Early prediction of chronic kidney disease based on ensemble of deep learning models and optimizers

Journal of Electrical Systems and Information Technology

Table 6 Performance evaluation of 6 months data produced by the three proposed individual models using different optimizers

Optimizer	Precision				Sensitivity				F1-score				Accuracy
Optimizer	CKD	Non-CKD	Macro avg	Weighted avg	CKD	Non-CKD	Macro avg	Weighted avg	CKD	Non-CKD	Macro avg	Weighted avg	Accuracy
CNN model
Adamax	0.93	0.96	0.95	0.95	0.96	0.93	0.95	0.95	0.95	0.95	0.95	0.95	0.95
Adam	0.89	0.93	0.91	0.91	0.93	0.88	0.91	0.91	0.91	0.91	0.91	0.91	0.91
SGD	0.79	0.70	0.75	0.75	0.65	0.83	0.74	0.74	0.71	0.76	0.74	0.74	0.74
Adadelta	0.64	0.82	0.73	0.73	0.89	0.50	0.70	0.69	0.74	0.62	0.68	0.68	0.69
Adagrad	0.71	0.85	0.78	0.78	0.88	0.64	0.76	0.76	0.79	0.73	0.76	0.76	0.76
LSTM model
Adamax	0.92	0.98	0.95	0.95	0.98	0.92	0.95	0.95	0.95	0.95	0.95	0.95	0.95
Adam	0.94	0.98	0.96	0.96	0.98	0.93	0.96	0.96	0.96	0.96	0.96	0.96	0.96
SGD	0.55	0.69	0.62	0.63	0.86	0.32	0.59	0.59	0.67	0.44	0.56	0.56	0.59
Adadelta	0.54	0.70	0.62	0.62	0.89	0.26	0.58	0.57	0.67	0.38	0.53	0.53	0.57
Adagrad	0.68	0.69	0.68	0.68	0.70	0.67	0.68	0.68	0.69	0.68	0.68	0.68	0.68
LSTM-BLSTM model
Adamax	0.95	0.99	0.97	0.97	0.99	0.95	0.97	0.97	0.97	0.97	0.97	0.97	0.97
Adam	0.95	0.97	0.96	0.96	0.97	0.95	0.96	0.96	0.96	0.96	0.96	0.96	0.96
SGD	0.72	0.60	0.66	0.66	0.45	0.83	0.64	0.64	0.55	0.70	0.62	0.63	0.64
Adadelta	0.60	0.74	0.67	0.67	0.84	0.44	0.64	0.64	0.70	0.55	0.62	0.62	0.64
Adagrad	0.65	0.74	0.70	0.70	0.80	0.58	0.69	0.69	0.72	0.65	0.68	0.68	0.69

The bold, underlined values represent the best optimizer's performance for each model