A Novel Intelligent System for Brain Tumor Diagnosis Based on a Composite Neutrosophic-Slantlet Transform Domain for Statistical Texture Feature Extraction

Table 4

Classification results obtained by different combinations of GLDS, GLRLM, and GLCM features with various classifiers from NS, SLT, and composite NS-SLT methods, respectively. The accuracy in bold indicates the best classification result.

Features	Classifier methods	Techniques	Performance metrics
			Accuracy (%)	Precision	Sensitivity	Specificity	AUC
GLDS+GLRLM	DT-NN	NS
		SLT
		NS-SLT
	SVM-NN	NS
		SLT
		NS-SLT
	KNN-NN	NS
		SLT
		NS-SLT
	NB-NN	NS
		SLT
		NS-SLT
GLDS+GLCM	DT-NN	NS
		SLT
		NS-SLT
	SVM-NN	NS
		SLT
		NS-SLT
	KNN-NN	NS
		SLT
		NS-SLT
	NB-NN	NS
		SLT
		NS-SLT
GLRLM+GLCM	DT-NN	NS
		SLT
		NS-SLT
	SVM-NN	NS
		SLT
		NS-SLT
	KNN-NN	NS
		SLT
		NS-SLT
	NB-NN	NS
		SLT
		NS-SLT
GLDS+GLRLM+GLCM	DT-NN	NS
		SLT
		NS-SLT
	SVM-NN	NS
		SLT
		NS-SLT
	KNN-NN	NS
		SLT
		NS-SLT
	NB-NN	NS
		SLT
		NS-SLT