Detection of fracture and quantitative assessment of displacement measures in pelvic X-RAY images

Fracture detection in cases of traumatic pelvic injuries is crucial for rapid and successful patient treatment. Initial diagnosis is typically made via X-ray images, which can be challenging and time-consuming to analyze due to their low resolution and the differing visual characteristics of fractures by their location. This paper presents a fracture detection method for the pelvic ring based on Discrete Wavelet Transform and boundary tracing applied to windows extracted from the ring, as defined by prior automated region segmentation via a deformable Spline/ASM model. Results so far are promising, and indicate that the approach can extract useful features for a trauma decision-making system to assist physicians and improve patient care.