Skip to main content

Table 2 Univariate and multivariate regression analyses between early recurrence and non-recurrence groups in the training cohort

From: Hepatocellular carcinoma: radiomics nomogram on gadoxetic acid-enhanced MR imaging for early postoperative recurrence prediction

Intercept and variables

Univariate analysis

 

Multivariate analysis

 

β

Odd ratios (95%CI)

P

β

Odd ratios (95%CI)

P

Intercept

–

–

–

1.599

–

< 0.001*

AFP level

1.918

6.809 (2.984–16.419)

0.000*

1.621

2.112 (0.488–9.974)

< 0.001*

Tumour size

0.905

2.472 (1.149–5.438)

0.021*

–

–

–

Multifocality

1.067

2.908 (1.274–6.912)

0.012*

–

–

–

Gross vascular invasion

1.408

4.090 (1.778–9.923)

0.001*

1.211

3.356 (1.308–9.023)

0.013*

Non-smooth tumour margin

1.159

3.189 (1.446–7.271)

0.004*

1.006

2.735 (1.104–6.989)

0.031*

Peritumoural hypointensity on HBP images

0.865

2.376 (1.097–5.271)

0.036*

–

–

–

Peritumoural enhancement

1.136

3.116 (1.436–6.955)

0.004*

–

–

–

Radiomics score

1.000

2.718 (1.691–4.370)

0.000*

0.889

2.433 (1.436–4.473)

0.002*

  1. Significant variables with P < 0.05 in the univariate analysis were included in the multivariate logistic regression analysis
  2. AFP alpha-fetoprotein, CI confidence internal
  3. β is the regression coefficient. * P value < 0.05