Figure 3

Support Vector Machine (SVM) results. Line below: confusion matrix of the SVM model in predicting the patient’s outcome, relative metrics, and misclassification of the model if one of the variables is kept out. Considering the true positive (TP) and false positive (FP) classifications sensitivity (TP/(TP + FN)) and specificity (TN/(TN + FP)) indicate how well the model identifies positive and negatives cases respectively; Accuracy ((TP + TN)/(TP + TN + FP + FN)) measures the proportion of correctly classified cases among the total number of cases; balanced accuracy ((Sensitivity + Specificity)/2) is the average of sensitivity and specificity; positive likelihood ratio (sensitivity/(1 − specificity)) represents how much more likely a positive result is to occur in people with the condition compared to those without it, while the negative likelihood ratio (1 − sensitivity)/specificity represent how much more likely a negative result is to occur in people without the condition compared to those with it; F1 score (2*(precision*sensitivity)/(precision + sensitivity)) measures the model’s accuracy, combining both precision and recall into a single metric, where precision (TP/(TP + FP)) is the ratio between true positives and the sum of true positives and false positives. Line above: misclassification rate of 100 SVM models based on cost and gamma values and measure relative to the SVM chosen model. Entropy R² is a goodness-of-fit measure for classification models. It is based on the concept of entropy, which represents the uncertainty in a dataset. Generalized R² measures the proportion of variance in the dependent variable explained by the model. It is an extension of the traditional R-square used for linear regression and can be applied to non-linear models like SVM. Mean-log p (Mean Negative Log Likelihood) measures how well the predicted probabilities from the SVM model match the actual outcomes. RASE (Root Average Squared Error) measures the average squared difference between the predicted and actual values. Mean Absolute Deviation is the average of the absolute differences between the predicted and actual values.