1.IntroductionPrecise noninvasive neuromodulation can potentially help better understanding the inner workings of the brain and tackle the rise of neurodegenerative diseases in an aging population. Deep brain electrical stimulation, which employs surgical implantation of electrodes, has become a treatment option for several neurological disorders including Parkinson’s disease and epilepsy.1,2 Noninvasive brain stimulation techniques based on electric or magnetic fields are incapable of targeting deeper structures without affecting more superficially located tissues.3 In contrast, sound waves can be focused into tiny tissue volumes with minimal collateral effects.4 Focused ultrasound (FUS) thus has the potential to noninvasively target nearly any brain area, both in animal models and humans.5–8 Unlike the mechanisms of electric and magnetic field interaction with the neurons, the effects of ultrasonic fields at the cellular, network, and whole brain level have not been fully understood,9–12 chiefly due to the lack of efficient methods for noninvasive real-time observation of ultrasound neuromodulation (USNM) effects. Understanding how the brain reacts to mechanical stimuli requires a new set of tools as ultrasound (US) produces multiple physical effects, including radiation force, heating, and cavitation.13,14 Despite the repeated evidence of neuronal activation upon US stimuli in isolated neurons and cell cultures,10,12,15,16 observing such responses in vivo remains challenging, partially due to the lack of consistency in the stimulation parameters used across different studies.17 Using mouse models expressing fluorescent calcium sensors18 with precise US delivery poses a technical challenge of combining precise FUS delivery with real-time optical imaging.19,20 Most USNM experiments in mice have therefore relied on motor evoked responses generated by directly sonicating large focal areas in the cortical and subcortical regions of the murine brain.21 Reduction in latency and increased response has been observed following electrical stimulation with FUS pre-treatment.19 A thermosensitve ion channel, TRPV1, was sensitized to respond to FUS-induced heating,22 yet no direct calcium signal was reported without genetic manipulations.22 Recently, we integrated high-resolution FUS delivery and simultaneous widefield fluorescence imaging to achieve and characterize highly precise FUS targeting in a living mouse brain.23 However, our initial observations were dominated by the thermal dynamics of fluoro-thermal tags (FTT) or propagating spreading depolarizations, and no localized neural activity has been isolated. Here, we further advance the ability of cortex-wide fluorescence imaging to observe responses to precisely steered localized FUS stimuli by characterizing the activation dynamics and developing a method to separate thermal fluorescence quenching and actual neural responses. While this is, to our knowledge, the first systematic attempt to compensate for thermal events in fluorescence neuroimaging—major thermal events are pervasive in modern neurotechnology, and thus associated confounds are likely present in a myriad of related studies using one- or two-photon imaging.9,24 2.Methods2.1.Experimental Setup and ProceduresThe fluorescence-guided focused ultrasound (FLUS) system has been designed to achieve simultaneous fluorescence imaging and precise noninvasive FUS stimulation of the murine brain. A wide-angle spherical US array (Imasonic, France) consisting of 512 transducer elements having a wide (3 to 9 MHz) effective bandwidth was employed for delivering FUS into the target location [Fig. 1(a)]. The array is capable of generating small focal spots (measuring down to ) through the mouse skull at any depth and location in the brain.25 At the same time, the exact location of the focus can accurately be tracked in three dimensions by means of real-time volumetric optoacoustic tomography (VOT) feedback performed with the same spherical array. For this, excitation of optoacoustic responses is performed with a pulsed laser beam (800 nm wavelength) guided by means of optical fiber bundles to the tissue surface. Absorption of a single 10 ns pulse duration laser pulse by tissue chromophores, such as hemoglobin, triggers the generation of tiny US vibrations, which are detected by the spherical array. The VOT images are then rendered at a real-time frame rate of 25 Hz established by the pulse repetition rate of the laser.26,27 The imaging feedback and US emission are automatically co-registered by the time-reversal principle since both are employing the same transducer array. As a result, the VOT images can be used to precisely navigate the location of the US stimulation target. During the experiments, the array was immersed in deionized water at room temperature and coupled to the sample using a thin ( thick) polyvinyl chloride film that is transparent to both light and US. Fluorescence imaging was performed simultaneously with the FUS emission through an 8-mm-diameter centrally located hole in the spherical array by means of a flexible fiberscope, attaining an effective 12-mm-diameter circular FOV at lateral resolution. A continuous wave laser at 488 nm is used for exciting the GCamp6f calcium-sensitive proteins expressed in the mouse brain. 