Validating the accuracy of deep learning for the diagnosis of pneumonia on chest x-ray against a robust multimodal reference diagnosis: a post hoc analysis of two prospective studies

European Radiology Experimental

Table 4 ROC-AUC and statistical comparisons of diagnostic performances (PACSCAN cohort)

		Clinician	Radiologist	AI	CT
	ROC-AUC	0.635 (0.571–0.699)	0.717 (0.658–0.776)	0.735 (0.667–0.802)	0.919 (0.884–0.954)
Clinicians	0.635 (0.571–0.699)
Radiologist	0.717 (0.658–0.776)	p = 0.022
AI	0.735 (0.667–0.802)	p = 0.021	p = 0.683
CT	0.919 (0.884–0.954)	p < 0.001	p < 0.001	p < 0.001

ROC-AUC results are reported in percent along with their 95% confidence interval. Statistical comparison of two sets of predictions by area under the receiver operating characteristic curve using method from Sun and Xu. [reference #[31]. AI, Artificial intelligence; CT, Computed tomography; LDCT, Low-dose computed tomography; ROC-AUC, Area under the receiver operating characteristic curve