Table 1.
Methods which used the Intra-patient paradigm.
| Work | # Classes | Feature set | Classifier | Effectiveness |
|---|---|---|---|---|
| Cristov & Bortonal, 200433 | 2 | Heartbeat-Intervals, VCG | NN | Acc = 99% |
| Özbay et al., 200634 | 10 | Raw-wave | MLP, Fuzzy Cluster, FCNN | Acc = 99% |
| Bortolan et al., 200735 | 2 | VCG and Morphological hyperbox + GA | Fuzzy Clustering | Acc = 99% |
| Ubeyli, 200736 | 4 | DWT | SVM, ECOC | Acc = 99% |
| Yu & Chen, 200712 | 5 | ICA, RR-interval | PNN | Acc = 99% |
| Yu & Chen, 200712 | 6 | Wavelet (statistics), RR-interval | PNN | Acc = 99% |
| Minhas & Arif, 200837 | 6 | Wavelet, RR-interval, PCA | kNN | Acc = 99% |
| Asl et al., 200838 | 6 | HVR, GDA | SVM | Acc = 100% |
| Chen et al., 20146 | 6 | RR-intervals | SVN, NN | Acc = 100% |
| Mert et al., 201439 | 6 | RR-intervals, HOS, | Bagged Decision Tree | Acc = 99% |
| 2nd order LPC coeff. | ||||
| Alickovic & Subasi, 201540 | 5 | autoregressive (AR) modeling | SVM, MLP, RBF, kNN | Acc = 99% |
| Li et al., 201741 | 6 | WPD | GA-BPNN | Acc = 99% |
Neural Network (NN); Principal Component Analysis (PCA); Generalized Discriminant Analyses(GDA); Error correcting output codes (ECOC); Independent Component Analysis (ICA); Probabilistic neural Network (PNN); Continues Wavelet Transform (CWT); Discrete Wavelet Transform (DWT); Discrete Cosine Transform (DCT); Higher order statistics (HOS); Linear Discriminants (LD); wavelet packet decomposition (WPD). Abbreviations: Acc: Accuracy.