[Retracted] Machine Learning Algorithms for Prediction of Survival Curves in Breast Cancer Patients

Table 3

Hospital variable quality analysis from value and 95% CI.

Variable	value Kaplan-Meier	Kaplan-Meier 95% CI	value Cox PH
Disaggregated age	<0.0001	0.576-0.999	0.4
Standard age	0.86	0.854-0.933	0.9
Age K-means	0.37	0.831-0.943	0.4
Disaggregated affected lymph nodes	<0.0001	0.725-0.96	0.005
Standard and paper affected nodes	<0.0001	0.771-0.942	<0.0001
Lymph nodes affected K-means	0.0063	0.843-0.929	0.01
Hierarchical lymph nodes	<0.0001	0.811-0.927	<0.0001
Lymph nodes removed disaggregated	0.0005	0.642-0.987	0.005
Hierarchical nodes	0.074	0.825-0.937	0.06
Disaggregated tumor size	<0.0001	0.648-0.982	0.1
Standard tumor size	0.07	0.838-0.935	0.08
Tumor paper size	0.027	0.823-0.939	0.03
Hierarchical tumor size	0.2	0.852-0.932	0.2
Tumor size K-means	0.07	0.837-0.936	0.07
Ki67 disaggregated	<0.0001	0.642-0.988	0.5
Ki67 standard	0.46	0.854-0.933	0.5
Ki67 paper	0.44	0.841-0.943	0.5
Ki67 hierarchical	0.68	—	0.7
Hormone receptor	0.79	0.855-0.931	0.8
Menopausal state	0.71	0.855-0.9331	0.7
Degree	0.43	0.842-0.935	0.4
Cancer subtype	0.29	0.854-0.933	0.3
Risk	0.0001	0.825-0.919	<0.0001
Hormone therapy	0.011	0.836-0.943	0.02