Assessment of Electrocardiogram Rhythms by GoogLeNet Deep Neural Network Architecture

Table 10

Classification results compared with the state-of-the-art PVC and APC heartbeats in terms of Acc, Se, Sp, and Pp (percentage, %).

Method	PVC				APC
	Acc	Se	Sp	Pp	Acc	Se	Sp	Pp
Ince et al. [26]	98.3	84.6	98.7	87.4	97.4	63.5	99	53.7
Kiranyaz et al. [18]	99	93.9	98.9	90.6	97.6	60.3	99.2	63.5
Zhang et al. [19]	99.7	97.1	99.9	98.1	99.3	85.9	99.7	88.7
Luo et al. [25]	95.5	60.4	97.9	66.8	96.2	15.4	99.3	47.3
Proposed	98.64	98.40	98.70	94.90	98.31	88.75	99.40	94.40