Automatic Detection of Atrial Fibrillation from ECG Signal Using Hybrid Deep Learning Techniques

Table 5

Performance comparison of the proposed model with the existing works.

	S. No.	Year	Author	Methodology	F1 score (%)
A	1	2017	Manuel et al. [10]	Multiclass SVM	73
B	2	2017	Rajpurkar et al. [11]	The deep CNN model has 34 layers that map ECG signal samples into arrhythmia heartbeat classes.	79.9
C	3	2017	Coppola et al. [12]	Hierarchical classification model	78.55
D	4	2017	Neha et al. [13]	A LSTM network, which learns patterns directly from precomputed QRS complex features that classify ECG signals	78
E	5	2017	Schwab et al. [14]	Ensemble RNN with the LSTM attention model	79
F	6	2017	Andreotti et al. [15]	ResNet CNN	79
G	7	2017	Jiménez-Serrano et al. [16]	Feedforward neural network (FFNN)	77
H	8	2022	(present work)	CNN-ResNet model	80.58
				Hybrid-ResNet and LSTM (bidirectional)	80.08
				ResNet and RBF	80.20