Screening method for early dementia using sound objects as voice biomarkers

TL;DR

This paper introduces a novel voice biomarker-based screening method for early dementia using sound object features from short vowel recordings, achieving high accuracy in distinguishing healthy individuals from those with MCI or Alzheimer's.

Contribution

The work's main novelty is the use of sound object features for more accurate and interpretable voice-based dementia screening from brief speech samples.

Findings

ROC AUC of 0.85 for MCI detection

Accuracy of 0.76 for MCI vs healthy

Effective screening with very short voice recordings

Abstract

Introduction: We present a screening method for early dementia using features based on sound objects as voice biomarkers. Methods: The final dataset used for machine learning models consisted of 266 observations, with a distribution of 186 healthy individuals, 46 diagnosed with Alzheimer's, and 34 with MCI. This method is based on six-second recordings of the sustained vowel /a/ spoken by the subject. The main original contribution of this work is the use of carefully crafted features based on sound objects. This approach allows one to first represent the sound spectrum in a more accurate way than the standard spectrum, and then build interpretable features containing relevant information about subjects' control over their voice. Results: ROC AUC obtained in this work for distinguishing healthy subjects from those with MCI was 0.85, while accuracy was 0.76. For distinguishing between healthy subjects and those with either MCI or Alzheimer's the results were 0.84, 0.77, respectively. Conclusion: The use of features based on sound objects enables screening for early dementia even on very short recordings of language-independent voice samples.