Musculoskeletal (MSK)AI / InformaticsResearch

Hybrid Deep Learning System Detects and Classifies Nasopalatine Canal on CBCT

European archives of oto-rhino-laryngology : official journal of the European Federation of Oto-Rhino-Laryngological Societies (EUFOS) : affiliated with the German Society for Oto-Rhino-Laryngology - Head and Neck Surgeryyesterday

A YOLOv10 model detected the nasopalatine canal on CBCT with 100% recall, and MobileNetV3Large classified its morphology with 81.48% accuracy in a retrospective study of 135 patients, showing feasibility for pre-procedural screening.

  • Retrospective analysis of 135 CBCT scans; YOLOv10 achieved 99.5% mAP50 for detection, and MobileNetV3Large had 81.48% accuracy for morphological classification.
  • Classification performance: precision 82.60%, F1 score 81.47%; EfficientNetV2B0 achieved 79.26% accuracy.
  • Limitation: findings are preliminary; external validation and expert verification are needed before clinical use.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.