A methodology to ensure and improve accuracy of Ki67 labelling index estimation by automated digital image analysis in breast cancer tissue

Table 6 Single linear regression models with reference values as explanatory variable (n = 164, P <0.0001 for all models and slope estimates)

Variable	R-square	Intercept estimate	InterceptP	Slope estimate	Slope standardized estimate
Ki67-DIA-2	0.90	5.9692	<0.0001	0.7588	0.9494
Ki67-DIA-1	0.89	-0.6389	0.5324	0.7892	0.9447
Ki67-DIA-0	0.86	-0.9576	0.3389	0.6667	0.9278
Ki67-VE-median	0.86	-3.3799	0.0242	1.0093	0.9316
Ki67-VE-1	0.74	-8.1114	<0.0001	0.8057	0.8514
Ki67-VE-2	0.82	0.6733	0.7180	1.0616	0.9049
Ki67-VE-3	0.85	0.1337	0.9382	1.0926	0.9210
Ki67-VE-4	0.79	-2.7516	0.0987	0.8545	0.8545
Ki67-VE-5	0.71	0.4763	0.8294	0.9245	0.8422

ISSN: 1465-542X