Preliminary Study on the Efficient Electrohysterogram Segments for Recognizing Uterine Contractions with Convolutional Neural Networks

Table 3

Test results with EHG segments of different durations.

Different durations	Resulting values				Parameters
	FP	FN	TP	TN	SE	SP	ACC
10_L6	107	793	263	949	0.25	0.90	0.57
10_L5	115	777	279	941	0.26	0.89	0.58
10_L4	162	534	522	894	0.49	0.85	0.67
10_L3	287	368	688	769	0.65	0.73	0.69
10_L2	190	308	748	866	0.71	0.82	0.76
10_L1	115	183	873	941	0.83	0.89	0.86
10_R1	124	206	850	932	0.80	0.88	0.84
10_R2	135	259	797	921	0.75	0.87	0.81
10_R3	261	349	707	795	0.67	0.75	0.71
10_R4	151	517	539	905	0.51	0.86	0.68
10_R5	268	887	169	788	0.25	0.84	0.55
10_R6	182	844	212	874	0.20	0.83	0.51
20_L3	241	726	330	815	0.31	0.77	0.54
20_L2	160	344	712	896	0.67	0.85	0.76
20_L1	130	108	948	926	0.90	0.88	0.89
20_R1	176	205	851	880	0.81	0.83	0.82
20_R2	169	371	685	887	0.65	0.84	0.74
20_R3	198	784	272	858	0.26	0.81	0.54
30_L2	210	687	369	846	0.35	0.80	0.58
30_L1	135	216	840	921	0.80	0.87	0.83
30_R1	154	238	818	902	0.77	0.85	0.81
30_R2	198	750	306	858	0.29	0.81	0.55