. 2019 Aug 26;33(1):121–130. doi: 10.1007/s10278-019-00260-w

Table 4.

Comparison of inter-annotator agreement vs. the ERT machine learning model

	Accuracy	FAVG	$F_{\bar{FU}}$	F_FU	$P_{\bar{FU}}$	$R_{\bar{FU}}$	P_FU	R_FU
Ann1–Ann3	0.812	0.795	0.736	0.853	0.974	0.593	0.752	0.987
ERT model	0.765	0.753	0.699	0.807	0.806	0.617	0.744	0.882
Baseline	0.581	0.551	0.436	0.666	0.778	0.304	0.529	0.900