Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Sleep Apnea01:21

Sleep Apnea

479
Sleep apnea is a condition where breathing stops intermittently during sleep, often leading to significant health issues. Each episode can last from 10 to 20 seconds or more and is frequently accompanied by a brief arousal from sleep. This disturbance, largely unnoticed by the individual, can lead to severe daytime fatigue. Commonly, individuals seek help after being informed by their partners about loud snoring and noticeable breathing pauses during sleep.
The condition is more prevalent among...
479

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Machine Learning-Based Survival Prediction Models for Young Patients With Gastric Cancer: Model Development and Validation Study.

JMIR cancer·2026
Same author

Estimating Sleep-Stage Distribution from Respiratory Sounds via Deep Audio Segmentation.

Sensors (Basel, Switzerland)·2025
Same author

Structure-Based Engineering of a PTPsigma Ectodomain for Enhanced Solubility and Productivity.

International journal of molecular sciences·2025
Same author

High-affinity promotor binding of YhaJ mediates a low signal leakage for effective DNT detection.

Frontiers in microbiology·2025
Same author

Understanding Morning Emotions by Analyzing Daily Wake-Up Alarm Usage: Longitudinal Observational Study.

JMIR human factors·2024
Same author

Disulfide Bond Engineering of Soluble ACE2 for Thermal Stability Enhancement.

International journal of molecular sciences·2024
Same journal

The Clinical, Humanistic, and Economic Burden of Idiopathic Hypersomnia in the United States: Analysis of the National Health and Wellness Survey.

Nature and science of sleep·2026
Same journal

A Nomogram Model for Predicting Moderate to Severe OSA in Western China: A Retrospective Analysis.

Nature and science of sleep·2026
Same journal

Content Validity and Psychometric Evaluation of the Idiopathic Hypersomnia Severity Scale (IHSS) in Adults with Idiopathic Hypersomnia.

Nature and science of sleep·2026
Same journal

The Prognostic Value of Early Treatment Factors for Cognitive Behavioral Therapy for Insomnia (CBT-I) Outcomes.

Nature and science of sleep·2026
Same journal

Follow-up Study on the Sleep Status of Anti-Epidemic Staff: A Study Based on Wearable Sleep Trackers.

Nature and science of sleep·2026
Same journal

Prevalence of Insomnia in Healthcare Profession - PIHEP Study: A Large-Scale National Study in Vietnam.

Nature and science of sleep·2026
See all related articles

Related Experiment Video

Updated: Jan 15, 2026

Asthma Detection Research Based on Voice Signal Processing and Machine Learning
04:04

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Published on: July 22, 2025

926

ApneaWhisper: Transformer-Based Audio Segmentation for Fine-Grained Non-Invasive Sleep Apnea Detection.

Yunu Kim1, Myeongbin Kim1, Jaemyung Shin2

  • 1Department of Applied Artificial Intelligence, Hanyang University ERICA, Ansan, Gyeonggi-do, Republic of Korea.

Nature and Science of Sleep
|October 10, 2025
PubMed
Summary
This summary is machine-generated.

ApneaWhisper, a Transformer model, noninvasively detects sleep apnea subtypes from audio with high accuracy. It shows promise for home monitoring by estimating the apnea-hypopnea index (AHI).

Keywords:
audio segmentationdeep learningsleep apneasleep breathingtransformerwhisper

More Related Videos

Drug-Induced Sleep Endoscopy DISE with Target Controlled Infusion TCI and Bispectral Analysis in Obstructive Sleep Apnea
07:54

Drug-Induced Sleep Endoscopy DISE with Target Controlled Infusion TCI and Bispectral Analysis in Obstructive Sleep Apnea

Published on: December 6, 2016

20.5K
Infant Auditory Processing and Event-related Brain Oscillations
06:34

Infant Auditory Processing and Event-related Brain Oscillations

Published on: July 1, 2015

16.9K

Related Experiment Videos

Last Updated: Jan 15, 2026

Asthma Detection Research Based on Voice Signal Processing and Machine Learning
04:04

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Published on: July 22, 2025

926
Drug-Induced Sleep Endoscopy DISE with Target Controlled Infusion TCI and Bispectral Analysis in Obstructive Sleep Apnea
07:54

Drug-Induced Sleep Endoscopy DISE with Target Controlled Infusion TCI and Bispectral Analysis in Obstructive Sleep Apnea

Published on: December 6, 2016

20.5K
Infant Auditory Processing and Event-related Brain Oscillations
06:34

Infant Auditory Processing and Event-related Brain Oscillations

Published on: July 1, 2015

16.9K

Area of Science:

  • Sleep Medicine
  • Artificial Intelligence in Healthcare
  • Biomedical Signal Processing

Background:

  • Sleep apnea is a common disorder with significant health risks.
  • Accurate diagnosis often relies on polysomnography (PSG), which is resource-intensive.
  • Noninvasive methods for sleep apnea detection and subtype classification are needed.

Purpose of the Study:

  • To introduce ApneaWhisper, a Transformer-based audio segmentation model.
  • To evaluate ApneaWhisper's efficacy in noninvasively detecting sleep apnea subtypes using PSG-Audio data.
  • To assess the model's capability for frame-level and clip-level sleep apnea event prediction.

Main Methods:

  • Utilized a PSG-Audio dataset from 284 patients.
  • Employed a pretrained Whisper encoder for 10 ms-resolution frame-level feature extraction.
  • Developed a lightweight Transformer decoder with token-based segmentation and classification head, fine-tuned with class-balanced cross-entropy loss.

Main Results:

  • Achieved a clip-level F1-score of 0.82 and a frame-level F1-score of 0.70 for sleep apnea detection.
  • Outperformed conventional baseline models (MFCC+DNN, VGGish+bi-LSTM, VAD-based).
  • Demonstrated promising ability in distinguishing between Obstructive Sleep Apnea (OSA), Mixed Sleep Apnea (MSA), Central Sleep Apnea (CSA), and hypopnea.

Conclusions:

  • ApneaWhisper offers precise apnea event localization, duration estimation, and subtype classification.
  • The model facilitates accurate apnea-hypopnea index (AHI) estimation, potentially reducing reliance on full PSG.
  • Challenges remain in distinguishing CSA and MSA; future work may involve multimodal integration and noise-robust training.