MANILA: A team of researchers from the Philippines and Taiwan has developed a cutting-edge artificial intelligence (AI) model that accurately identifies tooth and sinus structures in dental panoramic X-rays — achieving a remarkable 98.2% accuracy.
The research, led by the Ateneo Laboratory for Intelligent Visual Environments (ALIVE) in collaboration with Chang Gung Memorial Hospital, National Cheng Kung University, Chung Yuan Christian University, and Ming Chi University of Technology, introduces a deep learning-based YOLOv7 11n object detection model tailored for diagnosing odontogenic sinusitis.
✅ "This lightweight, real-time system reduces patient exposure to radiation, lowers costs, and supports more precise clinical decision-making," researchers noted.
Why It Matters
Odontogenic sinusitis is a lesser-known but serious condition caused by dental infections extending into the sinuses. It is often misdiagnosed as general sinusitis due to overlapping symptoms — a delay in diagnosis can lead to severe complications, including infections spreading to the face, eyes, and even the brain.
The AI model aims to fill this diagnostic gap by automatically determining if dental root apices are near the maxillary sinus floor. Through image enhancement, the system highlights sinus regions on dental panoramic radiographs (DPR), alerting both patients and clinicians to potential risk areas in real time.
Key Findings
- The YOLO 11n model showed a 96.1% classification accuracy, outperforming other AI tools.
- It improved diagnostic accuracy by 16.9% over non-enhanced methods.
- The model surpassed previous studies by at least 4%, setting a new benchmark in dental imaging.
Impact and Future Use
The AI tool has the potential to be widely adopted in dental clinics and ENT departments, significantly enhancing early diagnosis and treatment planning. It also allows seamless data sharing between dentists and otolaryngologists, streamlining interdisciplinary collaboration.
With its high speed, accuracy, and cost-effectiveness, the model is poised to become a critical asset in the era of AI-assisted diagnostics in healthcare.