2.2.In Vivo ExperimentsSeven GCaMP6f mice [C57BL/6J-Tg(Thy1-GCaMP6f) GP5.17Dkim/J, the Jackson Laboratory] were used for this study (three female and four male) aged between 5 and 6.5 weeks. The animals were housed in individually ventilated cages inside a temperature-controlled room, under a 12-h dark/12-h light cycle. Pelleted food (3437PXL15 and CARGILL) and water were provided ad libitum. All experiments were performed in accordance with the Swiss Federal Act on Animal Protection and approved by the Cantonal Veterinary Office Zurich. The mouse head was secured using a custom stereotactic frame (Narishige International Limited, London, United Kingdom) fixed by a holder to minimize motion artifacts for acquiring in vivo images during FUS stimulation. Blood oxygen saturation, heart rate, and mouse body temperature were continuously monitored. The core body temperature was maintained at using a homeothermic temperature controller coupled to a heating pad, both of which were controlled by PhysioSuite (Kent Scientific, Torrington, Connecticut). To ensure optimal US coupling, the hair on the mouse head was removed. We injected buprenorphine () subcutaneously and removed the scalp after 30 min. A 40% dilution of phosphate-buffered saline in ultrasound gel (Aquasonic Clear, Parker Laboratories Inc., Fairfield, New Jersey) was deposited on mouse’s scalp and brought into contact with the transparent membrane of the tank filled with degassed water to ensure unobstructed transmission of US into the mouse brain for imaging. All mice were sonicated under isoflurane anesthesia [3% (v/v) for induction and 1.2% (v/v) for maintenance] through an intact skull with 150-ms duration pulses at 3 MHz delivered in the mouse cortex [Fig. 1(c)]. In the repeated FUS stimulation studies peak pressure range was adjusted from 2.5 to 2.8 MPa and the constant time interval between two sequential stimulations was 10 s to minimize interference among consecutive stimulations. The array’s generated pressure at the focus was measured with a calibrated hydrophone through a mouse skull. The US intensity can be approximated from the pressure as , where , , and represent the pressure, density, and speed of sound, respectively.7 To test the temporal precision and repeatability of different FUS parameters, we applied 20 repeated stimuli in each experiment. No unusual behavior was observed during the experiments. 2.3.Ex Vivo ExperimentThe brain of one GCaMP 6f mouse was extracted and cut into 1 mm thick slices. The slices are immersed in a temperature-controlled water bath. The temperature was continuously monitored using a thermocouple (IT-23, Physitemp Instruments, Clifton, New Jersey) and recorded to a PC by means of a USB interface (NI 9213, National Instruments, Austin, Texas). Heating and subsequent cooling cycles are averaged together. Fluorescence was recorded using the same setup previously described in Sec. 2.1. 2.4.Data AnalysisThe pipeline for analyzing the data is depicted in Fig. 2. The time profiles from fluorescent recordings originate from the same location where the US was emitted and focal FTT-related dip was observed. As a pre-processing step, we denoised the image stack with a predictive Kalman filter in ImageJ with a bias of 0.5 for average sensitivity to momentary fluctuations. The filter is applied on a per-slice basis to the time-lapse sequence of raw fluorescent images (co-registered on the atlas). All other data analyses were conducted using MATLAB (2021b Mathworks, Massachusetts) and custom Python scripts (version 3.10.5). Fluorescence calcium recordings were band-pass-filtered between 0 and 8 Hz and normalized by calculating the relative change to the baseline with a moving baseline (0 to 0.05 Hz) to remove signal drifts due to laser energy fluctuations or photobleaching. A total of 20 stimulations separated by a period of 10 s are averaged to cancel noise and remove strong background signal variations due to dynamics, which were found to be an order of magnitude larger than the US-induced responses and obscured responses from single excitation events. An isotropic Gaussian filter with a kernel size of 1 pixel () was then applied to smoothen the image. To further increase the signal-to-background ratio (SBR), we took advantage of the high interhemispheric correlation of resting state dynamics and subtracted the signals recorded from the opposite hemisphere to the FUS delivery. Activation is localized based on the observed focal FTT dip in the region of interest with high precision in time and space. The calculated spatio–temporal signature of the FTT23 was subsequently subtracted from the processed profile in time and space. The processed time profiles were temporally smoothed by Savitzky–Golay filter with a filter window of 11 and polynomial order of 2 to fit the samples. For quantitative analysis, normalized peak amplitude was identified for each profile as the maximum percentage of relative change with respect to the baseline during 2 s after the US stimulation onset. 