Random-forest algorithm based biomarkers in predicting prognosis in the patients with hepatocellular carcinoma

Table 1 Correlation between NUF2 expressions with clinic-pathological characteristics of HCC

Clinicopathological variables	N	NUF2 expression		P value
Clinicopathological variables	N	Low (37)	High (58)	P value
Sex				0.482
Male	81	31	50
Female	14	6	8
Age, years				0.533
< 50	48	19	29
≥ 50	47	18	29
AFP, ng/L				< 0.001
< 200	53	29	24
≥ 200	42	8	34
HBsAg				0.439
Negative	44	12	32
Positive	51	25	26
Tumor size, cm				0.075
≤ 5	49	23	26
> 5	46	14	32
Tumor nodule number				0.001
Solitary	57	30	27
Multiple (≥ 2)	38	7	31
Cancer embolus				0.002
Absence	62	31	31
Presence	33	6	27
TNM stage				0.013
Early (I & II)	52	26	26
Late (III & IV)	43	11	32
Differentiation grade				0.571
Well	67	26	41
Poor	28	11	17

ISSN: 1475-2867