Resting-state functional connectivity

In an analogous fashion to fMRI, functional ultrasound can be used to map intrinsic brain connectivity (often called resting-state functional connectivity or resting-state networks), by detecting correlated fluctuations of spontaneous blood flow.

The high spatial and temporal resolution of fUS, and its much greater inherent sensitivity than BOLD-based fMRI, makes it a powerful way of studying neuropsychiatric diseases, and potentially aiding early diagnosis. Below are a few examples of fUS in action.

2D connectivity matrices

Generating connectivity matrices in two dimensions involves investigating responses from a single coronal plane, as illustrated here for a mouse model. This task is difficult using fMRI, because the small size of the mouse brain demands expensive, high-field fMRI scanners, and because maintaining stable physiological parameters (and hence efficient neurovascular coupling) is very difficult in anesthetized mice. In addition, the reliability of a connectivity matrix depends critically on accurately defining the ‘region of interest’ (ROI). Doing this using Iconeus One is quick and reliable, because it allows automated correlation to the Allen Mouse Brain Atlas, meaning results are much less vulnerable to bias.
This 2D resting-state functional connectivity study shows local and inter-hemispheric correlations at approximately –1.0 mm from Bregma in a mouse model – which would have been challenging to achieve using fMRI. Iconeus
This 2D resting-state functional connectivity study shows local and inter-hemispheric correlations at approximately –1.0 mm from Bregma in a mouse model – which would have been challenging to achieve using fMRI.

This 3D connectivity matrix for a mouse model was acquired non-invasively in 20 minutes, and shows strong interhemispheric connectivity patterns between four slices, with correlation coefficients up to 0.8. Reproduced from Bertolo et al., Journal of Visualized Experiments, 2021 (licensed under CC BY-NC-ND 3.0)

3D connectivity matrices

3D connectivity matrices are simply an extension of the 2D approach mentioned above – in essence, you’re determining the overall connectivity between (usually) closely-spaced slices. The fast switching between planes used in Iconeus One shortens the whole process to 20 minutes or less.

Seed-based correlation mapping

Another way of visualizing brain activity relationships is by correlating voxel activity across the whole brain with the activity of a small ‘seed’ region – known as seed-based (or ROI-based) functional connectivity.

This approach is not so dependent on the segmentation of the brain into functional regions, since you only need to define a single region.

Seed-based correlation mapping was used here in a mouse model, with each panel showing the brain regions correlated with a single ‘seed’ region (green arrow). Iconeus
Seed-based correlation mapping was used here in a mouse model, with each panel showing the brain regions correlated with a single ‘seed’ region (green arrow).
The two matrices on the left show the strength of correlations between fluctuations in CBV across 10 brain regions, and indicate a profound alteration in functional connectivity between the control (left) and a rat model of sustained inflammatory pain (center). On the right is a significance matrix, with white squares showing pairs of ROIs that differ significantly (and their p-values). Reproduced from Rahal et al., Scientific Reports, 2020 (licensed under CC BY 4.0).

Functional connectivity in pathological models

A different manifestation of functional connectivity is looking at how differences relevant to animal pathologies influence connectivity across the whole brain.

For example, functional ultrasound has been used to study the role of oxytocin in rat pups (Mairesse et al., Glia, 2019) and to investigate sensitivity to inflammatory pain in anesthetized rats, as shown here.

Investigating connectivity in moving animals

Connectivity matrices are normally developed for anesthetized animals, but an important development is using the light, robust Iconeus One probes to obtain results from awake mice (freely-moving or head-fixed), eliminating the bias of anesthetics.

As shown in the example below, this capability allows the effects on brain activity of anesthetics themselves to be studied, and it may also be relevant for analyzing genetically modified mice models.

In this functional connectivity study of an awake mouse, the strength of the relationship is directly proportional to the level of wakefulness (using medetomidine sedation and atipamezol reversal). Reproduced from Ferrier et al., Proceedings of the National Academy of Sciences, U.S.A., 2020 (licensed under CC BY-NC-ND 4.0)

Contact us

Interested by what you’ve read about functional activation mapping using fUS? Talk to one of our specialists about your application.

Get in touch