2.5.SimulationsFUS simulations are performed assuming linear US wave propagation with the software FieldII.28 The simulations were calibrated using hydrophone measurements performed under full water immersion in a water tank.23,25 Simulations of the spatio–temporal heat deposition dynamics induced by FUS are modeled using the bioheat model29 as implemented by Soneson.30 The input to the model corresponds to the spatial US field distribution using the same parameters used in the experiments with the thermal constants adapted from the literature.23,31 3.ResultsIn absence of USNM, the mouse brain under isoflurane anesthesia (1.2%) presents spontaneous resting-state calcium dynamics as the background signal [Fig. 1(c)]. Once the FUS pulse is applied, the FTT response occurs almost simultaneously, as indicated by the purple spot [Fig. 3(a), see Fig. S1 in the Supplementary Material for the raw response]. The FTT reflects the position and size of the US focus, and it also precedes the subsequent neural activation event that spreads over a wider area [light green spot in Fig. 3(a)]. However, the resting-state activity of the anesthetized mice creates a strong background visible in the images [light purple, Fig. 3(a)] and fluorescence time traces [Fig. 3(a) below]. Given the high inter-hemisphere (IH) correlation of the resting-state signals, one can subtract the left hemisphere to cancel the resting state and obtain a much clearer view of both the FTT and the subsequent localized activation [Fig. 3(b)]. The red-colored time trace of the raw fluorescence recording at the focal point depicts both events, with a rapidly rising activation following the FTT. No relevant activity was observed near the auditory cortex. To test our hypothesis on the thermal origin of the FTT, we use a bioheat model23 to correct for the thermal transients (Fig. 4). The FUS delivery can increase the temperature at the focus and its immediate vicinity [Fig. 4(a)]. The simulated US intensity at the focus and the pulse duration (150 ms) serve as input to the bioheat model to predict the spatio–temporal evolution of the temperature changes. The tightly focused heat source rapidly increases its temperature for the 80 to 150 ms time window, followed by heat diffusion at 300 to 600 ms. A more detailed analysis of the temporal signal evolution at different points surrounding the US focus [Fig. 4(b)] confirms the fast rise and slower decay of the FUS-induced temperature changes. The temperature dependence of the fluorescence brightness [Fig. 5(a)] has been further validated using ex-vivo brain slices of GCaMP-6f-expressing mouse. As expected, an increase in temperature results in quenching of the fluorescence.23 The dependence of the mean FTT decay on the FUS intensity [Fig. 5(b)] reveals a negative correlation (Pearson correlation ), confirming the US–thermal–fluorescence quenching process. Looking at FTT’s spatial footprint, the model is in reasonable agreement with the experimental data acquired in-vivo through the mouse skull. Changing the pressure after IH subtraction generates a pattern of two opposing phenomena [Fig. 6(a)]. On the one hand, we observe the deepening of the FTT followed by a stronger calcium response with the increase in pressure. The simulated thermal transient is subsequently subtracted from the measured signal and smoothed to clearly reveal the underlying FUS-evoked activation [Fig. 6(b)]. The activation was robust and consistent among mice with a peak latency of [Fig. 6(c)], as measured from the onset of the stimulation to the activation peak. These results are generally in agreement with previous GCaMP6f-based measurements of FUS stimuli in vitro10,12,32,33 in terms of activation rise time and signal shape. We next examined whether the observed responses were stationary across the experiment by comparing the mean responses for the first and last 10 stimulations. Our results (Fig. S2, Table S1 in the Supplementary Material) validate that responses did not fatigue during the experiments. 