Table 3 Experimental results of an ensemble of 10 fused multi-label models trained on the SData set (\(n=22\)) and evaluated on the GData set (\(n=157\)) for the detection of PD. The AUC in the age corrected GData column refers to the mean AUC across 10 randomly generated, age-matched GData subsets (as such, we do not report performance metrics values at the two operating points for this experiment).

From: Unobtrusive detection of Parkinson’s disease from multi-modal and in-the-wild sensor data using deep learning techniques