Automatic Measurement On CT Images For Patella Dislocation Diagnosis

With the development of medical imaging technology such as CT (Computerized Tomography), MRI (Magnetic resonance Imaging), PET (Positron emission Tomography), X-ray Imaging, Medical Imaging has played an indispensable part in clinical diagnosis and treatment of diseases. Medical Imaging could display the disease and the change of the body organization intuitively, and provide strong scientific security. With the increasing number of medical images and fast development of medical image processing technology, people are eager to utilize the computer technology to improve the efficiency and accuracy of medical diagnosis.This paper focuses on automatically measurement on CT image for patella dislocation diagnosis. Patella Dislocation is a common symptom, which often happen in youngsters and young women. The symptom happens when patella slips out of femoral trochlear. When patella deviated from femoral trochlear, we call the symptom patellar subluxation. Besides, when people get older, the shape of patella and femur will change because of the wear and tear in the knee joint, which will also cause patella dislocation, patella subluxation and arthritis.This work designed an automatic diagnostic system to patella dislocation, the input to the system is a set of patella CT images, and the output is the diagnosis results. This paper first introduced different kinds of segmentation techniques to medical images. Medical image segmentation technology is developing fast, mainly classified as threshold-based, region-based, edge-based, active contour models and other artificial intelligence-based. Yet because of the complexity of medical images, there is not one technique which could be used under all circumstances in medical image segmentation. Here we propose a new method to improve the accuracy of segmentation on CT patella images, which makes use of the similarity of adjacent frames and is combined with region-based active contour model.Besides, this paper automatically measures the segmented CT images, including angle and distance between femur and patella, which is essential to clinical diagnosis. In the end, we compare our work with manual diagnosis. As experimentally demonstrated, the results obtained with our system are highly consistent with those manually made by experienced doctors.