Artificial Intelligence in detecting obstructive sleep apnea using dental radiographs and craniofacial imaging: A mini review

Obstructive sleep apnea (OSA) is a common sleep disorder in which breathing repeatedly stops and starts during sleep. It is often not diagnosed early, but it is linked with serious health problems like high blood pressure and heart disease. The standard test for diagnosis, polysomnography, is costly and not easily available everywhere. Because of this, early detection is often missed.
Dental imaging like lateral cephalograms and cone-beam computed tomography (CBCT) is routinely used in dental practice. These images can also show changes in airway and facial structures that are related to OSA. This gives an opportunity for dentists to help in early screening of patients.
Recently, artificial intelligence (AI), including machine learning and deep learning techniques, has shown good potential in detecting OSA from dental radiographs and craniofacial images. Studies have shown that AI models can identify OSA risk using lateral cephalograms by analysing facial and airway features. AI-based CBCT studies have also shown that airway size and shape can be automatically measured and used for prediction. Some studies also show that combining anatomical features with AI can improve accuracy.
This mini review summarises the current research on the use of AI for detecting OSA using dental imaging. It explains different imaging methods, AI techniques, and their usefulness in clinical practice. Although results are promising, there are some limitations like small sample sizes, lack of large validation studies, and differences in imaging methods.
In conclusion, AI-based analysis of dental imaging has strong potential for early detection of OSA in dental clinics. However, more large studies are needed before it can be used in routine clinical practice.