4.DiscussionOur results show that the correction of rapid thermal confounds is a potentially crucial and feasible step toward direct evidence of neuronal network activation upon precise FUS stimulation in vivo. The significance and benefit of this processing solution is high: precise stimulations in both lateral and axial dimensions together with FTT-guided FUS delivery overcomes the limitations of previous studies.34 FTT-guided FUS delivery rules out such confounds as it can monitor precisely the US delivery location. Our study found that relatively high pressure levels are needed to activate the mouse brain at the 3 MHz frequency used in our experiments. The detection sensitivity and resolution limits of the imaging system may have limited our ability to measure weak activations at lower pressures. Nonetheless, in contrast to other high resolution imaging approaches over restricted (sub-millimeter) field of view, cortex-wide fluorescence allows minimizing the resting state background thus obtaining cleaner activation traces. Previous studies reported lower activation thresholds with significantly longer stimulations, e.g., for neurons over-expressing the TRPV1 ion channel where stimulation durations in the 7 s range were used.22 The relatively high neural activation thresholds can arguably also be attributed to the high temperature increase in the targeted spot, which in turn may have caused inhibitory effect thus partially canceling out excitatory effects.33,35–40 This balance of excitatory and inhibitory events is probably omni-present in this type of experiments and remains to be carefully explored using the tools introduced here. The 3 MHz excitation frequency used in our study produces obvious transient thermal effects and higher radiation forces as compared to lower frequencies. Conversely, the probability of inducing cavitation is also lower41 at maximum mechanical index (MI) of 1.6 used in our study, i.e., below the FDA-required safety limit of 1.9 for diagnostic US imaging. Photobleaching and laser heating due to the continuous wave laser excitation (see Sec. 2.1) are present in the signals but occur on much longer time scales than the transient ultrasonic heating. Therefore, baseline correction and cycle averaging should be used to remove slow photobleaching and thermal effects from the signals. Future work should aim at characterizing specific regimes for optimal FUS stimulation under specific experimental conditions and application-related requirements. Our flexible image-guided platform enables systematic testing over a wide parameter space in various brain regions. Due to the bulky arrayed US transducer setup, animal studies are mostly limited to stimulation under anesthesia or heavy sedation, which typically suppress the neural response to stimulation,11 or otherwise to head-restrained, awake animals. Furthermore, deep learning methods can be developed and integrated into the analysis pipeline for spatiotemporal enhancement and denoising of calcium imaging responses.42 5.ConclusionThis study introduced a non-invasive US stimulation technique with precise volumetric optoacoustic navigation and simultaneous fluorescence calcium recordings of the cortical responses. The method can target deep murine brain regions with high spatiotemporal resolution thus holding promise to advance the study of the nervous system and uncover new ways to treat neurological disorders. In addition, the careful handling of thermal confounds is crucial to the understanding of the stimulation processes and clearly differentiate between thermal and neural responses. We expect our method could also find application in other neurostimulation modalities that cause thermal transients and rely on fluorescence as readout of neural responses. Future studies will evaluate various underlying phenomena over a wide range of parameters. Code and Data AvailabilityThe data that support the findings of this study are available from the corresponding author upon reasonable request. Author ContributionsD.R. and S.S. conceived the experimental system. H.E. and A.O. developed the experimental system. H.E. and N.D. carried out experiments in mice with the help of A.O. H.E. developed the software. H.E. and N.D. performed data analysis and visualization. D.R. and S.S. supervised the study. H.E. and N.D. wrote the manuscript. All authors reviewed and edited the manuscript. AcknowledgmentsWe thank the valuable assistance of M. Reiss with the mouse experiments. This work was supported by the National Institutes of Health grants UF1-NS107680 and RF1-NS126102 (to S.S. and D.R.), 5R01-NS109885-02 (to S.S.), and the Swiss National Science Foundation Grant 310030_192757 (to D.R.). ReferencesK. Follett et al.,
“Pallidal versus subthalamic deep-brain stimulation for Parkinson’s disease,”
N. Engl. J. Med., 362
(22), 2077
–2091 https://doi.org/10.1056/NEJMoa0907083
(2010).
Google Scholar
M. J. Cook et al.,
“Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study,”
Lancet Neurol., 12 563
–571 https://doi.org/10.1016/S1474-4422(13)70075-9
(2013).
Google Scholar
A. Bhattacharya et al.,
“An overview of non-invasive brain stimulation: basic principles and clinical applications,”
Can. J. Neurol. Sci., 49 479
–492 https://doi.org/10.1017/cjn.2021.158
(2022).
