Assessing Freezing of Gait in Parkinson's Disease Using Analysis of Hypokinetic Dysarthria

Hypokinetic dysarthria (HD) and freezing of gait (FOG) are frequent symptoms of Parkinson's disease (PD). The aim of this work is to reveal pathological mechanisms common for HD and FOG, and use acoustic analysis of dysarthric speech to assess the gait difficulties in PD. We used a correlation analysis to investigate a relationship between speech features and FOG evaluated by freezing of gait questionnaire (FOG-Q). We found speech features quantifying reduced mobility of the articulatory organs significantly correlated with all parts of the questionnaire. Next, we built multivariate regression models to estimate the FOG-Q total score. With this approach, mean estimation error rate of 14.71% was achieved. We confirmed the previous findings of a close relationship between HD and FOG in PD. Furthermore, we showed it is possible to accurately (with the error of approximately 0.5 points) estimate FOG-Q using a reasonable number of conventional speech features.