Automated Feature Extraction in Brain Tumor by Magnetic Resonance Imaging Using Gaussian Mixture Models

Table 2

Performance metrics (%) based on the PCA features.

Classifier	Sequence	Accuracy	False alarm	Missed detection
NB	T1-WI	73.52	8.82	17.64
	T2-WI	79.41	2.94	17.64
	FLAIR	82.35	5.88	11.76
	Entire GBM	68.62	15.68	15.68
SVM	T1-WI	55.88	0	44.11
	T2-WI	61.76	8.82	29.41
	FLAIR	85.29	0	14.70
	Entire GBM	35.29	20.58	44.11
PNN	T1-WI	61.76	8.82	29.41
	T2-WI	52.94	14.70	32.35
	FLAIR	94.11	0	5.88
	Entire GBM	51.96	9.80	38.23