Recognition of root canal orifices in video sequences as a future support system during endodontic treatment

J Endod. 2009 Oct;35(10):1400-3. doi: 10.1016/j.joen.2009.06.013. Epub 2009 Aug 22.

Abstract

Introduction: The objective of this study was to show the practical application of computer-aided techniques for detecting root canal orifices through the access cavity using a video camera mounted on a microscope.

Methods: A minimum distance classification image recognition algorithm was tested in an in vitro study to assess the possibilities of computer-aided recognition of root canal orifices. A Motic DM143 digital stereo microscope (Motic Germany GmbH, Wetzlar, Germany) was used because it includes a video camera that can be connected via USB1.1 to any computer.

Results: The newly developed software is capable of communicating with a video camera and can automatically detect the root canal orifices in all teeth used in this study. A total of 165 extracted human teeth (molars and premolars) were used as test data to collect 8,250 images via screenshots for the evaluation of the detection quality. The software provided a detection sensitivity of 90.1%, with only 11.9% of the images as false-positive detections.

Conclusion: The study shows that computer-aided recognition of root canal orifices with video cameras is possible.

MeSH terms

  • Adult
  • Aged
  • Algorithms
  • Bicuspid / anatomy & histology
  • Dental Pulp Cavity / anatomy & histology*
  • False Positive Reactions
  • Humans
  • Image Processing, Computer-Assisted
  • Microscopy / instrumentation
  • Middle Aged
  • Molar / anatomy & histology
  • Pattern Recognition, Automated / methods*
  • Root Canal Preparation* / instrumentation
  • Root Canal Preparation* / methods
  • Sensitivity and Specificity
  • Software
  • Therapy, Computer-Assisted
  • Video Recording* / instrumentation