Related Experiment Video
Updated: May 5, 2026
07:54
Drug-Induced Sleep Endoscopy DISE with Target Controlled Infusion TCI and Bispectral Analysis in Obstructive Sleep Apnea
Published on: December 6, 2016
19.7K
Artificial intelligence facial recognition of obstructive sleep apnea: a Bayesian meta-analysis
Esther Yanxin Gao1,2,3,4, Benjamin Kye Jyn Tan5,6,7, Nicole Kye Wen Tan3
1Department of Otorhinolaryngology-Head & Neck Surgery, Singapore General Hospital (SGH), Singapore, Singapore.
Sleep & Breathing = Schlaf & Atmung
|November 30, 2024
Summary
Artificial intelligence (AI) using craniofacial photographs shows high accuracy for diagnosing obstructive sleep apnea (OSA). This AI approach offers a cost-effective screening tool, potentially improving accessibility in primary care settings.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Healthcare
- Sleep Medicine
Background:
- Conventional diagnosis of obstructive sleep apnea (OSA) via polysomnography is expensive and not widely accessible.
- Emerging artificial intelligence (AI) technologies offer novel methods for OSA diagnosis using readily available data.
- Craniofacial photographs represent a potential low-cost data source for AI-driven OSA screening.
Purpose of the Study:
- To evaluate the diagnostic accuracy of AI algorithms trained on craniofacial photographs for obstructive sleep apnea (OSA).
- To compare the performance of AI-based OSA diagnosis against conventional diagnostic criteria.
- To identify the most effective AI approaches for OSA detection from photographic data.
Main Methods:
- A systematic literature search was conducted across major scientific databases (PubMed, Embase, Scopus, Web of Science, IEEE Xplore).
- Observational studies of adults comparing AI diagnostic performance using craniofacial photographs against apnea-hypopnea index (AHI) criteria were included.
- A Bayesian bivariate meta-analysis was performed on eligible studies, with risk of bias assessment.
Main Results:
- Six studies comprising 10 AI models were included, demonstrating low risk of bias.
- AI models using craniofacial photographs achieved a pooled sensitivity of 84.9% and specificity of 71.2% for OSA diagnosis.
- Deep learning models, specifically convolutional neural networks, exhibited the highest accuracy (91.1% sensitivity, 79.2% specificity), comparable to home sleep apnea tests.
Conclusions:
- AI trained on craniofacial photographs demonstrates high diagnostic accuracy for obstructive sleep apnea (OSA).
- This AI-driven approach is a promising low-cost screening tool for OSA, enhancing accessibility.
- Future research focusing on deep learning with smartphone imagery could further improve feasibility in primary care.
Keywords:
Deep learningDiagnostic test accuracyMachine learningNeural networksSleep disordered breathingSnoringMore Related Videos
Related Concept Videos
01:21Sleep Apnea
914
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...
The condition is more prevalent among...
914
01:24Facial Feedback Hypothesis
953
Charles Darwin proposed that facial expressions are an evolutionary adaptation for communication. He argued that these expressions are not influenced by culture but are universal across species. For example, a snarling expression with exposed teeth signals a threat in many animals, including humans. Darwin also suggested that displaying an emotion can intensify the feeling. Smiling, for example, could enhance one's sense of happiness. This idea laid the foundation for understanding the role...
953