Google Scholar
R. F. Dallapiazza et al.,
“Noninvasive neuromodulation and thalamic mapping with low-intensity focused ultrasound,”
J. Neurosurg., 128 875
–884 https://doi.org/10.3171/2016.11.JNS16976 JONSAC 0022-3085
(2017).
Google Scholar
Y. Tufail et al.,
“Transcranial pulsed ultrasound stimulates intact brain circuits,”
Neuron, 66 681
–694 https://doi.org/10.1016/j.neuron.2010.05.008 NERNET 0896-6273
(2010).
Google Scholar
W. Legon et al.,
“Transcranial focused ultrasound modulates the activity of primary somatosensory cortex in humans,”
Nat. Neurosci., 17 322
–329 https://doi.org/10.1038/nn.3620 NANEFN 1097-6256
(2014).
Google Scholar
O. Naor, S. Krupa and S. Shoham,
“Ultrasonic neuromodulation,”
J. Neural Eng., 13 031003 https://doi.org/10.1088/1741-2560/13/3/031003 1741-2560
(2016).
Google Scholar
K. S. Lee et al.,
“Focused ultrasound stimulation as a neuromodulatory tool for Parkinson’s disease: a scoping review,”
Brain Sci., 12 289 https://doi.org/10.3390/brainsci12020289
(2022).
Google Scholar
Z. Cheng et al.,
“High resolution ultrasonic neural modulation observed via in vivo two-photon calcium imaging,”
Brain Stimul., 15 190
–196 https://doi.org/10.1016/j.brs.2021.12.005
(2022).
Google Scholar
E. Weinreb and E. Moses,
“Mechanistic insights into ultrasonic neurostimulation of disconnected neurons using single short pulses,”
Brain Stimul., 15 769
–779 https://doi.org/10.1016/j.brs.2022.05.004
(2022).
Google Scholar
P.-F. Yang et al.,
“Bidirectional and state-dependent modulation of brain activity by transcranial focused ultrasound in non-human primates,”
Brain Stimul., 14 261
–272 https://doi.org/10.1016/j.brs.2021.01.006
(2021).
Google Scholar
S. Yoo et al.,
“Focused ultrasound excites cortical neurons via mechanosensitive calcium accumulation and ion channel amplification,”
Nat. Commun., 13 493 https://doi.org/10.1038/s41467-022-28040-1 NCAOBW 2041-1723
(2022).
Google Scholar
M. Plaksin, E. Kimmel and S. Shoham,
“Cell-type-selective effects of intramembrane cavitation as a unifying theoretical framework for ultrasonic neuromodulation,”
eNeuro, 3 ENEURO.0136-15.2016 https://doi.org/10.1523/ENEURO.0136-15.2016
(2016).
Google Scholar
W. J. Tyler,
“The mechanobiology of brain function,”
Nat. Rev. Neurosci., 13 867
–878 https://doi.org/10.1038/nrn3383 NRNAAN 1471-003X
(2012).
Google Scholar
W. J. Tyler et al.,
“Remote excitation of neuronal circuits using low-intensity, low-frequency ultrasound,”
PLoS One, 3 e3511 https://doi.org/10.1371/journal.pone.0003511 POLNCL 1932-6203
(2008).
Google Scholar
B. Clennell et al.,
“Transient ultrasound stimulation has lasting effects on neuronal excitability,”
Brain Stimul., 14 217
–225 https://doi.org/10.1016/j.brs.2021.01.003
(2021).
Google Scholar
A. Guerra and M. Bologna,
“Low-intensity transcranial ultrasound stimulation: mechanisms of action and rationale for future applications in movement disorders,”
Brain Sci., 12 611 https://doi.org/10.3390/brainsci12050611
(2022).
Google Scholar
T.-W. Chen et al.,
“Ultrasensitive fluorescent proteins for imaging neuronal activity,”
Nature, 499 295
–300 https://doi.org/10.1038/nature12354
(2013).
Google Scholar
J. A. Fisher and I. Gumenchuk,
“Low-intensity focused ultrasound alters the latency and spatial patterns of sensory-evoked cortical responses in vivo,”
J. Neural Eng., 15 035004 https://doi.org/10.1088/1741-2552/aaaee1 1741-2560
(2018).
