Lung Cancer Classification Employing Proposed Real Coded Genetic Algorithm Based Radial Basis Function Neural Network Classifier

Table 6

Comparison of various lung cancer classification methods.

Image datasets	Metrics for image classification	Hybrid Logistic Regression-Artificial Neural Network Approach [6]	Hopfield Neural Network & Fuzzy Clustering Approach [7]	Back Propagation Neural Network Approach [8]	SRGWO-ELM Approach	Proposed; RCGA-RBFNN; Classifier; Approach
LIDC cancer datasets	Sensitivity	68.75	70.37	92.12	98.77	99.10
	Specificity	96.45	97.18	98.67	99.63	100.00
	Classification accuracy	88.12	89.07	91.11	96.72	98.36
	AUROC	0.9050	0.9010	0.9000	0.9800	1.00
Real time lung cancer datasets; (PSGIMSR)	Sensitivity	72.39	81.37	90.65	96.57	98.42
	Specificity	85.47	89.01	95.47	98.21	99.73
	Classification accuracy	84.69	87.65	90.65	95.69	97.31
	AUROC	0.7231	0.8721	0.8700	0.9758	0.9865