Investigating the Significance of New Features Extracted from Long Bones X-ray Images

Bone fracture is a medical condition where the continuity of the bone is broken. Fractures can be simple or complicated, and sometimes can be difficult to be diagnosed by specialists. Medical image processing plays an important role in improving the health care system by guiding physicians to diagnose different types of diseases, and to make the right decision regarding the appropriate treatment option. In this paper, we propose to investigate different conditions of long bones using new features that describe the physical properties of the bone. The continuous wavelet analysis, gradient analysis, bone edge detection, and Principal Component Analysis (PCA) are discussed in this work and used to extract features. By taking the tibia bone as an example, the features of the injured bone using these methods are extracted and compared with the features of the contralateral un-injured bone (i.e. the normal tibia bone) for the same patient, taking advantage of human body symmetry. The novelty of this work, the idea has not been used before in the field of features extraction from long bones or in fracture detection processes. This work will help researchers to choose the most significant features of long bones, before using them in fracture detection and classification processes.