Table 2 Methods which used the Inter-patient paradigm.
From: Inter-Patient ECG Heartbeat Classification with Temporal VCG Optimized by PSO
Work | Feature set | Classifier | Effectiveness |
|---|---|---|---|
de Chazal et al., 20042 | ECG-Intervals, Morphological | Weighted LD | Acc = 83%; Se N = 87%; Se S = 76%; Se V = 77% + P N = 99%; + P S = 38%; + P V = 82% |
Soria & Martinez, 200942 | RR-Intervals, VCG, morphological + FFS | Weighted LD | Acc = 90%; Se N = 92%; Se S = 88%; Se V = 90% + P N = 85%; + P S = 93%; + P V = 92% |
Llamedo & Martinez, 20113* | Wavelet, VCG + SFFS | Weighted LD | Acc = 93%; Se N = 95%; Se S = 77%; Se V = 81% + P N = 98%; + P S = 39%; + P V = 87% |
Mar et al., 201143 | Temporal Features, Morphological, statistical features + SFFS | Weighted LD MLP | Acc = 89%; Se N = 89%; Se S = 83%; Se V = 86% + P N = 99%; + P S = 33%; + P V = 75% |
Ye et al., 201244 | Morphological, Wavelet, RR interval, ICA, PCA | SVM | Acc = 86.4% Se N = 88%; Se S = 60%; Se V = 81% + P N = 97%; + P S = 53%; + P V = 63% |
Lin & Yang, 201445* | normalized RR-interval morphological features | weighted LD | Acc = 93%; Se N = 91%; Se S = 81%; Se V = 86% + P N = 99%; + P S = 31%; + P V = 73% |
Huang et al., 201446** | Random projection RR-intervals | Ensemble of SVM | Se N = 99%; Se S = 91%; Se V = 94% + P N = 95%; + P S = 42%; + PV = 91% |
