Title: Wavelet based texture analysis and classification of bone lesions from dental CT

Authors: T. Kalpalatha Reddy, N. Kumaravel

Addresses: SKR Engineering College, Electronics and Communications Engineering Department, Anna University, Chennai, India. ' SKR Engineering College, Electronics and Communications Engineering Department, Anna University, Chennai, India

Abstract: The research presented in this article is aimed at developing a computerised automatic imaging system for classification of bone samples in dental CT. The article focuses on using texture analysis for the classification of bone samples from CT scans. The approach consists of two steps: automatic extraction of the most discriminative texture features of regions of interest in the CT medical images and creation of a classifier. A comparative study of wavelets-based texture descriptors from three families of wavelets (Daubechies|, coiflets, symlets) coupled with the implementation of a decision tree classifier based on back propagation is carried on. Principal component analysis is compared against the classical sequential methods in terms of the best recognition rate achieved and the optimal number of features. The classification performance of 100% is achieved in PCA with optimal features compared to sequential methods.

Keywords: multi-resolution analysis; texture features; wavelets; computed tomography; principle component analysis; PCA; bone lesions; dental CT; automatic imaging systems; feature extraction; medical images; classifiers; feature recognition; classification performance; dental implants; alveolar jaw bone; missing teeth.

DOI: 10.1504/IJMEI.2010.035223

International Journal of Medical Engineering and Informatics, 2010 Vol.2 No.3, pp.319 - 327

Published online: 14 Sep 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article