EURASIP Journal on Advances in Signal Processing

Table 3 Average accuracy of the classifiers (LDA, k-NN, SVM) using the proposed data extraction

From: Support system for classification of beat-to-beat arrhythmia based on variability and morphology of electrocardiogram

	Average accuracy %
	R-R	\(\sigma ^{2}_{X}\)	γ _X	κ _X	\(\sigma ^{2}_{X} + \gamma _{X}\)	\(\sigma ^{2}_{X} + \kappa _{X}\)	γ_X+κ_X	\(\sigma ^{2}_{X} + \gamma _{X} + \kappa _{X}\)
Arrhythmia
LDA	84.40	92.07	91.43	93.02	96.73	99.45	99.28	99.78
k-NN	84.40	92.07	91.43	93.02	96.73	99.45	99.28	99.78
SVM	84.40	92.07	91.43	93.02	96.73	99.45	99.28	99.78
Atrial fibrillation
LDA	85.70	94.73	93.45	95.52	97.72	99.97	99.78	100
k-NN	85.70	94.73	93.45	95.52	97.72	99.97	99.78	100
SVM	85.70	94.73	93.45	95.52	97.72	99.97	99.78	100

LDA linear discriminant analysis, k-NN k-nearest neighbors, SVM support vector machine, \(\left (\sigma ^{2}_{X}\right)\) variance, γ_X skewness, κ_X kurtosis

Back to article page