Table 8.

Comparison of classification performance: image-level vs. exam-level evaluation.

Evaluation level	Accuracy	Precision	Sensitivity	Specificity	F1 score	AUROC
Image-level Classification (n = 213)	0.925	0.915	0.915	0.933	0.915	0.980
Exam-level evaluation (n = 101)	0.941	0.872	0.971	0.924	0.919	0.994
(95% Confidence interval)	(0.89–0.98)	(0.75–0.97)	(0.91–1.00)	(0.85–0.98)	(0.84–0.98)	(0.98–1.00)

Bold values indicate the highest performance for each classification metric (column).