Multimodal ultrasound radiomics model combined with clinical model for differentiating follicular thyroid adenoma from carcinoma

Table 3 Diagnostic performance of different models in the training and test groups

Group	Model	ACC	AUC	95% CI of AUC	Sen	Spe	PPV	NPV	P’	p^
Training group	Clinic Signature	0.723	0.802	0.725–0.880	0.708	0.756	0.861	0.548	0.035*	0.777
	Rad Signature	0.773	0.788	0.706–0.869	0.792	0.733	0.864	0.623	0.007*
	Nomogram	0.830	0.867	0.800–0.933	0.854	0.778	0.891	0.714		0.007*
Test group	Clinic Signature	0.717	0.745	0.616–0.874	0.757	0.652	0.778	0.625	0.023*	0.639
	Rad Signature	0.767	0.787	0.671–0.904	0.811	0.696	0.811	0.696	0.430
	Nomogram	0.783	0.833	0.729–0.937	0.865	0.652	0.800	0.750		0.430

ACC: Accuracy; AUC: Area Under the Curve; CI: Confidence Interval; Sen: Sensitivity; Spe: Specificity; NPV: Negative Predictive Value; PPV: Positive Predictive Value; p’: p value compared with Nomogram, p^: p value compared with Rad Signature; *p < 0.05

ISSN: 1471-2342