The weights/biases of kSVM are set as the variables, and the median square error (MSE) of the samples are set as the fitness function of PSO.
Here w and b are weights/biases of kSVM, respectively.
Traditional method uses trial-and-error to determine the optimal values of error penalty C and kernel parameter [sigma] of kSVMs. It will cause heavy computation burden, and cannot guarantee to find the optimal or even near-optimal solutions.
The results showed that the proposed DWT + PCA + KSVM method obtains quite excellent results on both training and validation images.
It indicates that our proposed method DWT + PCA + KSVM with GRB kernel performed best among the 10 methods, achieving the best classification accuracy as 99.38%.
In this study, we have developed a novel DWT + PCA + KSVM method to distinguish between normal and abnormal MRIs of the brain.
The proposed DWT + PCA + KSVM with GRB kernel method shows superiority to the LIN, HPOL, and IPOL kernels SVMs.
This technique of brain MRI classification based on PCA and KSVM is a potentially valuable tool to be used in computer assisted clinical diagnosis.
Confusion matrix of our DWT + PCA + KSVM method (Kernel chose LIN, HPOL, IPOL, and.