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

Table 5

Classification results for individual and combined texture features derived from SLT in the neutrosophic domain (composite NS-SLT). The accuracies in bold indicate the best classification result.

Statistical features	Classifier method
	DT-NN (%)	SVM-NN (%)	KNN-NN (%)	NB-NN (%)	Average accuracy (%)
GLDS	80.44	81.18	82.76	91.41	83.95
GLRLM	98.59	98.94	98.23	98.58	98.59
GLCM	96.10	97.63	97.65	95.29	96.67
Fusion of GLRLM and GLDS	98.23	98.92	97.88	98.57	98.40
Fusion of GLCM and GLDS	96.04	97.64	96.86	97.63	97.04
Fusion of GLCM and GLRLM	98.59	98.60	97.90	97.89	98.25
Fusion of GLCM, GLRLM, and GLDS	98.22	98.23	97.89	97.52	97.97