Fig. 8From: Classification of painful or painless diabetic peripheral neuropathy and identification of the most powerful predictors using machine learning models in large cross-sectional cohortsModel performance in the independent validation dataset. A: Scalar performance metrics. From left to right: Accuracy, the p.value of the negative binomial test with the null hypothesis that accuracy is not higher than the prevalence of the majority class, Balanced Accuracy, Mathews Correlation Coefficient, Area Under the Precision Recall Curve. B—D: Average performance estimates during training—red dots, versus performance achieved on the independent validation dataset—blue dotsBack to article page