Research Article

ECG Signal Classification Based on Fusion of Hybrid CNN and Wavelet Features by D-S Evidence Theory

Table 8

Comparison of different classification approaches.

AuthorECG beatsApproachDatabasePerformance (%)

Mehrdad Javadi et al. [5]Normal (N)
Premature ventricular contraction (PVC)
Others
Mixture of experts (ME) and negatively correlated learning (NCL)MIT-BIHSPn = 98.01
SEpvc = 92.27
SEother = 93.72
Acc = 96.02

Mondéjar-Guerra V [12]Normal (N)
Supraventricular ectopic beat (SVEB)
Ventricular ectopic beat (VEB)
Fusion (F)
SVMsMIT-BIHSEN = 95.9
PPVN = 98.2
SESVEB = 78.1
PPVSVEB = 49.7
SEVEB = 94.7
PPVVEB = 93.9
SEF = 12.4
PPVF = 23.6
Acc = 94.5

Shi hangrui [32]Left bundle branch block (LBBB)
Right bundle branch block (RBBB)
Atrial premature beat (APB)
Premature ventricular contraction (PVC)
Learning vector quantization (LVQ)MIT-BIHAccAPB = 84.2
AccPVC = 92.6
AccLBBB = 77.8
AccRBBB = 81.4
Acc = 84

Wang Run [33]N
LBBB
RBBB
PVC
Radial basis function (RBF)MIT-BIHAccN = 89.7
AccLBBB = 97.1
AccRBBB = 96.7
AccPVC = 93

Yıldırım et al. [34]13 classes
15 classes
17 classes
CNNMIT-BIHSE13 = 93.52
SP13 = 99.61
PPV13 = 92.52
Acc13 = 95.2
SE15 = 88.57
SP15 = 99.39
PPV15 = 90.48
Acc15 = 92.51
SE17 = 83.91
SP17 = 99.41
PPV17 = 89.52
Acc17 = 91.33

Li et al. [35]AAMIRFMIT-BIHSEN = 94.67
PPVN = 99.73
SES = 20.00
PPVS = 0.16
SEV = 94.20
PPVV = 89.78
SEF = 50.00
PPVF = 0.52
SEQ = 0.00
PPVQ = 0.00
Acc = 94.61

Amrita Rana et al. [36]N
LBBB
RBBB
APB
PVC
LSTMMIT-BIHAcc = 95.00

Anika alim et al. [37]N abnormalSVM and ANNMIT-BIHAccSVM = 87
AccANN = 94

Sherin M. Mathews et al. [38]SVEB
VEB
Restricted Boltzmann machine (RBM) and deep belief networks (DBN)MIT-BIHAccSVEB = 93.63
AccVEB = 95.87

OursN
V
R
L
A
CNN and SVR by D-S evidence theoryMIT-BIHSEN = 99.49
SPN = 99.88
PPVN = 99.49
AccN = 99.49
SEV = 98.56
SPV = 99.87
PPVV = 99.52
AccV = 99.04
SER = 98.48
SPR = 99.88
PPVR = 99.49
AccR = 99.98
SEL = 100
SPL = 99.87
PPVL = 99.51
AccL = 99.76
SEA = 98.96
SPA = 99.38;
PPVA = 97.45
AccA = 98.20
Acc = 99.64