Table 1.

95% CI and P-value of Wilcoxon signed rank test comparing the DSC results generated by the individual observers.

	Heart		Esophagus		Spinal cord		Lung_L		Lung_R
	95% CI	P	95% CI	P	95% CI	P	95% CI	P	95% CI	P
A vs. B	0.937~0.947	0.937	0.794~0.843	0.754	0.858~0.890	0.937	0.954~0.968	0.814	0.965~0.974	0.239
A vs. C	0.934~0.946	0.530	0.798~0.838	0.695	0.861~0.888	0.814	0.953~0.968	0.308	0.964~0.974	0.117
B vs. C	0.933~0.946	0.388	0.797~0.838	0.754	0.860~0.892	0.638	0.953~0.967	0.754	0.963~0.973	0.530
A vs. D	0.914-0.939	0.004	0.746-0.808	0.002	0.866-0.887	0.136	0.952-0.966	0.158	0.960-0.972	0.136
B vs. D	0.913-0.936	0.002	0.745-0.807	0.005	0.864-0.894	0.239	0.952-0.964	0.308	0.959-0.971	0.272
C vs. D	0.912-0.933	0.004	0.748-0.804	0.002	0.867-0.890	0.530	0.952~0.964	0.308	0.958-0.970	0.480

A, B, and C are the three observers, D is the CNN-based auto-contouring.