An Expert Diagnosis System for Parkinson Disease Based on Genetic Algorithm-Wavelet Kernel-Extreme Learning Machine

Table 4

The correct Parkinson diseases diagnosis performance comparing of the GA-WK-ELM method with classic ELM classifiers, which have different types of kernel function and the number of hidden neurons.

Used method	Type of the kernel function	Value of wavelet kernel parameter	Value of wavelet kernel parameter	Value of wavelet kernel parameter	The number of hidden neurons	Accuracy (%)
GA-WK-ELM	Wavelet	6	5	12	74	96.32
GA-WK-ELM	Wavelet	15	3	10	86	96.81
GA-WK-ELM	Wavelet	9	4	17	23	95.46
GA-WK-ELM	Wavelet	5	4	12	42	95.28
Classic ELM	Poly	—	—	—	76	89.31
Classic ELM	Hard limit	—	—	—	142	83.22
Classic ELM	Tangent sigmoid	—	—	—	164	91.75
Classic ELM	Tangent sigmoid	—	—	—	242	91.64
Classic ELM	Poly	—	—	—	56	83.85
Classic ELM	Radial basis	—	—	—	108	87.62
Classic ELM	Sigmoid	—	—	—	265	92.28
Classic ELM	Radial basis	—	—	—	356	91.54
Classic ELM	Radial basis	—	—	—	462	93.45