Supplementary Table S2: Hurst exponents of all IMFs of arrhythmic and normal subjects estimated by R/S technique.

Empirical Mode Decomposition (EMD) technique was applied to each of the filtered ECG signal (Supplementary Table S1) decomposing them into a finite number of Intrinsic Mode Functions (IMFs) following the sifting process. To retrieve the maximum number of IMFs, the number of iterations was set to 150. The number of IMFs retrieved from most of the ECG signals was 12. However, it ranges between 10 and 14. The values of Hurst Exponent (H) of all IMFs of the ECG time-series were estimated by R/S techniques in order to quantitatively assess the regularity and scaling of the data. The estimated H values for each IMFs corresponding to diseased and normal patients’ ECG data are tabulated in Supplementary Table S2.

经验模态分解（Empirical Mode Decomposition, EMD）技术被应用于每一路经滤波处理的心电图（Electrocardiogram, ECG）信号（详见补充表S1），通过筛分过程将其分解为有限个本征模函数（Intrinsic Mode Functions, IMFs）。为获取最大数量的本征模函数，迭代次数设置为150次。多数心电图信号可分解得到12个本征模函数，但其数量介于10至14之间。为定量评估数据的规律性与标度特性，采用R/S分析法估算了心电图时序所有本征模函数的赫斯特指数（Hurst Exponent, H）值。针对患病与正常患者的心电图数据，各本征模函数对应的估算赫斯特指数值已汇总于补充表S2中。