Table 4.

Studies of nasopharyngeal carcinoma (NPC) diagnosis using radiomics.

Author, Year, Reference	Image	Sample Size	Feature Selection	Modeling	Model Evaluation
Lv, W. (2018) [101]	PET/CT	106	Intra-class coefficient	LR with LOOCV	AUC 0.89
Du, D. (2020) [102]	PET/CT	76	MIM, FSCR, RELF-F, MRMR, CMIM, JMI, SC > 0.7	DT, KNN, LDA, LR, NB, RF, and SVM with radial basis function kernel	AUC 0.892

Logistic regression with leave-one-out cross-validation (LOOCV), mutual information maximization (MIM), Relief-F (RELF-F), conditional mutual information maximization (CMIM), Fisher score (FSCR), joint mutual information (JMI), Spearman’s correlation (SC).