Table 4.

Comparison of extreme learning machine with other two classifiers.

Method	Accuracy	Sensitivity	Specificity	AUC
SVM	79.87%	83.43%	75.54%	83.85%
Random forest	75.0%	82.84%	65.47%	81.99%
Extreme learning machine	79.87%	84.02%	74.82%	86.14%

Implementation of SVM was performed using third party library LIBSVM (https://www.csie.ntu.edu.tw/~cjlin/libsvm/), and the random forest was utilized with the third party library (http://code.google.com/p/randomforest-matlab). Both classifiers used the default settings.