Google Scholar
T. Sato, M. G. Shapiro and D. Y. Tsao,
“Ultrasonic neuromodulation causes widespread cortical activation via an indirect auditory mechanism,”
Neuron, 98 1031
–1041.e5 https://doi.org/10.1016/j.neuron.2018.05.009 NERNET 0896-6273
(2018).
Google Scholar
C. Aurup, H. A. Kamimura and E. E. Konofagou,
“High-resolution focused ultrasound neuromodulation induces limb-specific motor responses in mice in vivo,”
Ultrasound Med. Biol., 47 998
–1013 https://doi.org/10.1016/j.ultrasmedbio.2020.12.013
(2021).
Google Scholar
Y. Yang et al.,
“Sonothermogenetics for noninvasive and cell-type specific deep brain neuromodulation,”
Brain Stimul., 14 790
–800 https://doi.org/10.1016/j.brs.2021.04.021
(2021).
Google Scholar
H. Estrada et al.,
“High-resolution fluorescence-guided transcranial ultrasound mapping in the live mouse brain,”
Sci. Adv., 7 eabi5464 https://doi.org/10.1126/sciadv.abi5464
(2021).
Google Scholar
A. Kaszas et al.,
“Two-photon GCaMP6f imaging of infrared neural stimulation evoked calcium signals in mouse cortical neurons in vivo,”
Sci. Rep., 11 9775 https://doi.org/10.1038/s41598-021-89163-x
(2021).
Google Scholar
H. Estrada et al.,
“Spherical array system for high-precision transcranial ultrasound stimulation and optoacoustic imaging in rodents,”
IEEE Trans. Ultrasonics, Ferroelectr. Freq. Control, 68 107
–115 https://doi.org/10.1109/TUFFC.2020.2994877
(2020).
Google Scholar
X. L. Deán-Ben and D. Razansky,
“Portable spherical array probe for volumetric real-time optoacoustic imaging at centimeter-scale depths,”
Opt. Express, 21
(23), 28062
–28071 https://doi.org/10.1364/OE.21.028062 OPEXFF 1094-4087
(2013).
Google Scholar
X. L. Deán-Ben, A. Ozbek and D. Razansky,
“Volumetric real-time tracking of peripheral human vasculature with GPU-accelerated three-dimensional optoacoustic tomography,”
IEEE Trans. Med. Imaging, 32
(11), 2050
–2055 https://doi.org/10.1109/TMI.2013.2272079 ITMID4 0278-0062
(2013).
Google Scholar
D. Bæk, J. A. Jensen and M. Willatzen,
“Modeling transducer impulse responses for predicting calibrated pressure pulses with the ultrasound simulation program field II,”
J. Acoust. Soc. Am., 127 2825
–2835 https://doi.org/10.1121/1.3365317 JASMAN 0001-4966
(2010).
Google Scholar
H. H. Pennes,
“Analysis of tissue and arterial blood temperatures in the resting human forearm,”
J. Appl. Physiol. (1985), 1 93
–122 https://doi.org/10.1152/jappl.1948.1.2.93
(1948).
Google Scholar
J. Soneson,
“High intensity focused ultrasound simulator,”
(2023). https://www.mathworks.com/matlabcentral/fileexchange/30886-high-intensity-focused-ultrasound-simulator Google Scholar
C. M. Collins, M. B. Smith and R. Turner,
“Model of local temperature changes in brain upon functional activation,”
J. Appl. Physiol. (1985), 97 2051
–2055 https://doi.org/10.1152/japplphysiol.00626.2004
(2004).
Google Scholar
M. Duque et al.,
“Sonogenetic control of mammalian cells using exogenous transient receptor potential A1 channels,”
Nat. Commun., 13 600 https://doi.org/10.1038/s41467-022-28205-y NCAOBW 2041-1723
(2022).
Google Scholar
S. Sharabi et al.,
“Non-thermal focused ultrasound induced reversible reduction of essential tremor in a rat model,”
Brain Stimul., 12 1
–8 https://doi.org/10.1016/j.brs.2018.08.014
(2019).
Google Scholar
G. Li et al.,
“Imaging-guided dual-target neuromodulation of the mouse brain using array ultrasound,”
IEEE Trans. Ultrasonics, Ferroelectr. Freq. Control, 65 1583
–1589 https://doi.org/10.1109/TUFFC.2018.2847252
(2018).
