A computational study of brain fluid flow within the human ventricular system / A. Aroussi, B. Howden and M. Vloeberghs

A three dimensional (3D) model of the Human Ventricular System (HVS) is presented to investigate the flow of Cerebrospinal Fluid (CSF) within the human brain, using Computational Fluid Dynamics (CFD). The role of CSF is complex but can be split into two main areas: a) acting as a buffer between the brain and the skull, b) acting as a transport medium for compounds and nutrients. CSF can be modelled as a Newtonian Fluid and its flow through the HVS can be visualized using CFD. In this investigation a 3D geometric model of the HVS (Fig. 1) was constructed from MRI data. It is the only model of its type to date. The flow of CSF within the HVS is a complicated phenomenon due to the complex HVS geometry. Previous modelling has only looked at individual parts of the HVS rather than as a whole. Understanding the nature of CSF flow allows engineers and physicians to design medical techniques and drugs to treat various HVS complications, such as hydrocephalus resulting from a tumor in the HVS. During this investigation CSF flow rate was set as 500ml/day, to mimic real life conditions, and secreted into the model via three inlets, with a uniform velocity. The model also contained three outlets, modelled as pressure outlets. The flow pattern of the CSF was found to vary with location, with the fastest flow being found in the Cerebral Aqueduct; a maximum velocity of 10mm/s was observed in this area. CSF pressure also varied with geometry and a negative pressure gradient was observed in the z-direction. The highest pressure drop also occurred through the Cerebral Aqueduct. The results of the investigation show that CSF flow is laminar and that the velocity is fastest within the Cerebral Aqueduct.
AIso CSF flow velocity is substantially slower in the areas of the HVS that are furthest away from the inlets; in some areas flow is virtually stagnated.