Cardiac Tachyarrhythmia Detection by Poincaré Plot-Based Image Analysis

Abstract

Tachyarrhythmia detection through RR interval analysis could improve performance of monitoring devices. In this paper a Poincaré plot-based image approach is presented. Three cardiac rhythms were analyzed in this study: normal sinus rhythm (NSR), atrial fibrillation (AF) and atrial bigeminy (AB). Using different MIT-BIH databases, 27955, 3363 and 76 images were generated for NSR, AF and AB respectively using a 2-minute window with 50 % overlap. The 80 % of the data available for each rhythm was used to create a reference rhythm image atlas. The remaining 20 % was classified into one of the three categories using mutual information. The process was iterated 10 times, in which images used to construct the atlas and used to create the test set were randomly selected. AF was correctly classified 94.12 %±0.45, AB 72.00 %±11.24 and NSR 80.70 %±0.54. The results of the present study suggest that Poincaré plot-based image analysis is a promising path for classifying different rhythms using only ventricular activity.

Poincaré Plot Image and Rhythm-Specific Atlas for Atrial Bigeminy and Atrial Fibrillation Detection

“…Preliminary results considering only a type of Poincaré Image, bin size and time window as well as a simpler classification metric, were presented in [22]. In this work, the concepts of Poincaré Image and Poincaré Atlas are extensively assessed and the parameters influencing their computation are optimized for obtaining better classification results.…”

“…In this work, the concepts of Poincaré Image and Poincaré Atlas are extensively assessed and the parameters influencing their computation are optimized for obtaining better classification results. The improvements introduced in this work with respect to [22] include: the exploration of the bin size influence on Poincaré Images and Altases, the study of different types of Poincaré Images configuration, the optimization of the distance metric between Poincaré Images and Atlases and finally the assessment of the methodology's performance for reduced time windows that range from 120 s to 20 s (in [22] only a time window of 120 s was used). The major contribution of this study is thus, the definition of Poincaré Images and Atlases and the demonstration of their potential to detect and classify different cardiac rhythms with segments as short as 20 s.…”