Tooth Detection with Convolutional Neural Networks


Oktay A. B.

Medical Technologies National Congress (TIPTEKNO), Trabzon, Turkey, 12 - 14 October 2017 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/tiptekno.2017.8238075
  • City: Trabzon
  • Country: Turkey
  • Keywords: convolutional neural networks, deep learning, dental image, X-RAY IMAGES, HUMAN IDENTIFICATION, DENTAL RADIOGRAPHS, SYSTEM, CLASSIFICATION, TEETH
  • Yıldız Technical University Affiliated: No

Abstract

Detection of teeth from dental panoramic Xray images is the a crucial step for the computerized tooth applications. In this a paper, we present a method for detecting teeth in dental panoramic X-ray images with Convolutional Neural Networks (CNN). After finding the mouth gap, possible positions of three tooth type (incisors, premolars, and molars) are determined. Teeth are detected with a modified version of AlexNet architecture where multi-class classification is performed. The accuracy of the method is over 90% and the presented method can be used as the first step of the computerized dental applications.