Model-based correction of rapid thermal confounds in fluorescence neuroimaging of targeted perturbation

Abstract

Significance: An array of techniques for targeted neuromodulation is emerging, with high potential in brain research and therapy. Calcium imaging or other forms of functional fluorescence imaging are central solutions for monitoring cortical neural responses to targeted neuromodulation, but often are confounded by thermal effects that are inter-mixed with neural responses.

Aim: Here, we develop and demonstrate a method for effectively suppressing fluorescent thermal transients from calcium responses.

Approach: We use high precision phased-array 3 MHz focused ultrasound delivery integrated with fiberscope-based widefield fluorescence to monitor cortex-wide calcium changes. Our approach for detecting the neural activation first takes advantage of the high inter-hemispheric correlation of resting state ${\mathrm{Ca}}^{2+}$ dynamics and then removes the ultrasound-induced thermal effect by subtracting its simulated spatio-temporal signature from the processed profile.

Results: The focused $350\mu m$-sized ultrasound stimulus triggered rapid localized activation events dominated by transient thermal responses produced by ultrasound. By employing bioheat equation to model the ultrasound heat deposition, we can recover putative neural responses to ultrasound.

Conclusions: The developed method for canceling transient thermal fluorescence quenching could also find applications with optical stimulation techniques to monitor thermal effects and disentangle them from neural responses. This approach may help deepen our understanding of the mechanisms and macroscopic effects of ultrasound neuromodulation, further paving the way for tailoring the stimulation regimes toward specific applications.

Keywords: calcium imaging; mouse brain; neuroimaging; thermal effects; ultrasound neuromodulation.

Fig. 1

Overview of the FLUS experimental setup and data. (a) Schematic of the multimodal FLUS system. (b) The experimental protocol uses optoacoustic volumetric imaging to precisely navigate the US stimulation to the desired target (red arrow). (c) Fluorescence data are continuously recorded over 20 stimulation cycles (the simultaneous FUS emissions are marked with cyan bars). CW, continuous wave; OA, optoacoustic; US, ultrasound; FL, fluorescent. $\mathrm{\Delta }F/F_0$ corresponds to relative fluorescence intensity changes to the baseline.

Fig. 2

Data pre- and post-processing pipeline. The raw image stack is denoised with a predictive Kalman filter, band-pass-filtered between 0 and 8 Hz and normalized by calculating the relative fluorescence change relative to the moving baseline to remove signal drifts due to laser energy fluctuations or photobleaching. A total of 20 stimulations separated by a period of 10 s are then averaged to cancel noise and remove background from the ${\mathrm{Ca}}^{2+}$ dynamics. An isotropic Gaussian filter is then applied to smooth the image. Signals recorded from the opposite hemisphere to the FUS delivery are subtracted to further increase the SBR. The calculated spatio–temporal signature of the FTT is subsequently subtracted from the processed profile in time and space and temporally smoothed by Savitzky–Golay filter.

Fig. 3

Snapshots of the US-induced fluorescence changes showing the immediate FTT responses and 500 ms thereafter. (a) The data were averaged over 20 stimulation cycles. Color arrows indicate the points where the time traces (below) were extracted. (b) IH subtraction cancels out highly correlated resting state, hence revealing the FTT followed by localized activation in the stimulated area.

Fig. 4

Model-based simulation of thermal effects. (a) US thermal deposition and diffusion simulated using the bioheat model for a continuous 0.15 s duration US pulse. The image on the left shows the simulated US focus on the axial plane, followed by a temporal sequence of the corresponding temperature change. (b) Time evolution of the temperature change extracted at different distances from the US focus (see labels). A blue rectangle marks the sonication time.

Fig. 5

Validation of thermal model versus FTT signal. (a) Change in fluorescence brightness as a function of temperature measured from GCaMP 6f mouse brain slices in a saline solution bath. Experimental data points include error bars corresponding to the standard error of the mean, while the solid line shows an affine fit. (b) The mean of the FTT dip ($\mathrm{\Delta }{F}_{\mathrm{US}}$) during the sonication as a function of US intensity ($n=7$). Pearson correlation coefficient is indicated as $R$ in the plot. Shaded region corresponds to the 95% confidence interval. (c) Spatial FTT dip in simulation versus experiment.

Fig. 6

Correction for thermal responses reveals the underlying neural responses to US stimulation. (a) Interhemispheric traces in the stimulated region for different pressures. (b) The IH signal is corrected using thermal model. (c) The corrected US-mediated activation in $n=6$ animals. Gray curves depict traces from different mice, black curve corresponds to the mean response.