Table 2.

95% CI and P-value of Wilcoxon signed rank test comparing the MSD results generated by 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	1.41~1.75	0.937	0.85~1.64	0.480	0.67~0.85	0.638	0.95~1.52	0.754	0.98~1.52	0.754
A vs. C	1.50~1.93	0.182	0.91~1.55	0.480	0.72~0.87	0.433	0.99~1.55	0.480	1.02~1.50	0.209
B vs. C	1.52~1.96	0.158	0.91~1.56	0.272	0.68~0.88	0.530	1.03~1.53	0.308	1.02~1.50	0.480
A vs. D	1.78-2.79	0.003	1.19-1.88	0.034	0.72-0.85	0.530	1.02-1.79	0.071	1.07-1.73	0.158
B vs. D	1.81-2.81	0.002	1.19-1.89	0.034	0.68-0.86	0.937	1.06-1.78	0.433	1.08-1.73	0.272
C vs. D	1.96-2.93	0.010	1.26-1.79	0.019	0.73-0.88	0.583	1.10-1.81	0.875	1.10-1.71	0.272

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