Skip to main content

Table 6 Comparison of accuracy between existing predictors of difficult intubation and the best AI model

From: Creation of an artificial intelligence model for intubation difficulty classification by deep learning (convolutional neural network) using face images: an observational study

  1. The AUCs of the existing predictors of the difficult intubations in this study ranged from 0.558 to 0.673, and no predictor exceeded 0.700. In contrast, the best AI model produced in this study had an AUC of 0.864, which was superior to the AUC of the predictors of difficult intubation