Long-time average spectrum in individuals with Parkinson disease 1

Abstract

Background:

Various methods of acoustic analysis have been used to describe phonatory deficits in Parkinson disease (PD), including long-time average spectrum (LTAS) analysis.

Objective:

The aim of this study was to utilize the LTAS to investigate laryngeal deficits in individuals with PD.

Methods:

Twenty-eight individuals with idiopathic PD and 10 controls were participants for this study. An LTAS was generated from a standard reading sample and analyzed using the following measurements: Mean spectral energy (MSE), first spectral peak, spectral tilt (ST), three spectral slope (SS) measures (i.e., SS^F0-1kHz, SS^1-5kHz, and SS ratio) and four spectral moments (mean, standard deviation, skewness, and kurtosis).

Results:

Individuals with PD were found to demonstrate significantly lower MSE, spectral mean and spectral SD, along with higher spectral skewness compared to control speakers. Furthermore, ST was positively correlated with self-rated symptom severity and self-rated medication effectiveness for individuals with PD.

Conclusions:

The LTAS findings were consistent with either reduced vocal fold adduction or vocal fold bowing in individuals with PD compared to control speakers. Additionally, vocal fold adductory changes were one possible explanation for correlations between LTAS and self-ratings of medication effectiveness and disease severity.