A Neural Network for Estimation of Aortic Pressure from the Radial Artery Pressure Pulse

Abstract

A neural network is developed to estimate aortic pressure from the radial artery pressure pulse waveform. Invasively measured aortic and radial artery pressure in 51 adult subjects were used to train the network. Tests in a separate group of 21 subjects of similar age range showed a high correlation (r > 0.93) between measured and estimated systolic, diastolic and pulse pressure, with mean absolute errors (%) of 2.5 plus/minus 0.3, 3.5 plus/minus 0.6, 4.8 plus/minus 0.7 respectively. This method has potential applications in obtaining accurate estimates of central aortic pressure values from noninvasive radial artery pulse measurements. Such neural networks can be trained in specific subgroups (e.g. diabetics) to improve the estimation of central aortic pressure from the peripheral pulse.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 25, 2001
Accession Number
ADA410382

Entities

People

  • A. Avolio
  • A. Qasem
  • G. Frangakis

Organizations

  • University of New South Wales

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Arteries
  • Biomedical Engineering
  • Cardiovascular Physiological Phenomena
  • Cardiovascular System
  • Diseases And Disorders
  • Engineering
  • Frequency
  • Frequency Domain
  • Layers
  • Measurement
  • Neural Networks
  • Pressure Measurement
  • Risk Factors
  • Test Sets
  • Transfer Functions
  • Waveforms

Readers

  • Cardiovascular Physiology
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Neural Networks