Table 5.

Diagnostic performance and level of agreement in each Trial.

		Diagnostic performance			Cohen’s kappa	Kappa index	McNemar’s test
		Accuracy	Sensitivity	Specificity
Trial 1	Expert	0.81	0.61	0.91	0.54	Moderate	.001
	AI	0.80	0.52	0.94	0.51	Moderate	.000
Trial 2	Expert	0.69	0.57	0.93	0.42	Moderate	.000
	AI	0.73	0.97	0.23	0.25	Fair	.000
Trial 3	Expert	0.85	0.72	0.97	0.69	Substantial	.000
	AI	0.78	0.73	0.82	0.56	Moderate	.366

AI, artificial intelligence.