Table 6.

External validation performance of the best model (on the internal test set) and baseline logistic model for each outcome, displayed as mean ± std.dev. Note that transient nND was not recorded in the external data, so no results are available for that outcome.

Outcome	Model	ROC-AUC	Average Prec	Accuracy	Sensitivity	Specificity	PPV	NPV
mRS > 2	QDA	0.61 ± 0.03	0.08 ± 0.01	0.57 ± 0.04	0.59 ± 0.06	0.57 ± 0.05	0.06 ± 0.01	0.97 ± 0.01
	Baseline LR	0.66 ± 0.04	0.16 ± 0.04	0.69 ± 0.07	0.55 ± 0.10	0.69 ± 0.07	0.08 ± 0.01	0.97 ± 0.01
mRS-Diff. > 1	MLP	0.53 ± 0.01	0.05 ± 0.01	0.54 ± 0.05	0.44 ± 0.04	0.54 ± 0.05	0.03 ± 0.01	0.96 ± 0.00
	Baseline LR	0.53 ± 0.03	0.11 ± 0.05	0.64 ± 0.09	0.48 ± 0.12	0.65 ± 0.09	0.05 ± 0.01	0.97 ± 0.00
GOS < 5	GAM	0.58 ± 0.03	0.12 ± 0.02	0.59 ± 0.07	0.49 ± 0.13	0.60 ± 0.09	0.09 ± 0.01	0.94 ± 0.01
	Baseline LR	0.62 ± 0.02	0.16 ± 0.03	0.67 ± 0.05	0.47 ± 0.09	0.68 ± 0.06	0.11 ± 0.01	0.94 ± 0.01

mRS = modified Rankin Scale, GOS =  Glasgow outcome scale, nND = new neurological deficit, LR = logistic regression, QDA = quadratic discriminant analysis, MLP = multilayer perceptron, SVM = support vector machine, GAM = generalized additive model, ROC-AUC = area under receiver operating characteristic curve, PPV = positive predictive value, NPV = negative predictive value.