You are here

Article Text

Article menu
Download PDFPDF
F: Clinical studies: case reports, observational studies and trials
F52 Acoustic voice features in Huntington’s disease in native English, Spanish and polish speakers
  1. Vitória S Fahed1,
  2. Emer P Doheny1,
  3. Carla Collazo2,
  4. Joanna Krzysztofik3,
  5. Elliot Mann4,
  6. Philippa Jones4,
  7. Cheney Drew4,
  8. Grzegorz Witkowski3,
  9. Esther Cubo2,
  10. Monica Busse4,
  11. Madeleine M Lowery1,
  12. on behalf of DOMINO-HD consortium
  1. 1School of Electrical and Electronic Engineering, Insight Centre for Data Analytics, University College Dublin, Belfield, Ireland
  2. 2Hospital Universitario of Burgos, Spain
  3. 3Institute of Psychiatry and Neurology, Poland
  4. 4Centre for Trials Research, Cardiff University, UK

Abstract

Background Changes in voice are a characteristic symptom of Huntington’s disease (HD), and can be associated with lack of motor control (dysarthria) or cognitive impairment. Objective quantitative measures of voice that can be performed within the clinic or home may provide a method to monitor disease progression and help target interventions.

Aims This study aimed to compare acoustic voice features from voice recordings in English, Spanish and Polish-speaking in people with HD and controls.

Methods Voice was recorded using mobile devices (Samsung Tablets Tab A and S6 Lite, and Smartphone Huawei Mate Lite 10) in participants with HD and matched control groups across three cohorts: English- (n=29), Spanish- (n=36) and Polish- (n=25) speaking. Voice was recorded during sustained vowel phonation (/a:/), syllable repetitions (/pa/,/ta/,/ka/,/pataka/,/pati/), and reading of a language-specific passage (English,1 Spanish,2 Polish3). Acoustic voice features were estimated and compared across cohorts and biological sex. Statistical analysis was performed using linear mixed models.

Results Significant differences between HD and control participants were observed for the standard deviation of fundamental frequency, harmonics-to-noise ratio, jitter and shimmer (Figure 1). Effect of biological sex was observed on all features examined.

Abstract F52 Figure 1
Abstract F52 Figure 1

Sample signal for (a) control group and (b) participants with HD for the sustained vowel task. Comparison of (c) standard deviation (SD) of the fundamental frequency in Hz, (d) harmonics-to-noise ratio (HNR) in decibels, (e) local jitter in percentage and (f) shimmer in decibels, between control and HD cohorts considering biological sex for the sustained vowel task. * indicates a significant difference (p<0.05) between control and HD cohorts

Conclusions The results illustrate differences between HD and control participants, consistent across languages, suggesting an opportunity to develop a language-agnostic protocol (sustained vowel and syllable repetition) that can be implemented using mobile devices to evaluate voice impairment in HD.

References

  1. Patel et al. 2018.

  2. Orozco-Arroyave et al. 2016.

  3. Pettorino et al. 2017.

  • Acoustic voice features
  • Huntington’s disease
  • signal processing
  • English
  • Spanish
  • Polish.

https://doi.org/10.1136/jnnp-2022-ehdn.143

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.