Classification of Parkinson’s disease with freezing of gait based on 360° turning analysis using 36 kinematic features

Table 2 Model parameters of the 7 classifiers estimated by grid search

ML techniques	PDs vs. Cons (with 36 features)	PDs vs. Cons (with 5 features)	F vs. NF (with 36 features)	F vs. NF (with 6 features)
LR	C = 1.0	C = 10.0	C = 0.1	C = 1000.0
KNN	k = 2	k = 4	k = 6	k = 2
NB	–	–	–	–
LDA	n_components = 1	n_components = 1	n_components = 1	n_components = 1
QDA	reg_param = 0.5	reg_param = 0.3	reg_param = 0.3	reg_param = 0.001
SVM	C = 29.6, gamma = 0.001, kernel = rbf	C = 7.6, gamma = 0.1, kernel = rbf	C = 7.6, gamma = 0.01, kernel = rbf	C = 1e-5, gamma = 10.0, kernel = rbf
RF	max_depth = 10, n_estimators = 500	max_depth = 20, n_estimators = 500	max_depth = 20, n_estimators = 1500	max_depth = 30, n_estimators = 500

ML machine learning, PDs people with PD, Cons controls, F people with PD with FOG, NF people with PD without FOG, LR logistic regression, “C” is the inverse of regularization strength, KNN k-nearest neighbors, “k” is the number of neighbors, NB Naïve Bayes, LDA linear discriminant analysis, “n_components” is the number of components, QDA quadratic discriminant analysis, “reg_param” is the regularization of the per-class covariance, SVM support vector machine, “C” is the regularization parameter and “gamma” is the kernel coefficient, RF random forest, “n_estimators” is the number of trees in the forest and “max_depth” is the maximum depth of the tree

ISSN: 1743-0003