Clinical and radiological feature selection
Of the features, 18 radiological features and 17 clinical features were selected to form the predictors based on the result from Tables 1 and 2. Table 3 lists the features selected by univariate analysis and LASSO.
Table 3 Selected features in C, R, and CR models
Model and individual features Coefficients
R, n = 8 (41)*
  Intercept − 0.307
  Total number of mixed GGO in peripheral area 0.359
  Total number of consolidation − 1.262
  Total number of solid nodules with ground-glass opacities 0.452
  Interlobular septal thickening − 5.559
  Crazy paving pattern 3.566
  Tree-in-bud − 2.548
  Pleural thickening 3.265
  Offending vessel augmentation in lesions 5.504
C, n = 7 (26)*
  Intercept 29.273
  Respiration − 0.359
  Heart rate − 0.054
  Temperature − 0.289
  White blood cell count − 0.175
  Cough − 1.866
  Fatigue 2.855
  Lymphocyte count category − 0.028
CR, n = 10 (67)*
  Intercept 45.117
  Total number of mixed GGO in peripheral area 0.108
  Tree-in-bud − 1.853
  Offending vessel augmentation in lesions 6.000
  Respiration − 0.583
  Heart ratio − 0.084
  Temperature − 0.536
  White blood cell count − 0.471
  Cough − 0.997
  Fatigue − 0.228
  Lymphocyte count category − 2.177
C, R, and CR indicate the predicted model based on clinical features, radiological features, and the combination of clinical features and clinical radiological features, respectively
*n means corresponding selected features, and data in parentheses are total features. Coefficients: the estimate value of each feature in multivariate logistic regression model by “glm” package in R