Table 2.

Accuracy of different methods in each functional module.

Index	Fast-CNN				QRS based by P&T
	Se	PPV	Acc	F1	Se	PPV	Acc	F1
1	0.9953	0.9908	0.9863	0.9931	0.9958	0.9922	0.9881	0.994
2	0.9716	0.9941	0.966	0.9845	0.9857	0.9834	0.9695	0.9827
3	0.9752	0.9857	0.9616	0.9804	0.9079	0.9128	0.8354	0.9103
4	0.9953	0.9995	0.9948	0.9974	0.9995	0.9991	0.9986	0.9993
5	0.986	0.9754	0.9621	0.9807	0.9808	0.9586	0.941	0.9696
6	0.9645	0.9828	0.9484	0.9735	0.9774	0.9738	0.9524	0.9756
7	0.9919	0.986	0.9781	0.9889	0.9953	0.979	0.9745	0.9871
8	0.9881	0.9851	0.9736	0.9866	0.9902	0.9868	0.9772	0.9885
9	0.9827	0.9596	0.9437	0.971	0.9637	0.9529	0.9199	0.9583
10	0.9592	0.9929	0.9526	0.9895	0.9865	0.9924	0.9791	0.9757
11	0.9939	0.9747	0.9688	0.9842	0.9898	0.9099	0.9015	0.9482
12	0.9978	0.9974	0.9952	0.9976	0.9958	0.9908	0.9866	0.9933
13	0.991	0.9959	0.9869	0.9934	0.9943	0.9762	0.9708	0.9852
14	0.983	0.9862	0.9697	0.9846	0.9898	0.9728	0.9631	0.9812
15	0.9796	0.9572	0.9384	0.9682	0.985	0.9806	0.9662	0.9828
16	0.9402	0.9461	0.8924	0.974	0.9749	0.9731	0.9493	0.9432
17	0.9844	0.9818	0.9668	0.9831	0.9757	0.9636	0.941	0.9696
18	0.9489	0.9644	0.9168	0.9566	0.9758	0.9634	0.9409	0.9696
19	0.9815	0.9923	0.974	0.9869	0.9834	0.9814	0.9654	0.9824
20	0.9811	0.9907	0.9721	0.9858	0.9814	0.9816	0.9637	0.9815
AVR	0.9796	0.9819	0.9624	0.9807	0.9814	0.971	0.9542	0.9762

Convolutional Neural Network, CNN. Sensitivity, Se. Positive predictive value, PPV. Accuracy, Acc. F1- measure, change the value of F function by adjusting alpha, F1 when alpha = 1.