Block Term Decomposition of ECG Recordings for Atrial Fibrillation Analysis: Temporal and Inter-Patient Variability

Pedro Oliveira,Vicente Zarzoso

doi:10.14209/jcis.2019.12

Abstract

Responsible for 25% of strokes and 1/3 of hospitalizations due to cardiac related disturbances, atrial fibrillation (AF) is the most common sustained cardiac arrhythmia in clinical practice, considered as the last great frontier of cardiac electrophysiology. Its mechanisms are not completely understood, and a precise analysis of the atrial activity (AA) signal in electrocardiogram (ECG) recordings is necessary to better understand this challenging cardiac condition. Recently, the block term decomposition (BTD) has been proposed as a powerful tool to noninvasively extract AA in AF ECG signals. However, this tensor factorization technique was performed only in short ECG recordings and illustrated in single patients. To assess its performance and variability through different subjects, BTD is applied to a population of 10 AF patients in this paper. Also, its time variability is evaluated by means of long segments of AF ECG with varying observation window size. Experimental results show the consistency of BTD as an AA extraction tool, outperforming two well-known matrix-based methods in most of the observed cases for long and short AF ECG recordings.

Highlights

C ARDIAC arrhythmias are heart diseases characterized by abnormal electrocardiography waveforms
Its performance in a population of patients remains an open question. Aiming answer this open question and to provide results with more clinical relevance, this paper presents the performance of block term decomposition (BTD) in long atrial fibrillation (AF) ECG segments, as well as the analysis of the observation window size on its performance [14]
It can be seen that the maximum spectral concentration (SC) presents a high value (≥ 75%) for all the observed segments, and in all cases, superior to that of matrix techniques, meaning that, with the right initialization, BTD outperforms the matrix-based techniques for the considered window lengths

Summary

Introduction

C ARDIAC arrhythmias are heart diseases characterized by abnormal electrocardiography waveforms. Reducing life expectancy and increasing the healthcare bill, atrial fibrillation (AF) is the most common sustained cardiac arrhythmia encountered in clinical practice, especially affecting the eldery. Persistent AF represents a complex case of this arrhythmia, where extensive atrial remodelling has taken place due to sustained AF, significantly affecting atrial activity (AA) and AF perpetuation itself. The importance given to this challenging cardiac condition has increased in the last years and is expected to increase further, becoming a major public health and economical concern, as its mechanisms are complex and not completely understood. About 160 000 new patients are diagnosed with AF every year only in USA and similar numbers are reported in European countries. By 2050, it is estimated that AF will increase from 2.3 to about 12.1– 15.9 million patients only in USA, becoming a new epidemic [1].

Methods

Results

Conclusion