Table 1.

Characteristics of the patients in different sets

Characteristics	Total (LN = 462)	Training set (LN = 283)			Testing set (LN = 122)			N2-validation set (LN = 57)
	LNM+ (176) LNM-(286)	LNM+  (100)	LNM− (183)	P value	LNM+  (51)	LNM− (71)	P value	LNM+  (25)	LNM− (32)	P value
Age (mean ± SD)	63.77 ± 9.13	61.26 ± 8.88	65.49 ± 9.02	< 0.001	60.92 ± 8.90	65.06 ± 7.87	0.008	61.12 ± 7.66	65.59 ± 11.67	0.103
Tumor length	4.57 ± 1.46	4.62 ± 1.51	4.49 ± 1.43	0.477	4.84 ± 1.36	4.54 ± 1.53	0.272	4.41 ± 1.33	4.58 ± 1.63	0.668
Gender
Female	130 (28.1%)	40(40.0%)	38(20.8%)	0.001	11 (21.6%)	23 (32.4%)	0.182	11 (44.0%)	7 (21.9%)	0.085
Male	332 (71.9%)	60(60.0%)	145 (79.2%)		40 (78.4%)	48 (67.6%)		14 (56.0%)	25 (78.1%)
Subtype
ADC	266 (57.6%)	73(73.0%)	90 (49.2%)	0.007	30 (58.8%)	41 (57.7%)	0.819	21 (84.0%)	11 (34.4%)	< 0.001
SCC	182 (39.4%)	24(24.0%)	91 (49.7%)		18 (35.3%)	25 (35.2%)		4 (16.0%)	20 (62.5%)
LCC	14 (3.0%)	3(3.0%)	2 (1.1%)		3 (5.9%)	5 (7.1%)		0 (0.0%)	1 (3.1%)
Location
Right	249 (53.9%)	67 (67.0%)	97 (53.0%)	0.021	26 (51.0%)	32 (45.1%)	0.523	15 (60.0%)	12 (37.5%)	0.095
Left	213 (46.1%)	33 (33.0%)	86 (47.0%)		25 (49.0%)	39 (54.9%)		10 (40.0%)	20 (62.5%)
Diabetes
No	429 (92.9%)	97 (97.0%)	163 (89.1%)	0.006	49 (96.0%)	65 (91.5%)	0.296	24 (96.0%)	31 (96.9%)	0.862
Yes	33 (7.1%)	3 (3.0%)	20 (10.9%)		2 (4.0%)	6 (8.5%)		1 (4.0%)	1 (3.1%)
Smoking
No	194 (42.0%)	49 (49.0%)	67 (36.6%)	0.046	16 (31.4%)	35 (49.3%)	0.046	13 (52.0%)	14 (43.7%)	0.544
Yes	268 (58.0%)	51 (51.0%)	116 (63.4%)		35 (68.6%)	36 (50.7%)		12 (48.0%)	18 (56.3%)

P value is obtained by using variance analysis between-group

ADC adenocarcinoma, SCC squamous cell carcinoma, LCC large cell carcinoma