Texture analysis for trabecular bone x-ray images using anisotropic morlet wavelet and rényi entropy

Authors:
Ahmed Salmi EL Boumnini El Hassani;Mohammed El Hassouni;Rachid Jennane;Mohammed Rziza;Eric Lespessailles
Affiliations:
LRIT, Faculty of Science, University Mohammed V - Agdal, Rabat, Morocco;DESTEC-FLSHR, University Mohammed V - Agdal, Rabat, Morocco;PRISME Laboratory, University of Orléans, Orléans, France;LRIT, Faculty of Science, University Mohammed V - Agdal, Rabat, Morocco;Hospital of Orleans, IPROS, Orléans, France
Venue:
ICISP'12 Proceedings of the 5th international conference on Image and Signal Processing
Year:
2012

Citing 3
Cited 0

Image analysis with two-dimensional continuous wavelet transform

Signal Processing
An introduction to ROC analysis

Pattern Recognition Letters - Special issue: ROC analysis in pattern recognition
A fully-anisotropic Morlet wavelet to identify dominant orientations in a porous medium

Computers & Geosciences

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose a new method based on texture analysis for the early diagnosis of bone disease such as osteoporosis. Our proposed method is based on a combination of four methods. First, bone X-ray images are enhanced using the algorithm of Retinex. Then, the enhanced images are analyzed using the fully anisotropic Morlet wavelet. This step is followed by the quantification of the anisotropy of the images using the Rényi entropy. Finally, the Rényi entropies are used as entries for a neural network. Applied on two different populations composed of osteoporotic (OP) patients and control (CT) subjects, a classification rate of 95% is achieved which provides a good discrimination between OP patients and CT subjects.