TABLE 3.

Comparison of the diagnostic efficacy of the multimodal combinatorial model, radiomics model, MoCA scores, and Symptom1 in the training, test, and validation sets.

Characteristics	Training set				Test set				Validation set
	AUC	Sensitivity	Specificity	p‐Value	AUC	Sensitivity	Specificity	p‐Value	AUC	Sensitivity	Specificity	p‐Value
Multimodal combinatorial model	0.842	0.857	0.745	NA	0.829	0.733	0.821	NA	0.860	0.889	0.714	NA
Radiomics model	0.819	0.743	0.766	0.290a	0.793	0.733	0.846	0.391a	0.828	0.889	0.667	0.657a
MoCA	0.666	0.571	0.713	p < 0.001b ***	0.638	0.600	0.615	0.048b *	0.642	0.722	0.524	0.002b **
Symptom1	0.579	0.914	0.245	p < 0.001c ***	0.582	0.933	0.231	p < 0.001c ***	0.532	0.111	0.952	p < 0.001c ***

Note: The markers a, b, and c represent the diagnostic performance comparison of the multimodal combinatorial model with the radiomics model, MoCA, and Symptom1, respectively.

Abbreviation: NA, not available.

^* p < 0.05.

^** p < 0.01.

^*** p < 0.001.