Detrended Fluctuation Analysis of Arrhythmia: Scaling Exponent as an Index of Heart Wellness

Abstract

Well-established technologies to analyze biological signals including rhythmic heartbeat are available and accessible to scholars. However, stronger empirical evidence is required to justify the use of these technologies as practical tools in the field of biomedicine. Here we conducted analyses of heartbeat interval time series using an analytical technology developed across three decades—detrended fluctuation analysis (DFA)—to verify the power-law/scaling characteristics of signals that fluctuate in a regular, irregular, or erratic manner. We believe that DFA is a useful tool because it can quantify the heart condition by a scaling exponent, with a value of one (1) set as the default for a healthy state. This baseline value can be compared to a clinical thermometer, where the baseline is 37 °C for a physiologically healthy condition. Our study aimed to ascertain and confirm the utility of DFA in evaluating heart wellness, specifically in the context of studying arrhythmic heartbeat. We present case studies to confirm that DFA is a beneficial tool that quantifies the scaling exponent of a heart’s condition as “nonstationarily” beating and dynamically controlled. From an engineering perspective, we show that the heart condition can be classified into two typical categories: a healthy rhythm with a scaling exponent of one (1.0), and arrhythmia with a lower scaling exponent (0.7 or less).