Establishment and Validation of a Machine Learning Prediction Model Based on Big Data for Predicting the Risk of Bone Metastasis in Renal Cell Carcinoma Patients

Table 1

Univariate and multivariate logistic regression for the risk of bone metastasis in patients with renal cancer.

Characteristics	Univariate logistic regression			Multivariable logistic regression
	OR	CI		OR	CI
Brain metastases
No	Ref	Ref	Ref	Ref	Ref	Ref
Yes	14.72	12.24-17.7	<0.001	2.46	1.98-3.05	<0.001
Grade
Well differentiated
Moderately differentiated	1.89	1.14-3.13	0.014	1.62	0.97-2.69	0.064
Poorly differentiated	5.84	3.57-9.56	<0.001	3.08	1.87-5.08	<0.001
Undifferentiated; anaplastic	13.92	8.48-22.84	<0.001	4.47	2.69-7.42	<0.001
Unknown	21.09	13.05-34.09	<0.001	7.97	4.9-12.97	<0.001
Liver metastasis
No	Ref	Ref	Ref	Ref	Ref	Ref
Yes	15.57	13.54-17.9	<0.001	2.37	2.01-2.8	<0.001
N
N0	Ref	Ref	Ref	Ref	Ref	Ref
N1	10.08	8.99-11.3	<0.001	2.18	1.9-2.51	<0.001
N2	4.47	2.88-6.94	<0.001	1.58	0.97-2.58	0.067
NX	4.88	4.14-5.75	<0.001	1.64	1.34-2.01	<0.001
Pulmonary metastasis
No	Ref	Ref	Ref	Ref	Ref	Ref
Yes	18.6	16.8-20.61	<0.001	5.2	4.58-5.89	<0.001
T
T1	Ref	Ref	Ref	Ref	Ref	Ref
T2	4.53	3.93-5.23	<0.001	2.13	1.81-2.5	<0.001
T3	3.78	3.34-4.27	<0.001	1.84	1.59-2.13	<0.001
T4	10.76	9-12.88	<0.001	2.08	1.68-2.59	<0.001
TX	18.08	15.14-21.59	<0.001	3.11	2.51-3.86	<0.001