Skip to main content

Table 3 Uni-variate and Multi-variable COX Analysis of NVFs-associated factors in the training set

From: Development and validation of a machine learning model to predict imminent new vertebral fractures after vertebral augmentation

 

Uni-variate analysis

Multi-variate analysis

factors

Hazard Ratio (95% CI)

P

Hazard Ratio (95% CI)

P

Radiomics signature

13.682 (7.678, 25.802)

 < 0.001a

8.753 (2.746, 17.735)

0.001a

Age

1.040 (1.012, 1.069)

0.005a

1.002 (0.971, 1.034)

0.893

Sex

0.980 (0.548, 1.754)

0.946

-

-

Surgical procedure

1.534 (0.947, 2.485)

0.082

-

-

Number of treated vertebra(e)

0.780 (0.410, 1.482)

0.448

-

-

Location of treated vertebra(e)

1.631 (1.032, 2.576)

0.036a

1.348 (0.821, 2.212)

0.238

IVC

0.318 (0.197, 0.516)

 < 0.001a

0.422 (0.255, 0.698)

0.001a

Previous VF

0.488 (0.308, 0.774)

0.002a

0.834 (0.453, 1.534)

0.560

Previous Multi-VF

0.498 (0.302, 0.821)

0.006a

1.065 (0.567, 2.000)

0.845

  1. CI confidence interval, VF vertebral fracture, IVC intravertebral cleft
  2. aP < 0.05