Table 8.

Evaluation metrics for assessing the performance of ECG beat type classification using the deep learning model. The table presents various statistical measures, including sensitivity, precision, F1-score, and overall accuracy, calculated for each beat type class.

Metric	N/A	N (Normal)	V (PVC)	A (PAC)	Macro AVG	Micro AVG
True positive	3	4244	108	90	1111.2	1111.2
False positive	1	1	5	4	2.75	2.75
False negative	0	8	1	2	2.75	2.75
True negative	4452	203	4342	4360	3339.2	3339.2
Precision	0.75	0.9998	0.9558	0.9575	0.9157	0.9975
Sensitivity	1	0.9981	0.9908	0.9783	0.9918	0.9975
Specificity	0.9998	0.9951	0.9989	0.9991	0.9982	0.9992
Accuracy	0.9975	0.9975	0.9975	0.9975	0.9975	0.9975
F1-measure	0.8571	0.9989	0.9730	0.9677	0.949	0.9975