Multiple Sclerosis Slice Identification by Haar Wavelet Transform and Logistic Regression

Aim) Currently, scholars tend to use computer vision approaches to implement multiple sclerosis (MS) identification. (Method) In this study, we proposed a novel MS slice identification system, based on Haar wavelet transform, principal component analysis, and logistic regression. (Result) Simulation results showed the accuracies of our method using 2-level, 3-level, and 4-level decomposition are 83.251.62%, 89.721.18%, and 87.651.79%, respectively. (Conclusion) Our method with 3-level decomposition achieved the best.