Table 3.

Prediction of the severity of an acute affective episode: model and generalization on unseen patients.

Individuals with affective episodes and performance metric		Model	Generalization
Manic BDa
	Accuracyb (%)	70	56.67
	F1-score	0.6978	0.5279
	Precision	0.6979	0.5381
	Recall	0.7000	0.5667
	AUROCc	0.6980	0.5432
Depressed BD
	Accuracyb (%)	61.67	41.67
	F1-score	0.6171	0.3968
	Precision	0.6273	0.4085
	Recall	0.6167	0.4167
	AUROC	0.6115	0.4067
Mixed BD
	Accuracyb (%)	63.33	30
	F1-score	0.6333	0.2576
	Precision	0.6333	0.3004
	Recall	0.6333	0.3068
	AUROC	0.6333	0.3012
Depressed MDDd
	Accuracyb (%)	85	28.33
	F1-score	0.8492	0.2451
	Precision	0.8774	0.2581
	Recall	0.8500	0.2833
	AUROC	0.8672	0.2856

^aBD: bipolar disorder.

^bAccuracy expected by chance for a 3-class classification task is 1/3=33%. Thus, accuracies above 33% suggest that the model can predict outcomes better than random guessing, and higher values for accuracy indicate better predictive capacity of the model. Note that the test set was designed to have the same number of samples in each class. This is reflected in the values of F₁-score, precision, and recall being very close to each other and to that of accuracy.

^cAUROC: area under the receiver operating characteristic.

^dMDD: major depressive disorder.