Table 4 Feature influence in imaging model.

From: Radiomics-enhanced modelling approach for predicting the need for ECMO in ARDS patients: a retrospective cohort study

 

B

SE

Wald

p value

OR

95% CI for OR

Lower

Upper

Normal_vent [relative proportion]

-6.195

1.317

22.123

< 0.001

0.002

0.000

0.027

  1. This table presents the strength of the feature’s influence quantified by the Wald score and regression coefficient. B: regression coefficient used for model development; SE: standard error; OR: odds ratio; CI: confidence interval; normal_vent: proportion of normal ventilated lung volume, expressed as a fraction of the overall volume of the lung.
Back to article page