Table 7.

Intraclass Correlation Coefficient (ICC) and Standard Error of Measurement (SEM) scores between the ground truth and the AI models & human raters.

Model	Metric	Respect for Tissue	Time and Motion	Quality of Final Product	Overall Performance
Image Model	ICC(2,1)	-	-	0.888	-
	SEM(2,1)	-	-	0.257	-
Kinematic Model	ICC(2,1)	0.477	0.621	-	0.534
	SEM(2,1)	0.464	0.441	-	0.416
Multi-modal Model	ICC(2,1)	0.301	0.591	0.904	0.746
	SEM(2,1)	0.499	0.428	0.309	0.305
Rater 1	ICC(2,1)	0.717	0.779	0.823	0.616
	SEM(2,1)	0.476	0.414	0.512	0.502
Rater 2	ICC(2,1)	0.606	0.627	0.758	0.508
	SEM(2,1)	0.516	0.524	0.521	0.689
Rater 3	ICC(2,1)	0.797	0.797	0.924	0.789
	SEM(2,1)	0.377	0.423	0.308	0.379