The pulsing brain: state of the art and an interdisciplinary perspective

Abstract

Understanding the pulsing dynamics of tissue and fluids in the intracranial environment is an evolving research theme aimed at gaining new insights into brain physiology and disease progression. This article provides an overview of related research in magnetic resonance imaging, ultrasound medical diagnostics and mathematical modelling of biological tissues and fluids. It highlights recent developments, illustrates current research goals and emphasizes the importance of collaboration between these fields.

Keywords: brain tissue pulsation; magnetic resonance imaging medical diagnostics; mathematical modelling of biological fluids; mathematical modelling of biological tissues; ultrasound medical diagnostics.

Figure 1.

Graphical overview of MRI methods that are used to capture brain tissue motion. We distinguish between image-analysis-based methods and acquisition-based methods. Image-analysis-based methods aim to derive brain tissue motion from normal anatomical images that can in principle be acquired on any MRI scanner, whereas acquisition-based methods use dedicated MRI acquisition methods that are by design tuned to capture brain tissue motion and often need programming of the MRI acquisition software. To be able to use normal anatomical images for studying brain motion, the images are acquired as a time series, either synchronized to the heartbeat (‘cine’) or a real-time time series. Synchronization to the heartbeat often requires a special commercial software option, like the cardiac package on the MRI scanner, while real-time acquisitions require very fast imaging which is normally limited to a single slice. The output yielded by the various methods is indicated and varies from semi-quantitative displacement maps to truly quantitative maps of the displacement or velocity of the tissue, time-resolved over the cardiac cycle. It should be noted that both approaches can detect sub-voxel motions, i.e. displacements that are (much) smaller than the dimensions of a voxel (‘a 3D pixel’) in the image. Quantitative displacement or velocity maps can be used to subsequently compute tissue strains or strain rates by taking spatial derivatives.

Figure 2.

Brain tissue deformation as imaged by DENSE MRI. The maps represent the strain tensor that describes the tissue deformation induced by the heartbeat. Left, top: schematic depiction of a deformed voxel at peak systole relative to its shape at end-diastole. It illustrates how a voxel stretches along a certain one-dimensional direction (blue arrows), while at the same time, it shortens along another orthogonal direction (red arrows), which is known as the Poisson effect. Left, bottom: the RGB colour coding of directionality. Red: right to-left (RL); green: anterior-to-posterior (AP); and blue: feet-to-head (FH). Middle and right panels show the direction and magnitude of first principal strain (lengthening) and third principal strain (concomitant shortening perpendicular to the first principal strain) in each voxel for a healthy volunteer, respectively (unsmoothed data from DENSE measurements at 7 T MRI). Top: the direction of the eigenvector in each voxel using the directionally encoded colour scheme. Bottom: grey-scale maps representing the magnitude of the associated principal strains. For the third principal component, the absolute value of the eigenvalues was taken, resulting in a map with only positive values. Multiplying the eigenvectors with the eigenvalues results in an intensity colour map that summarizes both the direction and magnitude of the respective strain components. Figure adapted with permission from [, fig. 2] (CC BY-NC-ND 4.0).

Figure 3.

Illustrations of representative dynamic MRI techniques for capturing neurofluid pulsations, including dynamic diffusion-weighted imaging (dynDWI) for assessing the CSF pulsations in the perivascular subarachnoid space; functional MRI (fMRI) for capturing CSF inflow at the fourth ventricle; real-time and 4D PC-MRI for quantitatively measuring CSF flow at the C2/C3 level or the aqueduct; and balanced SSFP for semi-quantitatively measuring CSF flow along the ventricles and spinal canal. Abbreviations: ADC: apparent diffusion coefficient; FWHM: full width at half maximum; A.U.: arbitrary unit; sec: seconds; bSSFP: balanced steady-state free precession.

Figure 4.

Suggested framework for the development of imaging biomarkers for cerebral small vessel disease, highlighting the unmet needs in measuring and characterizing brain pulsatility. Accurate and biologically specific methods are still required to address gaps in understanding capillary pulsatility, distinguishing fluid volume from flow velocity pulsations at the microvascular level and identifying the origin of fluids from arterial, venous or perivascular compartments. Figure reproduced from [, fig. 1] (CC BY-NC-ND 4.0).

Figure 5.

The typical set-up of the TPI equipment on a participant, with measurements taken from the right temporal position (left). The colour map shows the amplitude of brain tissue motion in the direction of the ultrasound beam with the brain moving inward (towards the midline) in systole (right). The map covers an axial slice of the brain centred on the Willis polygon (x-axis: posterior–anterior, y-axis: lateral–medial). The set-up shown in the figure was used in the M-Pulse study reported in [154].

Figure 6.

An example of some of the information that could be available with ultrasound imaging methods if it were not for the aberrating and attenuating effects of the skull, or if eventually full correction for these effects were to become possible. This example shows a high-frequency ultrasound B-mode image (a), time integrated microvascular colour power Doppler image retrospectively overlaid on the B-mode image (b) and three example time sequences of pulsations in Doppler power in small blood vessels (c) for the regions of interest (ROI) drawn in yellow in (b). Images (courtesy of Professor Christopher Uff) were acquired during routine recovery investigation of a decompressive craniectomy following trauma by scanning through the skin overlying the window where a small patch of skull had been removed. A Canon Toshiba Aplio 500™ was employed, with an 18L7 probe and Superb Microvascular Imaging (SMI)™ software.