Abstract:Rapid and accurate detection of atrial arrhymias is important for implantable devices to treat cardiac arrhythmia. Most of previous methods focus on a single feature of electrogram signals, resulting in a low accuracy. Several other methods utilize multi-features, resulting in a long detection time. A novel and computationally efficient method is proposed to identify sinus rhythm (SR), atrial flutter (AFL) and atrial fibrillation (AF). Here 23 features are firstly extracted from time-domain, frequency-domain and time-frequency domain of signals. Then, a column vector of 23 features is established for each rhythm which is viewed as a point in a 23-dimensional space. Since a high-dimensional space is not suitable for classification in terms of computational efficiency, the linear discriminant analysis (LDA) is adopted, not only to reduce the dimensionality to 2, but also to enhance the classification performance. Finally, the rhythm identification is conducted using a bivariate Bayesian decision rule. The performance of this algorithm is evaluated with two databases. One is the endocardial electrogram database obtained from the canine atrium surface, including 300 episodes of SR, AFL, and AF respectively. For each class of rhythm, 200 episodes are randomly selected as training data and the others as testing data. The other is the MIT-BIH arrhythmia database with 150 rhythms of SR, AFL and AF respectively. Here 100 segments are randomly selected as training data and the others as testing data. Results show that the overall classification accuracy of SR, AFL, and AF is respectively 100%, 99% and 98% for the canine database, and 99%, 99% and 98% for the MIT-BIH database, which has approximate 10% improvement compared with those of monovariate methods. Since the detection time of this algorithm is less than those of multi-feature methods, the implementation of this algorithm in implantable devices may provide speedy and accurate detection of atrial arrhythmias.

A variation rate measure for the arrhythmia via Poincaré plot

Cardiac Arrhythmia Classification Based on 3D Recurrence Plot Analysis and Deep Learning.

Research on Analyzing Approach of Mapping Plots of Heart Rate Variability Based on the First Order Difference Sequence.

Research on Analysis of Arrhythmia Based on pRRx Serials Derived from ECG Signals

[A Study on Clustering Analysis of Arrhythmias].

Temporal Variation Measure Analysis: An Improved Second-Order Difference Plot

Heart Rate Variability Analysis Based on Modified Poincaré Plot

Research on Algorithm for Detecting Atrial Fibrillation Using RR Intervals

Automatic Atrial Fibrillation Detection Based on Heart Rate Variability and Spectral Features

Analysis of heart rate variability using first order difference plot based on RR intervals

A novel technique for analysing beat-to-beat dynamical changes of QT-RR distribution for arrhythmia prediction

Comprehensive Analysis of Heart Rate Variability Features for Accurate Prediction of Paroxysmal Atrial Fibrillation.

Accurate Detection of Atrial Fibrillation Events with R-R Intervals from ECG Signals.

Rapidly detecting disorder in rhythmic biological signals: A spectral entropy measure to identify cardiac arrhythmias

An improved Poincaré plot-based method to detect atrial fibrillation from short single-lead ECG

Three-Dimensional Poincaré Plot Analysis for Heart Rate Variability

Improved Cardiac Arrhythmia Prediction Based on Heart Rate Variability Analysis

Statistical Analysis of HRV Parameters for the Detection of Arrhythmia

Mmarrhythmia: Contactless Arrhythmia Detection Via Mmwave Sensing

[Prediction of Recurrence of Paroxysmal Atrial Fibrillation Based on RR Interval].

Bivariate Bayesian Detection Of Atrial Arrhythmias Based On Dimensionality Reduction Of Multi-Feature Space