Feature Selection Based on Machine Learning in MRIs for Hippocampal Segmentation

Table 5

Classification result on 57 features selected through Kolmogorov-Smirnov test using Näive Bayes Classifier using in -fold cross validation based on confusion matrix.

57 features; Kolmogorov-Smirnov test	Completeness of a class	Purity of a class	Contamination of a class
Hippocampus	84%	57%	43%
Not-hippocampus	52%	81%	19%
Efficiency	66%