Fig. 4From: Rapid identification of carbapenem-resistant Klebsiella pneumoniae based on matrix-assisted laser desorption ionization time-of-flight mass spectrometry and an artificial neural network modelThe area under the precision–recall curve (AUPRC) is used to assess the efficacy of machine learning models when the training or validation dataset is imbalanced. An AUPRC value of 0.90 is obtained in the external validation of the ANN model with an equal number of CSKP and CRKP isolates. This indicates that the model is robust despite the imbalanced distribution of the datasetBack to article page