Noninvasive sensors and AI to predict sleep apnea severity
A Multisensor Deep Neural Framework Combining Digital Auscultation, Oxygen Saturation, and Motion Data to Estimate the Apnea-Hypopnea Index in Obstructive Sleep Apnea
This project will try to use an electronic stethoscope, fingertip pulse oximeter, and under-mattress pressure sensors combined with AI to predict how severe sleep apnea is in adults scheduled for a sleep test.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 150 (estimated) |
| Ages | 30 Years to 75 Years |
| Sex | All |
| Sponsor | Fu Jen Catholic University Academic / other |
| Locations | 1 site (New Taipei City) |
| Trial ID | NCT07447999 on ClinicalTrials.gov |
What this trial studies
Researchers will collect synchronized overnight recordings from an electronic stethoscope (respiratory sounds), a fingertip pulse oximeter (continuous oxygen saturation), pressure-sensing mattresses, and standard polysomnography. Recorded signals will be preprocessed, temporally aligned, and features extracted to ensure data quality. A multimodal deep learning model will be trained to estimate the apnea‑hypopnea index and classify obstructive sleep apnea severity using polysomnography-derived AHI as the reference. The goal is to create a clinically viable, user-friendly monitoring tool that could support earlier screening and home-based sleep care.
Who should consider this trial
Good fit: Adults aged 30–75 with suspected obstructive sleep apnea who are scheduled for an overnight polysomnography and can provide informed consent.
Not a fit: People with arrhythmia, neuromuscular disorders, significant structural airway abnormalities, pregnancy, recent hospitalization, intolerance to the sensors, or inability to consent are excluded and would not receive benefit from this protocol.
Why it matters
Potential benefit: If successful, this could enable easier and lower-cost screening and home monitoring by estimating sleep apnea severity without a full overnight polysomnography.
How similar studies have performed: Prior single‑sensor and some multimodal machine‑learning studies have shown promising results, but widely validated multimodal AI tools for home obstructive sleep apnea detection remain limited.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * age 30-75 years * clinically suspected obstructive sleep apnea and scheduled for polysomnography * willing and able to provide written informed consent Exclusion Criteria: * intolerance to the electronic stethoscope or fingertip pulse oximeter * significant structural airway abnormalities * arrhythmia * neuromuscular disorders * pregnancy * hospitalization within the past 1 month * inability to provide informed consent or requiring legal guardian consent
Where this trial is running
New Taipei City
- Fu Jen Catholic University Hospital, Fu Jen Catholic University — New Taipei City, Taiwan (Recruiting)
Study contacts
- Principal investigator: Ke-Yun Chao, PhD — Fu Jen Catholic University
- Study coordinator: Ke-Yun Chao, PhD
- Email: C00152@mail.fjuh.fju.edu.tw
- Phone: +886-905-301-879
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.