Functional neuroimaging

The high spatial resolution of functional ultrasound allows task-based neuroimaging (also called functional activation mapping) to be used very successfully to investigate brain responses to various stimuli, including visual and auditory inputs.

This was one of the first applications of functional ultrasound, and since 2011 numerous studies have used our technology, as summarized below.

Whisker stimulation

Physiological stimulation to elicit a brain response is a long-established method of testing brain imaging technology, and in an early study (Macé et al., Nature Methods, 2011) we found that fUS delivered very good sensitivity and spatial resolution for whisker stimulation in the rat.
Transcranial imaging of the mouse brain during whisker stimulation.
Transcranial imaging of the mouse brain during whisker stimulation.

Visual stimulation

The high spatial resolution of fUS allows the creation of detailed retinotopic maps of animals performing complex visual tasks, as shown in a pioneering study of the rat (see the image below).

This activation map on a rat model, obtained with fUS, shows the result of exposure to a flickering screen, and indicates significant activation in the primary visual cortices and superior colliculi. Reproduced from Gesnik et al., NeuroImage, 2017 (licensed under CC BY-NC-ND 4.0)
Using functional ultrasound, a dramatic increase in the activity of the supplementary eye field (SEF) and anterior cingulate cortex (ACC) is observed upon a change in saccade behavior in a macaque. Reproduced from Dizeux et al., Nature Communications, 2019 (licensed under CC BY 4.0)

Researchers have also shown how fUS can be used to map the regions involved in complex tasks and rule handling in the supplementary eye field. Notably, SEF activation was observed in a single trial, without averaging (see the figure above). fUS neuroimaging is now also being investigated in non-human primates for potential applications in brain–computer interfaces (BCIs).

In this study of a rat, a sequence of electrical signals was sent into the right sciatic nerve, and the response in the left somatosensory cortex monitored using fUS. Note that the responses are 20–30% above the baseline – much greater than the 3–5% typically seen with fMRI. Reproduced from Osmanski et al., Nature Communications, 2014 (licensed under CC BY 4.0)

Electrical stimulation

Electrical stimulation is another commonly investigated input, and the study here shows how the responses from fUS are much greater than for fMRI. This means that a single activation is sufficient to achieve good sensitivity, avoiding the need to average multiple experiments in order to increase the signal-to-noise ratio (as with fMRI).

Auditory stimulation

Researchers have also begun to study audition in awake ferrets, an excellent model for hearing research. 

Using fUS, they’ve been able to construct highly-detailed sonotopic maps (see the figure). As a result, they were able to obtain independent auditory response curves from voxels just 100 µm apart – an excellent demonstration of the resolution of fUS.

Exposure to pure tones of five frequencies resulted in areas of heightened activity, as shown in this view of a tilted parasagittal slice of the visual cortex (VC) and auditory cortex (AC) in the ferret. Reproduced from Bimbard et al., eLife, 2018 (licensed under CC BY 4.0)
Functional activation mapping using fUS was used to image the activity in the dorsal horn (DH) of the spinal cord in the rat, following stimulation of C-fibers in the hind paw. Note the very large hemodynamic response. Reproduced from Claron et al., Pain, 2021 (licensed under CC BY-NC-ND 4.0)

Pain stimulation

As well as being used to study the brain, fUS can also be used to study the response of the spinal cord to pain induced elsewhere in the body. This is very difficult with fMRI because of the movement due to breathing, but a combination of fUS neuroimaging and ultrasound localization microscopy (ULM) avoids this problem. Responses can therefore be mapped on an impressively small scale (see the figure), with obvious applications to spinal cord damage.

Natural behavior

Finally, pilot studies have shown that fUS probes can be used to measure brain-wide activity in freely-moving rats while exhibiting their natural behavior. This is in contrast to fMRI, which requires the animal to be restrained or anesthetized.

The following two examples – both from the same group – also show that fUS is compatible with EEG recordings, resulting in additional insights into brain activity.

By using fUS in combination with EEG, ‘vascular surges’ (strong but transient increases of CBV) during sleep phases were detected for the first time in rodents. This phenomenon had previously only been seen in humans, and suggests that it has an important evolutionary benefit. Reproduced from Bergel et al., Nature Communications, 2018 (licensed under CC BY 4.0)
This fUS–EEG study of a rat running back and forth in a corridor shows that locomotion is reliably followed by an increase in CBV in a sequence involving the retrosplenial cortex, dorsal thalamus and hippocampus. Interestingly, after 10–20 runs, blood volume in the cortex decreased sharply while the hippocampal responses increased linearly. This hemodynamic ‘reshaping’ might reflect the different neural processes involved in early-stage active exploration versus late-stage repetitive behavior. Reproduced from Bergel et al., Nature Communications, 2021 (licensed under CC BY 4.0)

Contact us

Would you like to discuss using fUS to carry out task-based neuroimaging in your research? Let’s set up a call.