Multimodal Deep Learning Model for Predicting the Apnea-Hypopnea Index in Obstructive Sleep

THE QUICK READ

What this study involves

The registry does not currently label this study as recruiting. Ask about its status before planning a screening visit.

Study type: Observational
Phase: Not applicable / not listed
Age range: 30 years - 75 years
Sex eligible: All sexes
Healthy volunteers: Not accepted
Participants: 150 Actual

These are starting points, not an eligibility decision. The research team must review the full requirements.

About the research

This study aims to develop a multimodal deep learning model that integrates noninvasive signals to predict the severity of obstructive sleep apnea. By establishing a clinically viable and user-friendly monitoring tool, the study seeks to enhance early screening accessibility and support the development of home-based sleep care systems.

Read the full description

Obstructive sleep apnea is a common sleep disorder closely associated with cardiovascular, metabolic, and neuropsychiatric comorbidities. It is characterized by repeated upper airway collapse during sleep, leading to intermittent hypoxia and sleep fragmentation. Although polysomnography remains the diagnostic gold standard for obstructive sleep apnea, its high cost, complexity, and limited accessibility pose challenges for large-scale screening and early identification. Recent advancements in noninvasive sensing technologies-such as electronic stethoscopes, wearable oximeters, and under-mattress pressure sensors-have enabled low-burden physiological monitoring solutions, offering new opportunities for simplified obstructive sleep apnea detection. In this study, synchronized multimodal physiological data will be collected during overnight sleep, including respiratory sounds, continuous saturation measurements, and standard polysomnography waveforms. Signal preprocessing and feature extraction will be performed to ensure data quality and temporal alignment. A deep learning model will be developed using these multimodal signals as inputs. The apnea-hypopnea index will be derived from overnight polysomnography. The model will be trained to estimate apnea-hypopnea index values and classify obstructive sleep apnea severity according to established clinical thresholds.

Who may be able to participate?

The list below is the sponsor’s full registry wording. It can include conditions, prior treatment, test results and other requirements. A match in search does not mean you qualify.

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

Study population

patient of Affiliated University Hospital

Understand screening

Treatment, groups and participation

The registry may describe treatments and study duration without a complete visit calendar. Ask the study team for the exact schedule, procedures, risks, costs and alternatives. Keep your usual care plan unless your treating clinician tells you otherwise.

Assignment: Not reported
Masking: Not reported

electronic stethoscope DEVICE

digital device amplifying and recording cardiopulmonary sounds

fingertip pulse oximeter DEVICE

a small device placed on the finger to measure blood oxygen saturation (SpO₂) and pulse rate noninvasively.

pressure-sensing mattresses DEVICE

using ballistocardiography (BCG) for monitoring respiration and heart rate

Prepare your questions

What researchers are measuring

Outcome measures describe what is being studied. They are not proof that a treatment works.

Primary measures

apnea-hypopnea index, sound waveforms, and the correlation between apnea-hypopnea index and ballistocardiography waveforms

Time frame: one night

Not reported in this registry record.

No results are flagged as posted.

Review the complete result tables and limitations in the official record.

Open official results ↗

Research locations (1)

The overall study status and a local site’s status can differ. Confirm that location’s availability before traveling.

Find a location in this study

1 locations

Fu Jen Catholic University Hospital, Fu Jen Catholic UniversityNew Taipei City, TaiwanStatus unknown

View location on map ↗

Study team and timeline

Lead sponsor: Fu Jen Catholic University
Study start: Sep 5, 2025 Actual
Primary completion: Aug 17, 2026 Actual
Study completion: Aug 17, 2026 Actual
First posted: Mar 4, 2026 Actual
Last updated: Nov 10, 2026 Actual

Official study contacts

No central contact is listed. Check the locations and the official record.

Official study title

A Multisensor Deep Neural Framework Combining Digital Auscultation, Oxygen Saturation, and Motion Data to Estimate the Apnea-Hypopnea Index in Obstructive Sleep Apnea

A HUMAN NEXT STEP

Ask Wellness Clinical Research

Your message goes to WCR, including when this study is run by another organization. WCR may offer general guidance but may not operate this trial, have access to its team or be able to enroll you.

Source: ClinicalTrials.gov API v2. Information is submitted by sponsors and investigators. NLM does not approve the safety or science of all listed studies. Registry disclaimer.

Retrieved: Nov 11, 2026 · Last registry update: Nov 10, 2026