Fig. 5: Model for selection of best digital speech biomarkers to predict medication state change and accuracy performance.
From: Digital speech biomarkers can measure acute effects of levodopa in Parkinson’s disease
a Regression coefficients, 95% Confidence Intervals, and Relative importances plots from the backward stepwise logistic regression model to predict changes in the medication state in PD patients. b Receiver operating characteristic (ROC) curve of the medication state change model and respective performance. *p < 0.05, **p < 0.01. NSR Net speech rate, RST Rate of speech timing, Int Kurt Kurtosis of intensity, sd F0 Standard deviation of fundamental frequency, MPT Maximal phonation time, LTAS Long-term averaged spectrum, skew Skewness, CI Confidence Interval, AUC Area under the curve, Sens Sensitivity, Spec Specificity, PPV Positive predive value, NPV Negative predictive value. Created in https://BioRender.com.
