Characterization of coherent flow structures in brain ventricles

Halvor Herlyng, Shawn C. Shadden · 2026

Read Characterization of coherent flow structures in brain ventricles by Halvor Herlyng, Shawn C. Shadden (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-18849.txt.

Introduction

Herlyng and Shadden map coherent flow structures in brain ventricles. Cerebrospinal fluid moves on many scales, in space and in time. Snapshots alone make unsteady transport hard to read. Finite-time Lyapunov exponent fields give a time-integrated view instead. Models cover adult human and embryonic zebrafish ventricles built from imaging data. Human flow mixes heart pulsation, fluid secretion, and beating cilia. Ridges in the exponent fields mark coherent structures that match lab findings. Full Navier-Stokes runs resolve fine transport traits that Stokes flow misses.

Catalog record

CREATORHalvor Herlyng, Shawn C. Shadden
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE77688
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.18849

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Characterization of coherent flow structures in brain ventricles Halvor Herlyng, Shawn C. Shadden Abstract The dynamic flow of cerebrospinal fluid (CSF) in brain ventricles exhibits flow features on several scales, both spatially and temporally. Most analysis of this complex flow and the accompanying transport has used instantaneous (Eulerian) flow variables. Such analysis makes understanding of unsteady transport challenging. Here, we analyze brain ventricular CSF flow both in a Eulerian sense and from the Lagrangian perspective – a time-integrated view of the flow. With geometries generated from imaging data, we model CSF flow in adult human and embryonic zebrafish brain ventricles. In the human brain we model flow governed by cardiovascular pulsations, CSF secretion and motile cilia. The flow driven by cardiovascular pulsations is derived from a damped linear elastic model of brain ventricle deformations, as a result of applying displacement boundary conditions derived from experimental data. In the zebrafish brain we consider flow driven solely by motile cilia. The tissue and flow models are implemented and solved with finite element methods. We use the resulting velocity fields to compute finite-time Lyapunov exponent (FTLE) fields and use these fields to characterize Lagrangian coherent structures, which can be approximated by ridges in the FTLE fields. These coherent structures demonstrate prominent flow features in the brain ventricles congruent with findings in experimental research. In the human brain ventricles, we also investigate the role of inertia by comparing flow models governed by the Navier-Stokes and the Stokes equations. Comparisons show that solving the Stokes equations is adequate to compute integrated flow variables like stroke volumes, but that the Stokes approximation fails to resolve intricate features of flow and advective transport that are present in the solution to the Navier-Stokes equations, features that could be important to elucidating transport. 1 Introduction Brains are constantly in motion. These motions are tightly coupled to the cardiac and respiratory cycles, bodily movement and neural activity. This brain motion is a key contributor to cerebrospinal fluid (CSF) flow [ 29 , 6 , 89 ] . CSF is produced by choroid plexus in the brain ventricles [ 23 ] , from where the fluid flows into the spinal cord or to the subarachnoid space [ 68 ] . The flow of CSF maintains brain homeostasis, delivers nutrients and removes waste, and serves as a cushion for the brain [ 31 , 51 , 56 ] . Meanwhile, several pathologies afflicting the brain are related to disruption of CSF flow, such as hydrocephalus, where ventricular enlargement results from blockage of CSF flow [ 56 ] . The nature of CSF flow has been studied extensively in recent decades, both experimentally and computationally. Among the physiological processes that affect CSF flow are: blood vessel and respiration dynamics, secretion and absorption of CSF, beating of motile cilia, neural activity, and posture changes. [ 51 , 89 , 66 , 32 , 54 , 22 , 93 , 57 , 6 ] . With these physical processes occurring over various time and length scales, CSF flow also consists of motions on different scales [ 51 ] . Although CSF is fundamental to brain functioning, the underlying mechanisms of CSF flow are still not completely understood [ 51 ] . Because of the negative consequences of disruption in ventricular CSF flow, increasing our understanding of this flow is important to prevent or treat associated pathologies. In the brain ventricles, pulsations originating from cardiac and respiratory dynamics cause perturbations to blood volume and tissue deformation in the brain that result in CSF flow [ 28 , 94 , 57 , 5 , 36 , 93 , 51 , 35 ] . Back-and-forth motion of CSF with vortical flow features has been visualized in human brains [ 88 ] , and the pulsatility is clearly present in measurements of flow rates [ 5 , 57 , 93 ] . At smaller scales, visualizations of ventricular CSF flow trajectories in animals have demonstrated intricate features such as stagnation and separation points, and vortex-structured flows created by motile cilia [ 34 , 32 , 66 ] . Cilia-driven CSF flow has also to a lesser extent been studied in humans [ 95 , 97 , 82 ] , demonstrating local effects of cilia on CSF flow patterns. Flow patterns such as separatrices and vortices can be elucidated by special material structures known as Lagrangian coherent structures (LCS). The LCS constitute transport barriers in fluid advection, and organize stretching and folding in the mixing of passive tracers [ 40 , 39 ] . LCS can be characterized by computing finite-time Lyapunov exponent (FTLE) fields through integration of fluid parcel trajectories; ridges of these FTLE fields are good approximations of LCS [ 80 ] . Characterizing LCS thus provides a time-integrated view of fluid transport [ 80 , 40 , 39 ] . The explanatory power of LCS has previously been demonstrated in a host of applications, including blood flow [ 81 , 77 ] , atmospheric flow [ 38 ] , flow in airways [ 58 ] and oceanic flows [ 10 ] . Despite the predictive power of studying flow from a Lagrangian perspective–where fluid parcel trajectories are tracked over time–only a limited amount of previous ventricular CSF flow studies have considered this perspective [ 97 ] . Most studies are rather based on flow analysis with an Eulerian perspective [ 54 , 17 , 45 , 90 , 82 ] , in which flow variables such as velocity and volumetric flux are measured at a fixed point in space and time. Instantaneous flow variables may however deceive interpretation of time-dependence in the flow [ 80 ] , especially in flows that contain motion at multiple scales, such as ventricular CSF flow. Therefore, further knowledge of flow and transport in brain ventricles, and the precise contributions of specific flow mechanisms such as cardiovascular pulsations, CSF secretion, and beating of motile cilia to mixing and transport of CSF, could be gained from Lagrangian analysis. For example, do LCS exist in brain ventricular CSF flow, and if so, how are the LCS organized? How are the LCS affected by flow mechanisms such as cardiovascular pulsations, motile cilia and CSF secretion? Does fluid inertia play a significant role in establishing the LCS and thus the advective transport within the brain ventricles? In this work, we characterize LCS in brain ventricular CSF flow using two computational models generated with imaging data. One geometry represents the brain ventricles of an adult human. The other geometry represents the brain ventricles of embryonic zebrafish, an animal whose ventricular system shares many similarities with the human brain ventricles. Using finite element methods, we simulate CSF flow in these two geometries. The CSF velocity fields are used to compute particle trajectories. These trajectories are used to compute finite-time Lyapunov exponent (FTLE) fields, which we use to characterize LCS. In the human brain, we model flow driven by cardiovascular pulsations, choroid plexus secretion and motile cilia, and study the impact of each mechanism on flow patterns and advective transport. The results indicate the presence of LCS that are mostly determined by cardiovascular pulsations. For the zebrafish brain, we model flow driven by motile cilia, and demonstrate that FTLE

Read the complete text