ECG Segmentation and P-Wave Feature Extraction: Application to Patients Prone to Atrial Fibrillation

Abstract

This paper presents an automatic analysis method of the P-wave, based on lead II of a 12 lead standard ECG, which will be applied to the detection of patients prone to atrial fibrillation (AF), one of the most frequent arrhythmias. It focuses first on the segmentation of the electrocardiogram P-wave, which is performed in two steps: first, detection of the QRS complexes, then association of a wavelet analysis method and a hidden Markov model to represent one heat of the signal. After segmentation, the P-wave is isolated and a set of parameters, which have the ability to detect patients prone to AF, is calculated from it. The detection efficiency is validated on an ECG database of 145 patients including a control group and a study group with documented AF. A discriminant analysis is applied and the results obtained show a specificity and a sensitivity between 65% and 70%.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 25, 2001
Accession Number
ADA409557

Entities

People

  • Jean-christophe Cornilly
  • Jean-jacques Blanc
  • Jean-marc Boucher
  • Ronan Lepage

Organizations

  • Télécom Paris

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Automatic
  • Cardiac Arrhythmias
  • Classification
  • Coefficients
  • Computer Vision
  • Databases
  • Detection
  • Discriminant Analysis
  • Electrocardiography
  • Estimators
  • Health Services
  • Hidden Markov Models
  • Markov Models
  • Medical Personnel
  • Models
  • Physicians
  • Probability

Readers

  • Computer Vision.
  • Trauma or Military Medicine
  • Wave Propagation and Nonlinear Chaotic Dynamics.

Technology Areas

  • AI & ML