Skip to main content

Table 2 Univariate and multivariate logistic regression analysis of clinical and CECT semantic features of patients

From: Deep learning radiomics-based prediction model of metachronous distant metastasis following curative resection for retroperitoneal leiomyosarcoma: a bicentric study

Variable

Univariate Logistic Analysis

Multivariate Logistic Analysis

OR (95% CI)

P value

OR (95% CI)

P value

DLR-score

13.445 [7.879, 73.527]

< 0.001

13.936 [5.669, 107.238]

< 0.001

Age

1.005 [0.979, 1.033]

0.72

-

-

Sex

0.589 [0.280, 1.225]

0.159

-

-

Ki-67 index

1.008 [0.990, 1.026]

0.396

-

-

Tumor size > 10 cm

8.969 [3.602, 25.882]

< 0.001

7.943 [1.881, 44.452]

0.009

Clinical N stage

1.454 [0.534, 3.920]

0.457

-

-

Cystic spaces or necrosis

1.604 [0.322, 11.954]

0.591

-

-

Degree of enhancement

1.641 [0.593, 4.551]

0.335

-

-

Enhancement pattern

2.594 [1.050, 6.622]

0.041

1.725 [0.351, 8.909]

0.502

Tumor contours

1.714 [0.711, 4.145]

0.227

-

-

Adjacent organ involvement

1.147 [0.553, 2.386]

0.712

-

-

  1. Abbreviations: DLR, deep learning radiomics; CECT, contrast-enhanced computed tomography; OR, odds ratio; CI, confidence interval