Detecting femur fractures by texture analysis of trabeculae

30% of women and 13% of men worldwide suffer from osteoporotic bone fractures worldwide. In large hospitals, doctors need to visually inspect a large number of X-ray images to identify the fracture cases, which typically constitute about 12% of all the X-ray images examined. Automated fracture detection can help to screen for obvious cases and flag suspicious cases for closer examinations. This paper describes a method of detecting femur fractures by analyzing trabecular texture patterns. Test results show that it is more accurate than an existing method based on neck-shaft angle. Moreover, combining the methods further improve the overall performance of fracture detection.