Evaluation of Methods for Estimating Fractal Dimension in Motor Imagery-Based Brain Computer Interface

Table 2

Computation times (time it took for feature extraction and classification) corresponding to the maximum classification accuracies obtained by the combination of fractal dimension estimation methods and classifiers with the two and three electrode configurations (and the parameters (k and k _max) used to obtain these values).

	Computation time (s)
	Katz’s method		Higuchi’s method		R/S method		Renyi’s entropy
	C3, C4	C3, Cz, C4	C3, C4	C3, Cz, C4	C3, C4	C3, Cz, C4	C3, C4	C3, Cz, C4
FKNN	0.17; (𝑘=9)	0.23; (𝑘=9)	1.03; (𝑘=7, 𝑘max=18)	1.5; (𝑘=5, 𝑘max=18)	7.37; (𝑘=9)	11.07; (𝑘=9)	2.35; (𝑘=9, 𝑞=4)	2.60; (𝑘=9, 𝑞=5)
SVM	0.34	0.34	1.07; (𝑘max=13)	1.4; (𝑘max=13)	7.36	10.99	1.90; (𝑞=2)	4.87; (𝑞=4)
LDA	0.12	0.21	0.83; (𝑘max=13)	1.26; (𝑘max=13)	7.32	10.99	1.84; (𝑞=3)	3.01; (𝑞=18)