Skip to main content

Table 4 Sensitivity, specificity, and accuracy of the proposed methodology compared to the literature for arrhythmia and AF classification, evaluated for MIT-BIH Normal Sinus Rhyth, MIT-BIH Arrhythmia, and MIT-BIH Atrial Fibrillation databases

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

Autor (year)

Featureα

Classifierβ

SENS %

SPEC %

ACC %

Arrhythmia

This work (2018)

Voltage variation

LDA

99.64

99.91

99.78

  

k-NN

99.64

99.91

99.78

  

SVM

99.64

99.91

99.78

Mihandoost et al. (2018) [3]

Sparse decomposition

SVM

91.47

85.88

99.11

Raj et al. (2018) [42]

Spectral analysis

SVM, k-NN

91.47

85.88

99.11

Jovic et al. (2017) [6]

AlphEn. HRV

Random forest

91.10

97.01

91.20

Kim et al. (2016) [7]

HRV from 5s

SVM

91.69

Elhaj et al. (2016) [8]

PCA, DWT, ICA, HOS

NN

98.90

98.90

98.90

Martis et al. (2012) [23]

DWT, HOS

NN

98.61

98.41

94.52

Atrial Fibrillation

This work (2018)

P wave voltage variation

LDA

100

100

100

  

k-NN

100

100

100

  

SVM

100

100

100

Andersen et al. (2019) [24]

30 R-R

CNN, RNN

99.82

87.94

89.30

Xia et al. (2018) [25]

STFT, SWT

CNN

98.79

97.87

98.63

Kennedy et al. (2016) [26]

CoSEn+CV+RMSSD+MAD

Random forest

92.80

98.30

Orchard et al. (2016) [43]

P wave absence

Proposed algorithm

95.00

99.00

 

Petrenas et al. (2015) [27]

R-R

Threshold

97.10

98.30

Zhou et al. (2014) [28]

SD+SE

Threshold

97.53

98.26

98.16

  1. The symbol (–) represent the values not specified in the works
  2. AlphEn alphabet entropy, HRV heart rate variability, PCA principal component analysis, DWT discrete wavelet transform, ICA independent component analysis, HOS higher order spectra, CoSEn coefficient of sample entropy, CV coefficient of variance, RMSSDroot mean square of the successive differences, MAD median absolute deviation, SD symbolic dynamics, SE Shannon entropy, RR R-R intervals, STFT short-term Fourier transform, SWT stationary wavelet transform, LDA linear discriminant analysis, k-NN k-nearest neighbors, SVM support vector machine, NN neural network, CNN convolutional neural network, RNN recurrent neural network