Diagnostic ultrasound enhances, then reduces, exogenously induced brain activity of mice

Abstract

Transcranially delivered diagnostic ultrasound (tDUS) applied to the human brain can modulate those brains such that they became more receptive to external stimulation relative to sham ultrasound exposure. Here, we sought to directly measure the effect of tDUS on mouse brain activity subjected to an external stimulation-a blinking light. Using electrocorticography, we observed a substantial increase in median brain activity due to tDUS plus a blinking light relative to baseline and relative to sham tDUS plus a blinking light. Subsequent brain activity decreased after cessation of tDUS but with continuation of the blinking light, though it remained above that demonstrated by mice exposed to only a blinking light. In a separate experiment, we showed that tDUS alone, without a blinking light, had no observable effect on median brain activity, but upon its cessation, brain activity decreased. These results demonstrate that simultaneous exposure to tDUS and blinking light can increase the receptivity of the visual cortex of mice exposed to that light, and that prior exposure to tDUS can reduce subsequent brain activity. In each case, these results are consistent with published data. Our results on mice echo published human results but do not directly explain them, since their test subjects received less intense diagnostic ultrasound than did our mice. Given the near ubiquity of diagnostic ultrasound systems, further progress along this line of research could one day lead to the widespread use of diagnostic ultrasound to intentionally modulate human brain function during exogenous stimulation.

Keywords: LIFU; diagnostic ultrasound; electrocorticography; focused ultrasound; neuromodulation; ultrasound; ultrasound stimulation; visual stimulation.

Figure 1

Schematic of experimental setup. (A) Shows setup from the side. (B) Shows setup from above. Dotted line shows the position and orientation of the ultrasound scan head. Dots in the mouse brain of (B) denote the position of ECoG electrodes.

Figure 2

Schematic representation of the experiment for a given mouse. Here, Trial 1 refers to mice exposed only to (sham) blinking light. Trial 2 refers to mice exposed to (sham) blinking light plus (sham) tDUS. Trial 3 refers to mice exposed only to a (sham) blinking light. The Light+US cohort received each of blinking light and tDUS. The Light-only cohort experienced only a blinking light. The US-only cohort received only tDUS.

Figure 3

Schematic of analysis, here applied to sample representative data. (A) Sample baseline-normalized ECoG data from one light event (from top to bottom): raw data, 1 second block RMS (tick marks denote center of time segment), 0.25-second running RMS (used in our analysis) and 0.1-second running RMS (tick marks denote end of time segment). (B) Representative set of 55 brain-activity events collected during the entire Trial 1 for a single mouse. Each row contains 0.25-second running RMS brain-activity data. The red box outlines the entire set of brain activity events for this mouse. The dotted green box outlines the first two seconds of that data. (C) Results of averaging the 0.25-second running RMS of each event over the entire ten seconds for each event. (D) Results of averaging the 0.25-second running RMS of each event over the first two seconds of each event. (E) Calculation of the median of all 55 calculations shown in (D). We used this median value calculated for each mouse to normalize all the event data for each mouse in all Trials.

Figure 4

Median RMS brain activity in the brains of mice as a function of their cohort and Trial types, for event lengths of 10 s. Each line shows the median of a 0.25 s running RMS of brain activity during each event, normalized by baseline brain activity. Light stimulation, lasting 0.1 s, occurred at the start of each event.

Figure 5

Statistical significance versus length of subset of events for Trial 1 cohorts, data normalized by baseline brain activity. (A) Mann-Whitney analysis p-value versus event length of median RMS brain activity for the US+Light and Light-only cohorts during the first Trial. (B) Kruskal-Wallis analysis of the US+Light, Light-only, and US-only cohorts. The red line denotes the log₁₀ of the effective p-value (α = 0.005, adjusted for multiple comparisons of n = 10 pairs of data) below which values denote statistical significance between cohorts and above which comparisons do not differ in a statistically significant fashion. Grey areas denote statistically significant cases as a function of subset of event length.

Figure 6

Log₁₀ of the p-value for Mann-Whitney analysis of intra-cohort brain activity observed during Trial 1 versus Trial 2, as a function of subset of event length. (A) Shows this comparison within the US+Light cohort. (B) Shows this comparison within the Light-only cohort. (C) Shows this comparison within the US-only cohort. The red line denotes the log₁₀ of the effective p-value (α = 0.0017, adjusted for multiple comparisons of n = 30 pairs of data) below which values denote statistical significance between cohorts and above which comparisons do not differ in a statistically significant fashion. Grey areas denote statistically significant cases as a function of subset of event length.

Figure 7

Log₁₀ of p-value for Mann-Whitney analysis of intra-cohort brain activity observed during Trial 2 versus Trial 3 as a function of event length. (A) Shows this comparison within the US+Light cohort. (B) Shows this comparison within the Light-only cohort. (C) Shows this comparison within the US-only cohort. The red line denotes the log₁₀ of the effective p-value (α = 0.0017, adjusted for multiple comparisons of N = 30 pairs of data) below which values denote statistical significance between cohorts and above which comparisons do not differ in a statistically significant fashion. Grey areas denote statistically significant cases as a function of subset of event length.

Figure 8

Representative intra-cohort comparison of brain activity versus Trial number. Meaningful differences occur as a function of event length between Trial 2 and Trial 3 for the US+Light and Light-only cohorts. NS denotes insignificant difference, while asterisk denotes statistical significance with p < 0.00056, adjusted for n = 90 multiple comparisons using Mann-Whitney. The central mark of each boxplot denotes the median value of the distribution of the data while the bottom and top of the box mark the 25th and 75th percentiles of the data, respectively. Whiskers enclose greater than 99% of the data.

Figure 9

Representative intra-trial comparison of brain activity versus cohort type. Meaningful differences occur as a function of event length only during Trial 3. NS denotes insignificant difference, while asterisk denotes statistical significance, with p < 0.00083, adjusted for n = 60 multiple comparisons using Mann Whitney.