Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
3278018 | Revista Española de Enfermedades Metabólicas Óseas | 2009 | 7 Pages |
Abstract
Fractal dimension (box-counting method), lacunarity (sliding box-method) and entropy properties (Grey Level Co-occurrence matrix) were calculated in three regions of the proximal femur in a sample of 20 X-ray images in anteroposterior projection. The Wilcoxon related samples test was applied to compare the values of each variable for every region. Correlation coefficients were calculated to analyze the relationships between variables (CI 95%). Significant differences for entropy values were found between head and neck areas. However, no statistically differences variations in fractal dimension or lacunarity were observed. Significant strong inverse correlations were found between the fractal dimension and the lacunarity in all regions and between the entropy and the fractal dimension and lacunarity in proximal shape region. These results suggest that entropy measurements may be useful for characterizing trabecular bone texture from proximal femur X-ray images.
Related Topics
Health Sciences
Medicine and Dentistry
Endocrinology, Diabetes and Metabolism
Authors
J. RÃos-DÃaz, J.R. Caeiro Rey, J.J. MartÃnez Payá, S. DapÃa Robleda, M.E. del Baño Aledo,