Development of a Novel Deep Learning-Based Prediction Model for the Prognosis of Operable Cervical Cancer

Table 1

Patient demographic and clinicopathological characteristics.

Features	Characteristics	No.	%
Age (years)	Mean	43.0
	SD	13.0
Race	White	10927	78.3
	Black	1417	10.2
	American Indian	108	0.8
	Chinese	1379	9.9
	NA	121	0.8
Marital status	Single	3204	23.0
	Married	7340	52.6
	Separated	241	1.7
	Divorced	1636	11.7
	Widowed	901	6.5
	NA	630	4.5
Primary site	Endocervix	2520	18.0
	Exocervix	10978	78.7
	Overlapping lesion	454	3.3
Stage	IA	5967	42.8
	IB	5999	43.0
	IIA	662	4.7
	IIB	1243	8.9
	NA	81	0.6
Lymph node metastasis	No metastasis	11175	80.1
	Regional lymph node	1264	9.1
	Aortic/distant lymph node	202	1.4
	NA	1311	9.4
Positive lymph node numbers	0	11427	81.9
	1	1130	8.1
	2	530	3.8
	3	183	2.4
	4	334	1.3
		348	2.5
Resected lymph node numbers	Mean	10.8
	SD	13.7
Tumor diameter (mm)	Mean	23.8
	SD	33.1
Depth of invasion	Inner 1/3	4141	29.7
	Middle 1/3	1100	7.9
	Outer 1/3	2210	15.8
	NA	6501	46.6
Histology	Squamous	9971	71.5
	Adenocarcinoma	2501	17.9
	Other types	1480	10.6
Differentiation	Poor	3475	24.9
	Moderate	3505	25.1
	Well	1416	10.2
	NA	5556	39.8
Surgery	Local excision	2466	17.6
	TH	1632	11.7
		838	6.0
		1953	14.0
		7073	50.7

NA indicates not available.