DeepFHR: intelligent prediction of fetal Acidemia using fetal heart rate signals based on convolutional neural network

Table 7 Summary of related works conducted for the intelligent assessment of the fetal state using FHR signals obtained from CTG

Author	Database	Distribution (N/P)	Method			Performance(%)
Author	Database	Distribution (N/P)	Feature extraction	Feature selection	Classifier	Performance(%)
Krupa et al. 2011 [13]	Private	30/60	EMD	/	SVM	Acc:87 Se:95 Sp:70
Spilka et al.2012 [12]	Private	123/94	33 Set1, Set2, Set3	PCA,IG	NB,SVM,DT	Se:73.4 Sp:76.3 Fm:71.5
Czabanski et al. 2012 [14]	Private	146/43	7 Set1	/	WFS+ LS-SVM	Acc:92.0 QI:88.2
Fanelli et al. 2013 [15]	Private	61/61	2 Set3	/	ST	AUC:75
Xu et al. 2014 [40]	Private	255/255	64 Set1, Set2, Set3	GA	SVM	Se:83 Sp:66 AUC:74
Dash et al. 2014 [41]	Private	60/23	8 Set1	/	GM,NB	Se: 61 Sp:82
Spilka et al. 2014 [42]	CTU-UHB	175/377	33 Set1,Set2, Set3	/	LCA + RF	Se:72 Sp:78
Doret et al. 2015 [11]	Private	30/15	12 Set2, Set3	/	ST	AUC:87
Comert et al. 2016 [43]	CTU-UHB	60/40	18 Set1, Set2	/	ANN	Acc: 87.0 Se:88.7 Sp:85.1
Stylios et al. 2016 [44]	CTU-UHB	508/44	54 Set1, Set2, Set3	AUC	LS-SVM	Se:68.5 Sp:77.7
Comert et al. 2016 [16]	CTU-UHB	272/280	11 Set2, Set3	/	ANN	Acc: 92.40 Se:95.89 Sp:74.75
Georgoulas et al. 2017 [45]	CTU-UHB	508/44	33 Set1, Set2, Set3	AUC	LS-SVM	Se:72.12 Sp:65.30
Comert et al. 2018 [31]	CTU-UHB	439/113	IBTF	GA/	LS-SVM	Se:63.45 Sp:65.88
Li et al. 2018 [21]	Private	3012/1461	FHR + 1D CNN			Acc:93.24
Comert et al. 2018 [22]	CTU-UHB	508/44	STFT+2D CNN			Se:56.15 Sp:96.51 QI:73.61
Current work	CTU-UHB	447/105	CWT + 2D CNN			Acc:98.34 Se:98.22 Sp:94.87 QI:96.53 AUC:97.82

ISSN: 1472-6947