Classification of Gait Patterns in Patients with Neurodegenerative Disease Using Adaptive Neuro-Fuzzy Inference System

Table 8

Performance comparison of several state-of-the-art methods for discriminating ND gaits from normal gaits.

	Features	Classifier	Evaluation method	Overall accuracy (%)
ALS vs.CO	Swing-interval turns count; averaged stride interval [1]	LS-SVM	LOO	89.66
	Entropy and coherence extracted from the wavelet approximation of the gait signal [6]	LDA	LOO	86.2
	ANFIS models for left and right stride interval, left and right stance interval, and double support interval (proposed)	Distance rule	LOO	93.10
PD vs.CO	Swing-interval turns count; gait rhythm standard deviation [7]	LS-SVM	LOO	90.32
	Constant RBF networks learned via deterministic learning [12]	Distance rule	LOO	87.1
	ANFIS models for left and right stride interval, left and right stance interval, and double support interval (proposed)	Distance rule	LOO	90.32
HD vs.CO	Entropy and coherence extracted from the wavelet approximation of the gait signal [6]	LDA	LOO	86.10
	Statistical features such as minimum, maximum, average, and standard deviation [30]	SVM	Random subsampling	90.28
	ANFIS models for left and right stride interval, left and right stance interval, and double support interval (proposed)	Distance rule	LOO	94.44
ND vs.CO	Entropy and coherence extracted from the wavelet approximation of the gait signal [6]	LDA	LOO	80.4
	Constant RBF networks learned via deterministic learning [12]	Distance rule	ATAT	93.75
	ANFIS models for left and right stride interval, left and right stance interval, and double support interval (proposed)	Distance rule	LOO	90.63

LS-SVM: least squares support vector machines. LDA: linear discriminant analysis. ATAT: all-training-all-testing. LOO: leave-one-out.