Table 5.

Prediction performance of AI-100 on RDR, RMD, and GCS across cameras.

	Camera brand	Topcon	Syseye	iCare	Canon
RDR	Sensitivity	0.900 (0.555, 0.998)	1.000 (0.292, 1.000)	0.900 (0.555, 0.998)	1.000 (0.398, 1.000)
	Specificity	0.930 (0.875, 0.966)	0.967 (0.924, 0.989)	0.986 (0.950, 0.998)	0.968 (0.910,0.993)
	AUC	0.919 (0.781, 1.000)	1.000 (0.999, 1.000)	0.906 (0.721, 1.000)	0.992 (0.976, 1.000)
RMD	Sensitivity	0.982 (0.905, 1.000)	0.852 (0.729, 0.934)	0.982 (0.905, 1.000)	0.861 (0.705, 0.953)
	Specificity	0.907 (0.831,0.957)	0.970 (0.914,0.994)	0.885 (0.804, 0.941)	0.952 (0.865, 0.990)
	AUC	0.978 (0.956, 1.000)	0.966 (0.939, 0.993)	0.985 (0.972, 0.999)	0.968 (0.937, 1.000)
GCS	Sensitivity	0.909 (0.587, 0.998)	0.889 (0.518, 0.997)	1.000 (0.735, 1.000)	1.000 (0.292, 1.000)
	Specificity	0.944 (0.892, 0.975)	0.924 (0.867, 0.961)	0.957 (0.909, 0.984)	0.958 (0.896, 0.988)
	AUC	0.941 (0.826, 1.000)	0.973 (0.932, 1.000)	0.996 (0.989, 1.000)	0.993 (0.976, 1.000)

The numbers in the parentheses are lower and upper bounds of the respective 95% CIs.