Subscribe Now Subscribe Today
Science Alert
 
Blue
   
Curve Top
Journal of Applied Sciences
  Year: 2009 | Volume: 9 | Issue: 11 | Page No.: 2106-2114
DOI: 10.3923/jas.2009.2106.2114
 
Facebook Twitter Digg Reddit Linkedin StumbleUpon E-mail

Fully Automatic Extraction of Panoramic Dental Images from CT-Scan Volumetric Data of the Head

H. Akhoondali, R.A. Zoroofi and G. Shirani

Abstract:
In this study, a fully automatic technique is proposed for the extraction of panoramic dental images from CT-scan images of the teeth. The process is performed in three steps: Mandibular region isolation, mandibular curve creation and panoramic image extraction. First the mandibular region is isolated. To do this the maxilla and mandible are separated from each other by finding the slice separating them. Then an image of the maximum intensity projection of the slices containing the mandible is obtained. The projection of the mandible is then isolated from the other bones. Using the isolated mandible region an initial curve is created. Nine control points are then selected on the curve. These control points are used to create a fine curve using cubic splines. The formula for each spline segment is used to find the perpendicular lines to the spline on each point on the spline. These perpendicular lines are then used in all slices (mandible and maxilla) to create the panoramic images. The algorithm has many advantages over earlier methods. Results using multiple datasets have shown that the algorithm is robust.
PDF Fulltext XML References Citation Report Citation
 RELATED ARTICLES:
  •    Radiographic Study of the Location of Mental Foramen in a Randomly Selected Asian Indian Population on Digital Panoramic Radiographs
How to cite this article:

H. Akhoondali, R.A. Zoroofi and G. Shirani, 2009. Fully Automatic Extraction of Panoramic Dental Images from CT-Scan Volumetric Data of the Head. Journal of Applied Sciences, 9: 2106-2114.

DOI: 10.3923/jas.2009.2106.2114

URL: https://scialert.net/abstract/?doi=jas.2009.2106.2114

COMMENTS
01 September, 2014
Mayank Sharma:
Nice approach sir
I am following the your paper for similar work.
But i am stuck at one point.
Is segmentation is done at whole set of volute as a 3d segmentation or is it at individual images ?
Which algorithm you used for segmentation ?
COMMENT ON THIS PAPER
 
 
 

 

 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 

Curve Bottom