Skip to main content
. Author manuscript; available in PMC: 2015 Sep 1.
Published in final edited form as: Neuroimage. 2014 May 2;98:50–60. doi: 10.1016/j.neuroimage.2014.04.068

Table 3.

Comparison of different methods in terms of their classification performance (%) and correlation with a diagnosis phenotypic score, SRS. For logistic regression, only probabilistic output is used for ranking while the distance to the optimal hyperplane (wT x) is also considered for SVM.

Method SVM Logistic Regression PUNCH
Classification Performance 0.95 0.96 0.95
Method SVM SVM (wT x) Logistic Regression Clustering PUNCH
Correlation with SRS 0.42 0.63 0.46 0.34 0.76