Abstracts Category : Other

Computational modelling of neurovascular coupling and the BOLD signal.

by Elshin Joel Mathias

Normal brain function is critically dependent on a continuous supply of blood flow so thateven a few seconds of disrupted blood flow can hinder brain function. Our body has employedtight regulatory mechanisms to maintain cerebral blood flow and avoid hyperemia or ischemia.One of the primary mechanisms in which the neurons communicate with the cerebral bloodvessels to increase local blood flow to the region in need is termed as neurovascular coupling.Disordered neurovascular coupling has been observed in several brain pathologies such ashypertension, stroke and Alzheimers disease. It is likely that the disruption of the interactionsbetween neurons and the cerebral vasculature contributes to the initiation and progressionof brain dysfunction. Hence understanding this coupling mechanism is of vital importance.Experimental exploration suggests that there are more than one factor controlling the vesseldilation during neural activation. Based on the observation that there are different vasoactiveions in different brain regions of the brain, researchers suggested that neurovascular couplingis brain region dependent. Another recent experiment illustrated that it is not just brainregion dependent but also specific to the functions in a certain region of the brain as the sameregion can be connected to many other regions. This suggests that neurovascular couplinghas to be explored based on the function that induces the vascular response. We have notyet arrived at a quantitative relationship of the cellular structures controlling neurovascularcoupling primarily due to the complexity of the parameters involved and the difficultyof measuring them in the highly heterogeneous brain. Hence, developing mathematicalmodels simultaneously based on experimental data which can be validated with repetitiveexperiments may help establish a quantitative relationship of the neurovascular couplingmechanism. To test the model built based on a hypothesis , the model must also be related towell established and accessible experimental technique. fMRI BOLD technique is one ofthe widely available non invasive experimental technique used to map the active regions inthe brain. Even though it is used to study the functions of the brain, the method employedprimarily utilizes the neurovascular response. So modelling the fMRI BOLD response basedon the neurovascular reponse will enable to simulate an output that can be compared with theexperimental data. The aim of this study is to model a certain hypothesis of neurovascularcoupling based on experimental data and use it to model the fMRI BOLD response and compare it to experimental data. A mathematical model was created based on differentexisting models to simulate the experimentally well supported K+ signalling mechanism ofneurovascular coupling and its associated fMRI BOLD response. Our model predicts thevariations of the BOLD response such as initial dip, positive and negative BOLD signals,post stimulus undershoot arising due to the