Sleep Wakefulness Determinations From Heart Rate Data. Volume III.

Abstract

During the past years several projects have been conducted at the University of Texas at Austin by members of the Bio-Medical Engineering Program investigating the automated classification of levels of wakefulness. The primary design and goal of these projects was rapid, inexpensive determination of levels of wakefulness performed accurately using easily derived physiologic parameters. It was felt that by combining some of the procedures and results of previous studies with the procedures developed from the last two years of this research, a conglomerate algorithm which had the capabilities desired could be developed. During the third year of this research, an altered algorithm has been developed from previous algorithms to classify REM(+) - NREM sleep stages from minute-by-minute heart rate. The reclassification of the two training nights yielded accuracies of 51.30% and 63.68% for night one of LES and night one of FER, respectively. Accuracies from the remaining data of subject LES yielded 60.11% to 66.50%, of subject FER 45.99% to 63.68%. Subject OWN, whose data were not used in any training, yielded accuracies from 52.27% to 58.60%.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 31, 1977
Accession Number
ADA045817

Entities

People

  • A. J. Welch
  • L. E. Taylor
  • P. C. Richardson
  • T. P. Daubek

Organizations

  • University of Texas at Austin

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Accuracy
  • Cardiovascular Physiological Phenomena
  • Computers
  • Data Science
  • Databases
  • Detection
  • Digital Computers
  • Electrical Engineering
  • Engineering
  • Eye Movements
  • Frequency
  • Heart Rate
  • Information Science
  • Medical Engineering
  • Pattern Recognition
  • Physiological Phenomena
  • Regression Analysis

Readers

  • Circadian Sleep-Wake Regulation and Chronobiology
  • Computational Fluid Dynamics (CFD)
  • Regression Analysis.