[Retracted] Predicting the Spread of Vessels in Initial Stage Cervical Cancer through Radiomics Strategy Based on Deep Learning Approach

Table 1

A selected patient training and testing phase attributes.

Patients characteristics	Training phase N = 150		-value	Testing phase N = 55		-value	-value
	+ive lymphovascular invasion	-ive lymphovascular invasion		+ive lymphovascular invasion	-ive lymphovascular invasion
Patients age/year			0.60			0.53	0.98
Average age	55	56		53	60
Age ranges	27–55	27–60		30–55	35–65
Stages			0.62			<0.0002	0.45
Early stage IB	20 (50.2)	40 (52.6)		15 (30.4)	35 (70.2)
Late stage IB	15 (42.5)	48 (50.1)		18 (40.2)	20 (40.2)
Stage IIB	8 (18.5)	12 (13.2)		12 (52.6)	8 (10.2)
MRI lymph node status			<.002			0.002	0.70
Positive	20 (7.9)	30 (40.2)		10 (55.2)	50 (92.1)
Negative	150 (95.7)	52 (68.2)		12 (60.8)	15 (20.5)
Menstrual status			0.542			0.442	0.89
Postmenopausal	15 (40.3)	55 (56.2)		6 (30.2)	30 (52.7)
Premenopausal	28 (65.2)	48 (50.2)		15 (80.5)	35 (56.8)
Maximum cancer diameter			0.002			0.008	0.55
≤5 cm	25 (60.2)	80 (88.5)		8 (52.8)	42 (68.5)
>5 cm	20 (45.6)	18 (17.06)		10 (54.8)	9 (15.9)
Lymphovascular invasion			<.002			.001	<.002
Positive	88 (35.9)	46 (59.8)		18 (22.6)	15 (35.9)
Negative	170 (59.8)	35 (40.8)		97 (89.0)	28 (70.2)