A robust Parkinson’s disease detection model based on time-varying synaptic efficacy function in spiking neural network

Table 4 Performance comparison for K-fold Cross Validation with different values of K for dataset #1

Technology	Accuracy (in %)	Sensitivity (in %)	Specificity (in %)	MCC	Precision	F1 Score	Gmean
K = 3
MLP-NN	94.36	89.03	96.79	0.8502	0.8860	0.8798	0.9262
RBF-NN	84.61	51.32	95.06	0.5450	0.7906	0.6125	0.6937
RNN	96.5	94	98.05	0.95	0.9988	0.9682	0.96
LSTM	93.62	80.87	99.96	0.91	100	0.893	0.899
SEFRON [Dataset#1]	99.49	96.97	100	0.9816	1	0.9841	0.9845
K = 5
MLP-NN	94.36	85.68	97.53	0.8373	0.8933	0.8641	0.9116
RBF-NN	83.08	49.62	93.61	0.4913	0.7110	0.5802	0.6743
RNN	96.67	95.83	98	0.9063	1.0	0.9787	0.969
LSTM	97.85	88.9	77	0.93	0.971	0.93	0.812
SEFRON [Dataset#1]	99.48	96	100	0.9763	1	0.9778	0.9789
K = 8
MLP-NN	94.35	83.54	98.09	0.8432	0.9292	0.8700	0.9019
RBF-NN	83.63	49.27	94.15	0.5208	0.7833	0.5978	0.6780
RNN	91.94	80.4	98.21	0.82	0.77	0.89	0.87
LSTM	95.52	92.84	98	0.9017	0.9622	0.93	0.943
SEFRON [Dataset#1]	98.96	95.63	100	0.9702	1	0.9756	0.9768
K = 10
MLP-NN	95.39	88.42	98.13	0.8756	0.93	0.8997	0.9294
RBF-NN	84.68	49.67	95.95	0.5270	0.7583	0.5946	0.6809
RNN	92.11	100	88.69	0.84	0.78	0.86	0.93
LSTM	93.87	84.56	98.33	0.90	0.9273	0.89	0.899
SEFRON [Dataset#1]	99.47	95	100	0.9687	1	0.9667	0.9708

ISSN: 1471-2377