Feasibility of a Machine Learning-Based Smartphone Application in Detecting Depression and Anxiety in a Generally Senior Population.

For the three different designated specificity levels of 0.70, 0.75, and 0.80, the AUC performance (as represented by varying sensitivity) for PHQ-8 prediction has an increasing trend as speaking length increases, from the shortest speaking length of ~20 s to 180 s ( Figure 6 ).

also in 63,971 other papers

However, for GAD-7 there was minimal change in performance at the specificity level of 0.80, while the two lower designated specificity levels showed a trend toward improved sensitivity with speaking length ( Figure 7 ).

also in 53,322 other papers