Skip to main content

Table 6 AI model prediction performance using training data without Braden metrics

From: Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence

AI technique Original (row, imbalanced) data Post-processed (rebalanced) cohorts
LASSO linear regularized model Model Model
Performance of testing data Performance of testing data
Confusion matrix Result false Result true Confusion matrix Result false Result true
  Prediction False 4775 150 Prediction False 4111 48
  Prediction True 21 67 Prediction True 685 169
  Accuracy 96.59%   Accuracy 85.38%  
  95% CI 96.05% 97.07% 95% CI 84.37% 86.35%
  Kappa 0.425   Kappa 0.265  
  Sensitivity 0.30876   Sensitivity 0.7788  
  Specificity 0.99562   Specificity 0.8571  
  Prevalence 0.04329   Prevalence 0.04329  
Random forest