Skip to main content
Fig. 6 | Orphanet Journal of Rare Diseases

Fig. 6

From: Machine learning in Huntington’s disease: exploring the Enroll-HD dataset for prognosis and driving capability prediction

Fig. 6

The ROC curve calculated on the test set for the driving capability. The ROC curve (blue line) is shown with a simulated random model (dashed green line) and multiple advisory thresholds (T), indicated by the blue dots. The thresholds 0.91 and 0.06 show that a predicted value above 0.91 and below 0.06 results in a correct prediction for the positive class and negative class respectively in 99% of the test cases. The default threshold of 0.5 shows that an AUC of 0.929 is achieved and a TPR of 0.84 and FPR of 0.13

Back to article page