ResNet-50 for 12-Lead Electrocardiogram Automated Diagnosis

Table 3

Results obtained by different research in relation to the proposed work.

Author	Year	Number of records	Model	Preprocessing	Number of classes	Accuracy (%)	Precision (%)
Antonio et al. [34]	2020	2,322,513	DNN	No	6		92.36
Ahsanuzzman et al. [35]	2020	48	LSTM and RNN	Yes	1	97.57
Obeidat et al [36]	2021	2000	CNN and LSTM	Yes	6	98.22	98.26
Adedinsewo et al. [38]	2020	6613	CNN	No	1	85.9	74
Xiong et al. [39]	2020	8528	ResNet-16	Yes	4	82
Dongdong et al. [19]	2021	6877	ResNet-34	Yes	9	96.6
Proposed work	2021	42,511	ResNet-50	Yes	27	97.63	89.67