Research Article

A New Approach to Diagnose Parkinson’s Disease Using a Structural Cooccurrence Matrix for a Similarity Analysis

Table 1

Results from the best classifiers and combinations.

MeanderSpiralM/S
Acc (%)Acc (%)Acc (%)

Bayes
72.77 ± 9.9169.57 ± 11.0765.95 ± 2.05
75.11 ± 3.9976.36 ± 4.2668.38 ± 2.18
77.72 ± 4.2782.01 ± 5.5371.39 ± 2.39
OPF
75.54 ± 3.7669.73 ± 3.6364.28 ± 2.05
68.59 ± 3.7670.49 ± 3.6361.13 ± 2.57
77.50 ± 2.7575.71 ± 4.3967.21 ± 1.98
SVM
80.05 ± 1.8078.04 ± 2.7467.63 ± 2.73
75.00 ± 2.9078.04 ± 2.8565.30 ± 2.92
82.23 ± 3.0285.54 ± 3.6274.13 ± 2.27