Effective diagnosis of Alzheimer’s disease by means of large margin-based methodology

Table 2 Statistical measures of performance of LMNN-based techniques in comparison with other reported methods for SPECT database

SVM-linear classifier	Accuracy (%)	Sensitivity (%)	Specificity(%)
VAF	83.51	83.93	82.93
PCA	86.56	91.07	80.49
GMM	89.69	90.24	89.29
Gaussian kernel PCA+LMNN Transformation	91.75	91.07	92.68
Gaussian kernel PLS+LMNN Transformation	90.72	91.07	90.24
PLS+LMNN Transformation	92.78	91.07	95.12
LMNN-RECT	80.28	70	87.80
LMNN-Classifier Accuracy (%)	Euclidean	Mahalanobis	Energy
PCA	80.54	81.63	87.65
PLS	84.33	89.56	88.67

SVM classifier: Comparison of the methods reported in this work with VAF, GMM and PCA operation points. LMNN-based techniques parameters: linear SVM classifier with 6 components. VAF parameters: linear SVM classifier, GMM parameters: σ = 6 RBF-SVM classifier with 8 components and PCA parameters: σ = 6 RBF-SVM classifier with 16 components. LMNN Classifier with 6 components: Euclidean, Mahalanobis and Energy.

ISSN: 1472-6947