in case of transcranial direct current stimulation) or combinations of TFUS with other brain stimulation techniques. This will require safety studies that specifically test this parameter space in order to inform an international consensus on accepted settings and procedures, similar to established non-invasive brain stimulation methods [
      • Rossini P.M.
      • Burke D.
      • Chen R.
      • Cohen L.G.
      • Daskalakis Z.
      • Di Iorio R.
      Non-invasive electrical and magnetic stimulation of the brain, spinal cord, roots and peripheral nerves: basic principles and procedures for routine clinical and research application. An updated report from an IFCN Committee.
      ]. Along similar lines, in the few TFUS studies performed in humans so far, the type and extent of follow up exams differed strongly [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ,
      • Lee W.
      • Kim H.
      • Jung Y.
      • Song I.U.
      • Chung Y.A.
      • Yoo S.S.
      Image-guided transcranial focused ultrasound stimulates human primary somatosensory cortex.
      ,
      • Lee W.
      • Chung Y.A.
      • Jung Y.
      • Song I.U.
      • Yoo S.S.
      Simultaneous acoustic stimulation of human primary and secondary somatosensory cortices using transcranial focused ultrasound.
      ,
      • Legon W.
      • Bansal P.
      • Ai L.
      • Mueller J.K.
      • Meekins G.
      • Gillick B.
      Safety of transcranial focused ultrasound for human neuromodulation.
      ]. This suggests a need for guidelines that provide a secure framework for experimental settings and practical procedures, including mandatory safety screening and appropriate follow-up procedures. For example, the importance of establishing best practices also for apparently simple procedures was highlighted in a recent review [
      • Farzaneh A.
      • McLoughlin I.V.
      • Chauhan S.
      • Ter-Haar G.
      Bio-effects and safety of low-intensity, low-frequency ultrasonic exposure.
      ] of low-intensity low-frequency US (20–100 kHz), showing that US can cause skin damage due to inertial effect cavitation in the coupling gel if non-degassed gel is employed.
      Along similar lines, guidelines are important to prevent intensity hotspots that can occur due to unintended standing waves and focusing effects of the skull. While these effects more likely emerge in small animals [
      • Younan Y.
      • Deffieux T.
      • Larrat B.
      • Fink M.
      • Tanter M.
      • Aubry J.F.
      Influence of the pressure field distribution in transcranial ultrasonic neurostimulation.
      ], they have been shown to be also relevant in non-human primates for targets close to the skull base such as the amygdala [
      • Folloni D.
      • Verhagen L.
      • Mars R.B.
      • Fouragnan E.
      • Constans C.
      • Aubry J.F.
      • Rushworth M.F.
      • Sallet J.
      Manipulation of subcortical and deep cortical activity in the primate brain using transcranial focused ultrasound stimulation.
      ]. Moreover, a retrospective modeling study [
      • Baron C.
      • Aubry J.F.
      • Tanter M.
      • Meairs S.
      • Fink M.
      Simulation of intracranial acoustic fields in clinical trials of sonothrombolysis.
      ] suggests that unwanted secondary hotspots might have been the cause of intracerebral hemorrhages that occurred in a clinical trial on transcranial low frequency ultrasound for sonothrombolysis [
      • Daffertshofer M.
      • Gass A.
      • Ringleb P.
      • Sitzer M.
      • Sliwka U.
      • Els T.
      • Sedlaczek O.
      • Koroshetz W.J.
      • Hennerici M.G.
      Transcranial lowfrequency ultrasound-mediated thrombolysis in brain ischemia. Increased risk of hemorrhage with combined ultrasound and tissue plasminogen activator. Results of a phase II clinical trial.
      ] and that resulted in the early termination of the trial. Finally, the reviewed studies differed in regards to the choice of the stated safety-relevant parameters and the way those were assessed. A more standardized reporting of the relevant pulse parameters and of all safety indices of the FDA guidelines is a prerequisite for the development of future TFUS guidelines for human applications. Accurate estimation of the TFUS intensity after cranial transmission is particularly challenging in humans, as it has to rely on hydrophone measurements based on “representative” skull samples or computer simulations [
      • Lee W.
      • Kim H.C.
      • Jung Y.
      • Chung Y.A.
      • Song I.U.
      • Lee J.H.
      • Yoo S.S.
      Transcranial focused ultrasound stimulation of human primary visual cortex.
      ,
      • Lee W.
      • Kim H.
      • Jung Y.
      • Song I.U.
      • Chung Y.A.
      • Yoo S.S.
      Image-guided transcranial focused ultrasound stimulates human primary somatosensory cortex.
      ,
      • Legon W.
      • Sato T.F.
      • Opitz A.
      • Mueller J.
      • Barbour A.
      • Williams A.
      • Tyler W.J.
      Transcranial focused ultrasound modulates the activity of primary somatosensory cortex in humans.
      ]. The uncertainty range of the intensity estimates obtained by these procedures seems still unclear [
      • Robertson J.
      • Martin E.
      • Cox B.
      • Treeby B.E.
      Sensitivity of simulated transcranial ultrasound fields to acoustic medium property maps.
      ,
      • Pasquinelli C.
      • Montanaro H.
      • Neufeld E.
      • Lee H.J.
      • Thielscher A.
      Impact of the skull model on simulated TFUS beam profiles.
      ], and contributes to variations in the values reported across studies. As such, it seems useful that future studies additionally state intensity values for a pure water background to ensure good comparability of the baseline TFUS parameters.

      Conflicts of interest

      None declared.

      Acknowledgements

      We thank the anonymous reviewers for their helpful comments. CP was supported by a PhD stipend of the Technical University of Denmark . HRS holds a professorship in Precision Medicine at the Institute of Clinical Medicine, Faculty of Health and Medical Sciences, Copenhagen University, sponsored by Lundbeckfonden. AT and HRS were supported by a synergy grant from the Novo Nordisk Foundation (Interdisciplinary Synergy Program 2014; grant number NNF14OC0011413 ).

      Appendix A. Supplementary data

      The following is the Supplementary data to this article: