Towards an Accessible Speech-based Obstructive Sleep Apnea Screening Tool for Underserved Populations.

Area of Science:

• Respiratory Medicine
• Speech Science
• Public Health

Background:

• Obstructive sleep apnea (OSA) is a prevalent chronic respiratory disorder with significant comorbidities.
• Polysomnography (PSG) is the gold standard for OSA diagnosis but is inconvenient and inaccessible for many.
• Speech analysis offers a potential low-cost, accessible screening alternative for OSA.

Purpose of the Study:

• To investigate spectral speech characteristics differences between individuals with and without OSA in a homeless population.
• To explore the impact of body posture on these speech differences.
• To highlight potential speech markers for developing an accessible OSA monitoring tool.

Main Methods:

• A small cohort (n=18) of individuals experiencing homelessness with and without OSA in Toronto, Canada, was studied.
• Spectral speech characteristics were analyzed.
• The effect of body posture on speech differences was examined.

Main Results:

• Significant differences in spectral speech characteristics were observed between individuals with and without OSA.
• Body posture influenced these speech differences.
• Vowel articulation showed potential for OSA detection.

Conclusions:

• Speech analysis shows promise as an accessible screening tool for OSA, particularly in underserved populations.
• This research is the first to examine speech characteristics related to OSA in people experiencing homelessness.
• Further research can refine speech-based OSA monitoring tools for broader accessibility.