Feature extraction and classification of heart sound using 1D convolutional neural networks

EURASIP Journal on Advances in Signal Processing

Table 3 Results of the different numbers of convolution layers in 1D CNNs

Different layers	Convolutional kernel shape	Accuracy rate (%)	F-score (%)
3 layers	1×13	94.32	96.02
5 layers		95.21	97.31
3 layers	1×26	95.12	96.87
5 layers		96.33	98.65
3 layers	1×39	96.01	96.54
5 layers		99.01	99.10
3 layers	1×52	93.31	94.26
5 layers		94.24	94.89
3 layers	1×65	92.03	94.01
5 layers		94.32	95.23