Fig. 5: Multilayer Perceptron (MLP) model results.
From: AI-QuIC machine learning for automated detection of misfolded proteins in seed amplification assays
a Model confusion matrix represents the ratio of how many samples from each true class were classified in each predicted class. b The Receiver Operator Characteristic (ROC) illustrates the sensitivity-specificity tradeoff of the model, with an ideal model having an area under the curve (AUC) of 1. c The examples include a single false positive a given model misclassified along with correctly classified reference samples, highlighting how distinguishable a misclassified false positive was.