Google Scholar
A. Banerjee, R. Egger and M. A. Long,
“Using focal cooling to link neural dynamics and behavior,”
Neuron, 109 2508
–2518 https://doi.org/10.1016/j.neuron.2021.05.029 NERNET 0896-6273
(2021).
Google Scholar
P. C. Petersen, M. Vöröslakos and G. Buzsáki,
“Brain temperature affects quantitative features of hippocampal sharp wave ripples,”
J. Neurophysiol., 127 1417
–1425 https://doi.org/10.1152/jn.00047.2022 JONEA4 0022-3077
(2022).
Google Scholar
M. Ganguly et al.,
“Thermal block of action potentials is primarily due to voltage-dependent potassium currents: a modeling study,”
J. Neural Eng., 16 036020 https://doi.org/10.1088/1741-2552/ab131b 1741-2560
(2019).
Google Scholar
M. Gotoh et al.,
“Brain temperature alters contributions of excitatory and inhibitory inputs to evoked field potentials in the rat frontal cortex,”
Front. Cell Neurosci., 14 593027 https://doi.org/10.3389/fncel.2020.593027
(2020).
Google Scholar
I. Leake,
“Turning up the heat,”
Nat. Rev. Neurosci., 20 447
–447 https://doi.org/10.1038/s41583-019-0203-8 NRNAAN 1471-003X
(2019).
Google Scholar
H. Guo et al.,
“Ultrasound does not activate but can inhibit in vivo mammalian nerves across a wide range of parameters,”
Sci. Rep., 12 2182 https://doi.org/10.1038/s41598-022-05226-7
(2022).
Google Scholar
X. Qian et al.,
“Noninvasive ultrasound retinal stimulation for vision restoration at high spatiotemporal resolution,”
BME Front., 2022 9829316 https://doi.org/10.34133/2022/9829316
(2022).
Google Scholar
X. Li et al.,
“Reinforcing neuron extraction and spike inference in calcium imaging using deep self-supervised denoising,”
Nat. Methods, 18 1395
–1400 https://doi.org/10.1038/s41592-021-01225-0 1548-7091
(2021).
Google Scholar
BiographyNeda Davoudi is currently a senior scientist at ETH AI Center in Zurich. She received her PhD in the Information Technology and Electrical Engineering Department from ETH Zurich. She received her MSc degree in biomedical computing from Technical University of Munich. She worked as a researcher in the Institute for Biological and Medical Imaging at Helmholtz Center. She is an active reviewer in biomedical imaging communities. Her research interests are centered on the applications of machine learning in medicine. Héctor Estrada received his PhD in physical acoustics from the Universidad Politecnica de Valencia and joined the Institute for Biological and Medical Imaging, Helmholtz Zentrum München as a postdoctoral fellow. He currently works as a senior scientist at the Razansky lab at the University and ETH Zurich, where he develops models and techniques for transcranial ultrasound and optoacoustics. Ali Özbek studied electrical engineering and information technology in Technische Universität München developing methods for light focusing through a scattering medium with optoacoustic feedback. He received his PhD from the Department of Information Technologies and Electrical Engineering, ETH Zurich working on ultrafast optoacoustic image acquisition and compressed sensing with applications in cardiac and brain imaging fields and developing tools for ultrasound neuromodulation. Shy Shoham, SPIE fellow, is a professor of neuroscience and ophthalmology at NYU School of Medicine, and co-director of NYU Tech4Health institute. His lab develops photonic, acoustic, and computational tools for neural interfacing. He received his BSc degree from Tel Aviv University, his PhD in bioengineering from the University of Utah, and was a Lewis-Thomas postdoctoral fellow at Princeton University. He serves on the editorial boards of SPIE Neurophotonics and Journal of Neural Engineering and has co-edited the Handbook of Neurophotonics. Daniel Razansky (fellow – SPIE, IEEE, Optica) is a full professor of biomedical imaging at the University of Zurich and ETH Zurich. He received his biomedical and electrical engineering degrees from the Technion - Israel Institute of Technology and completed postdoctoral training at the Harvard Medical School. His lab pioneered a number of bioimaging technologies for pre-clinical research and clinical diagnostics, among them the multi-spectral optoacoustic tomography and hybrid optoacoustic ultrasound imaging. |
Fluorescence
Calcium
Brain
Thermal modeling
Ultrasonography
Thermal effects
Model based design