Repository logo
 

Non-Invasive Blood Pressure Measurement Algorithm for All Age Groups

Files

Whole thesis(2.08 MB)

Date

2012

Authors

Supervisor

Al-Jumaily, Ahmed
Lowe, Andrew

Item type

Thesis

Degree name

Master of Engineering

Journal Title

Journal ISSN

Volume Title

Publisher

Auckland University of Technology

Abstract

The oscillometric method is the most common technique used in commercial non-invasive blood pressure monitoring. In this technique, devices record the pressure oscillations in the cuff as the cuff pressure decreases from supra-systolic to sub-diastolic. The values for systolic pressure (SP) and diastolic pressure (DP) are determined by analyzing the shape of the oscillometric envelope. In many oscillometric algorithms, fixed percentile algorithms are used to determine SP and DP but their accuracy has been questioned. In this research an algorithm has been developed based on a beat-by-beat pattern recognition approach using time and frequency domain signal processing to extract features and an artificial neural network (ANN) is designed for classification of each beat as supra-systolic, sub-diastolic or in between. Normalized beat shape is successful at determining SP and DP and it also shows good agreement with recommended gold standard blood pressure auscultatory measurement.

Description

Keywords

Blood Pressure, Non-invasive, Oscillometric, Artificial neural network, Pattern recognition, Cuff Pressure

Source

DOI

Publisher's version

Rights statement

Permanent link

https://hdl.handle.net/10292/4551

Collections

Masters Theses