Parametric model derivation of transfer function for noninvasive estimation of aortic pressure by radial tonometry

Aortic pressure can be estimated noninvasively by applying a transfer function (TF) to radial tonometry signals. This study compares the performance of prior approaches, based on Fourier transform and inverted aortic-to-radial model, with direct radial-to-aortic autoregressive exogenous (ARX) model....

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on biomedical engineering 1999-06, Vol.46 (6), p.698-706
Main Authors: Fetics, B., Nevo, E., Chen-Huan Chen, Kass, D.A.
Format: Article
Language:English
Subjects:
Adult
Aged
Algorithms
Aorta - physiology
Bias
Biological and medical sciences
Biomedical measurements
Blood
Blood pressure
Blood Pressure Determination - methods
Calibration
Cardiology
Cardiovascular system
Female
Fourier Analysis
Fourier transforms
Frequency
Humans
Investigative techniques of hemodynamics
Investigative techniques, diagnostic techniques (general aspects)
Male
Mathematical models
Medical sciences
Middle Aged
Models, Cardiovascular
Parametric statistics
Physiological models
Predictive Value of Tests
Pressure
Pulse measurements
Radial Artery - physiology
Regression analysis
Reproducibility of Results
Signal Processing, Computer-Assisted
Steady-state
Supine Position
Systole
Transfer functions
Transient analysis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Aortic pressure can be estimated noninvasively by applying a transfer function (TF) to radial tonometry signals. This study compares the performance of prior approaches, based on Fourier transform and inverted aortic-to-radial model, with direct radial-to-aortic autoregressive exogenous (ARX) model. Simultaneous invasive aortic pressure and radial tonometry pressure were recorded during rest in 39 patients in the supine position. Individual radial-aortic TF's were estimated from 20 patients, and the average TF was used to predict aortic pressures in the remaining 19 patients. The direct average TF yielded accurate aortic systolic pressure estimation (error 0.4/spl plusmn/2.9 mmHg) and good reproduction of the aortic pressure waveform (root mean squared error 2.2/spl plusmn/0.9 mmHg). The inverted reverse TF (aortic radial) yielded comparable results, while the Fourier-based TF had worse performance. Individual direct TF provided improved predictive accuracy only for indexes which are based on higher frequency components of the waveform (augmentation index, systolic time period). An ARX average TF can be used to accurately estimate central aortic pressure waveform parameters from noninvasive radial pulse tracings, and its performance is superior to previous techniques.
ISSN:0018-9294
1558-2531
DOI:10.1109/10.764946