A research team from the Philippines and Taiwan has developed an artificial intelligence-assisted diagnostic system capable of identifying tooth and sinus structures in dental X-rays with 98.2% accuracy.
The study, published by the Ateneo Laboratory for Intelligent Visual Environments (ALIVE) in collaboration with Taiwan’s Chang Gung Memorial Hospital, National Cheng Kung University, Chung Yuan Christian University, and Ming Chi University of Technology, introduces a deep learning model designed to enhance the diagnosis of odontogenic sinusitis.
The YOLO (You Only Look Once) 11n model, a lightweight, real-time object detection system, aims to lower patients’ radiation exposure and reduce diagnostic costs. “The detection method developed in this study effectively reduces the radiation dose patients receive during CT imaging and serves as an auxiliary system, providing dentists with reliable support for the precise diagnosis of odontogenic sinusitis,” the researchers wrote.
Odontogenic sinusitis is an inflammation of the sinuses caused by dental issues, often misdiagnosed as general sinusitis due to similar early symptoms. Left untreated, the infection can spread to the face, eyes, and even the brain. The condition is typically diagnosed by both dental professionals and otolaryngologists.
According to the study, the AI model is designed to determine whether dental root apices are close to the sinus floor. It also highlights sinus locations through image enhancement. “This technology instantly identifies the proximity between tooth roots and the sinus floor when capturing a dental panoramic radiograph (DPR),” the study states. “It alerts patients to potential risks and facilitates case information sharing with otolaryngologists, providing additional reference data for clinical diagnosis.”
The research found that the YOLO 11n model outperformed other models in detecting odontogenic sinusitis. The system demonstrated a classification accuracy of 96.1%, improving diagnostic performance by 16.9% compared to non-enhanced methods and surpassing previous studies by at least 4%.
With its high accuracy and efficiency, the AI model has the potential to become a widely used diagnostic tool in dental and ear, nose, and throat (ENT) clinics, aiding in the early detection and treatment of odontogenic sinusitis.
