Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.

  

System Identification of a Cerebral Blood Flow Model

Author(s):

N. Müller
Conference/Journal:

Master Thesis, FS16
Abstract:

In the last two decades, extensive research was done on the subject of blood flow in the human brain. The goal is to better understand the self-regulating capability of the brain, called cerebral autoregulation. The gained knowledge allows the detection of malfunctioning in the human brain, therefore preventing permanent, severe brain damage. The goal of this thesis is to design a system identification algorithm, which generates the cerebral blood flow dynamics, using measurements of the blood pressure (BP). The relation between the BP and the cerebral blood flow velocity is known to be a high-pass filter where the low frequencies are damped in a not yet identified manner. To quantify the dynamics, system identification methods are used to find a suitable model. When applying these methods, the preprocessing of the data is crucial. Hence, in a first step, different preprocessing approaches are analyzed and optimized. Measurements often contain artifacts which need to be removed in the preprocessing as well. Both linear and nonlinear methods are considered in order to characterize the system in the course of this thesis. Further, these methods are compared by analyzing their prediction errors in the time and frequency domain. The obtained results show that in the frequency domain the best linear methods are subspace identification and Welch’s method. In fact, subspace identification yields a good estimate in the time domain as well which is why a polynomial nonlinear state space model is chosen as a nonlinear estimation method.

Supervisors: Tony Wood, John Lygeros, Roy Smith

Year:

2017
Type of Publication:

(12)Diploma/Master Thesis
Supervisor:



File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2017:IFA_5639
}
Permanent link