Fig. 1: Predicted positive cases per cognitive group.
From: Evaluating spoken language as a biomarker for automated screening of cognitive impairment
Distribution of predicted probabilities for ADRD detection on the test set across MMSE scores, considering 10 bootstrap repeats. Central points represent the mean predicted probability per cognitive group (CN, mild, moderate, severe) and error bars show the standard deviation across all observations within each cognitive group. Lower MMSE values indicate worse cognition.
