On the influence of image enhancement on fractal-based automatic osteoporosis detection from calcaneum x-rays

The goal of this research is to investigate the effectiveness of various classical automatic image enhancement methods in the framework of calcaneum bone characterization in digitally-acquired x-ray images. The decision on the presence or absence of osteoporosis signs is taken within a nearest neighbor decision rule by using fractal parameters, such as the fractal dimension. A successful image processing chain (acquisition, enhancement and analysis) could lead to an automatic scoring procedure of the bone degradation in the diagnosis of osteoporosis or other bone-modifying diseases, by using cheap imaging equipment, such as consumer digital cameras and general-purpose computers.