Skip to main content

Table 5 Values obtained from patient face images

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 model with the best AUC value, in the range of 10–30 epochs, batch size 16–32, was created for each body position. Datasets 1–5 were used to divide the data into training data and test data. For each dataset, the AUC was calculated, the median and 95% confidence interval of the AUC for that dataset are shown, and the sensitivity, specificity, and accuracy of the median AUC are shown
  2. **The name of dataset that produced the median AUC
  3. *Median AUC>0.700