You are here

Fast Feature Extraction and Non Iterative Multi Modal Image Registration for Orthopaedic Trauma using Local Phase Features

Project Type: 
PDF-led

The purpose of this research project is to determine the effectiveness and demonstrate the clinical feasibility of using 3D ultrasound imaging modality in image guided interventions especially in orthopaedic trauma applications.

Project Leader(s): 

Postdoctoral fellow: Dr. Ilker Hacihaliloglu, Department of Orthopaedics, University of British Columbia

Lead faculty member: Dr. David Wilson, Department of Orthopaedics, University of British Columbia

The Canadian National Trauma Registry have recorded that out of 109,738 major injuries occurring in 1999, 4531 had a pelvis fracture. Traditional intraoperative imaging modality in orthopaedic surgery is 2D fluoroscopy which makes identification of 3D bone surfaces very difficult and exposes the patient and the surgical team to harmful ionizing radiation. Ultrasound has traditionally been used to image the body's soft tissue, organs, and blood flow in real time. However, the use of ultrasound imaging in orthopaedic surgery is relatively uncommon because extracting useful information from ultrasound images is still very challenging due to poor image quality. The purpose of this research proposal is to determine the effectiveness and demonstrate the clinical feasibility of using 3D ultrasound imaging modality in image guided interventions especially in orthopaedic trauma applications. We will develop automatic and accurate new methods that will extract useful information from three dimensional ultrasound images in milliseconds which is not achieved before. Ultimately we aim to improve performance of the surgery by providing better assessment which will in turn improve the surgical outcome and decrease the total operation time and total cost. The proposed system will be also invaluable for all image guided intervention systems that make us of this modality.