Table 2.

Inter-reader variability of lesion matching versus the performance of the automated lesion matching method (auto) by disease cohorts. Data are reported as median (range). P-values are tests for significant differences between IRV and automated matching performance (Wilcoxon paired tests).

	Precision	Recall	F1 score	N d
NSCLC (N = 10)
IRV	0.97 (0.67, 1.00)	1.00 (0.50, 1.00)	0.98 (0.57, 1.00)	0.5 (0, 4)
Auto	0.92 (0.80, 1.00)	0.89 (0.71, 1.00)	0.91 (0.75, 1.00)	2.5 (0, 8)
p	0.74	0.26	0.40	0.18
Head and neck (N = 10)
IRV	1.00 (1.00, 1.00)	1.00 (1.00, 1.00)	1.00 (1.00, 1.00)	0 (0, 0)
Auto	1.00 (0.50, 1.00)	1.00 (0.33, 1.00)	1.00 (0.40, 1.00)	0 (0, 3)
p	0.11	0.11	0.11	0.08
Lymphoma (N = 10)
IRV	1.00 (0.25, 1.00)	1.00 (1.00, 1.00)	1.00 (0.40, 1.00)	0 (0, 3)
Auto	1.00 (0.92, 1.00)	1.00 (0.80, 1.00)	1.00 (0.86, 1.00)	0 (0, 4)
p	1.00	0.18	1.00	0.41
Advanced Cancers (N = 10)
IRV	0.92 (0.69, 1.00)	0.86 (0.74, 0.96)	0.89 (0.72, 0.96)	5.5 (2, 58)
Auto	0.88 (0.59, 1.00)	0.86 (0.60, 1.00)	0.87 (0.63, 1.00)	15.5 (0, 59)
p	0.24	0.95	0.86	0.53
ALL (N = 40)
IRV	1.00 (0.25, 1.00)	1.00 (0.50, 1.00)	1.00 (0.40, 1.00)	0 (0, 58)
Auto	0.97 (0.50, 1.00)	0.92 (0.33, 1.00)	0.94 (0.40, 1.00)	2 (0, 59)
p	0.14	0.05	0.12	0.06