Multiscale Residual Network Based on Channel Spatial Attention Mechanism for Multilabel ECG Classification

Table 5

Comparison results of different research work on the CCDD.

Literature	ECG categories	Classifier	Performance
[32]	2	CBRNN	Spe = 76.32%; Se = 75.52%; Acc = 87.69%
[44]	2	Ensemble deep learning	Spe = 86.86  3.51%; Se = 80.23  4.49%; Acc = 84.84  1.82%
[45]	2	LCNN	Spe = 83.84%; Se = 83.43%; Acc = 83.66%
[46]	2	Heart rate and LCNN fuse	Spe = 84.45%; Se = 85.19%; Acc = 84.77%
[47]	2	ResNet50	Spe = 91.63%; Se = 87.73%; Acc = 89.43%
[24]	7 multilabel	Ensemble multilabel classification model	Se (Rec) = 71.6%; Acc = 75.2%; Pre = 80.8%; F1 = 75.2%
This work	9 multilabel	CSA-MResNet	Spe = 98.7%; Se (Rec) = 85.9%; Acc = 97.1%; Pre = 90.6%; F1 = 88.2%