Related Experiment Video
Updated: Jun 18, 2026
12:43
A Protocol for Comprehensive Assessment of Bulbar Dysfunction in Amyotrophic Lateral Sclerosis (ALS)
Published on: February 21, 2011
Objective dysphonia quantification in vocal fold paralysis: comparing nonlinear with classical measures
Max A Little1, Declan A E Costello, Meredydd L Harries
1Systems Analysis, Modeling and Prediction Group, University of Oxford, Oxford, United Kingdom.
Journal of Voice : Official Journal of the Voice Foundation
|November 11, 2009
Summary
New nonlinear voice analysis methods show promise for assessing vocal fold paralysis treatment. These novel measures are more reliable than traditional ones, aiding in monitoring patient recovery after surgery.
Area of Science:
- Acoustic analysis of voice
- Speech pathology
- Biomedical engineering
Background:
- Classical acoustic voice analysis uses jitter, shimmer, and NHR, but has limitations with severe dysphonia.
- Alternative nonlinear random measures have shown potential for various vocal pathologies.
Purpose of the Study:
- To evaluate the performance of nonlinear random measures against classical measures in a cohort with unilateral vocal fold paralysis (UVFP).
- To assess voice changes pre- and postoperatively following medialization thyroplasty surgery.
Main Methods:
- Analyzed voice recordings from 17 UVFP patients and 11 healthy controls.
- Utilized jitter, shimmer, NHR, recurrence period density entropy (RPDE), detrended fluctuation analysis (DFA), and correlation dimension.
- Systematized preanalysis editing for improved reliability.
Main Results:
- Nonlinear measures, particularly RPDE, demonstrated greater stability and reliability in healthy controls.
- RPDE and jitter effectively detected pre- to postoperative voice improvements in UVFP patients.
- All tested measures distinguished between controls and patients (treated/untreated), with significant clinical importance (AUC > 0.7).
- GRBAS ratings confirmed significant improvement in overall dysphonia grade post-surgery (AUC=0.946).
Conclusions:
- Systematized preanalysis editing enhances the reliability of nonlinear random measures for voice analysis.
- Nonlinear measures show potential utility in monitoring treatment effectiveness for UVFP and possibly other forms of dysphonia.
