Toward automatic computer aided dental x-ray analysis using level set method

  • Authors:

  • Shuo Li;Thomas Fevens;Adam Krzyżak;Chao Jin;Song Li

  • Affiliations:

  • Medical Imaging Group, Department of Software Engineering and Computer Science, Concordia University, Montréal, Québec, Canada;Medical Imaging Group, Department of Software Engineering and Computer Science, Concordia University, Montréal, Québec, Canada;Medical Imaging Group, Department of Software Engineering and Computer Science, Concordia University, Montréal, Québec, Canada;Medical Imaging Group, Department of Software Engineering and Computer Science, Concordia University, Montréal, Québec, Canada;School of Stomatology, Anhui Medical University, Hefei, Anhui, P. R. China

  • Venue:
  • MICCAI'05 Proceedings of the 8th international conference on Medical Image Computing and Computer-Assisted Intervention - Volume Part I
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

A Computer Aided Dental X-rays Analysis (CADXA) framework is proposed to semi-automatically detect areas of bone loss and root decay in digital dental X-rays. In this framework, first, a new proposed competitive coupled level set method is proposed to segment the image into three pathologically meaningful regions using two coupled level set functions. Tailored for the dental clinical environment, the segmentation stage uses a trained support vector machine (SVM) classifier to provide initial contours. Then, based on the segmentation results, an analysis scheme is applied. First, the scheme builds an uncertainty map from which those areas with bone loss will be automatically detected. Secondly, the scheme employs a method based on the SVM and the average intensity profile to isolate the teeth and detect root decay. Experimental results show that our proposed framework is able to automatically detect the areas of bone loss and, when given the orientation of the teeth, it is able to automatically detect the root decay with a seriousness level marked for diagnosis.