Deep Learning based Tooth Multi-Disease Detection in Dental Diagnostics Di s Problemleri Tanisi i in Derin grenme tabanli oklu-Hastalik Tespiti


Hussein A., GÜNEÇ H. G., Aydin K. C., Ates H. F.

33rd IEEE Conference on Signal Processing and Communications Applications, SIU 2025, İstanbul, Türkiye, 25 - 28 Haziran 2025, (Tam Metin Bildiri)

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/siu66497.2025.11112385
  • Basıldığı Şehir: İstanbul
  • Basıldığı Ülke: Türkiye
  • Anahtar Kelimeler: Computer vision, deep learning, transformers, object detection, dental diseases, panoramic radiographs
  • Sağlık Bilimleri Üniversitesi Adresli: Evet

Özet

This work uses deep learning to automatically classify a set of dental pathologies from wide-field dental X-rays named panoramic radiographs. 9,573 adult and child patients' X-ray images form our dataset, each of which is manually annotated to 19 different dental pathologies. The proposed method leverages advanced deep learning models to diagnose a set of oral diseases and achieves state-of-the-art results. YOLO and DETR models are compared for their dental problem detection and classification accuracy. This complete AI-based method produces quick and ac-curate diagnoses of oral health, which allows dental practitioners to provide more informed decisions quickly and reliably. With evidence-based interpretation of AI results, we believe that the proposed method is a sensible way of supplementing dentists' diagnoses.