Intelligent Analysis of Premature Ventricular Contraction Based on Features and Random Forest

Table 7

Comparison with other literatures.

Methods	Recordings	Measures
		Acc (%)	PPV (%)	Se (%)	Sp (%)	γ (%)
Shyu et al. [27]	111, 115, 116, 119, 221, 230, 231	97.04	—	99.02	—	—
Li et al. [6]		98.0	66.0	99.7	—	—
Zarei et al. [5]		99.35	90.94	100	99.33	99.33
Proposed method		99.32	99.44	99.19	99.45	98.64
Lim [28]	115, 116, 119, 221, 230, 231	99.8	—	99.2	—	—
Li et al. [6]		99.7	79.0	99.6	—	—
Zarei et al. [5]		99.74	96.65	100	99.72	99.72
Proposed method		99.81	99.34	99.09	99.34	98.43
Llamedo and Martínez [8]	All recordings of DS2	98.16	87.97	82.94	99.21	82.15
Zhang et al. [9]		98.63	92.75	85.48	99.54	85.02
Zarei et al. [5]		98.77	86.48	96.12	98.96	95.08
Zhou et al. [7]		99.41	93.55	97.59	99.54	97.13
Proposed method		96.38	95.46	97.88	97.56	95.45