Bone joint X-ray film automatic identification and grading method

Abstract

The invention provides a bone joint X-ray film automatic identification and grading method. According to the method, bone joint X-ray film image samples are graded according to a bone joint X-ray filmimage template library, and the bone joint conditions are judged, wherein methods for extracting contour line shape vectors of the bone joint X-ray film image template library and the image samples are same. Firstly, bone joint X-ray film images are preprocessed, and contour lines of key parts of bone joints are extracted by applying wavelet denoising, median filtering and Log edge detection operator methods. Then, the contour lines of the key parts are captured by utilizing a neighborhood feature algorithm, and filtering is performed by adopting a connected component labeling method. Finally, the contour line shape vectors of the bone joint X-ray film image samples are extracted by adopting an improved Fourier descriptor algorithm, the shape vectors of the bone joint X-ray film image template library are matched by a classification algorithm, and the bone joint conditions of the image samples are judged. The bone joint X-ray film automatic identification and grading method can replace manual identification, and is rapid and high in precision.

Description

technical field [0001] The invention relates to a method for automatic identification and grading of X-ray films of bones and joints, and relates to orthopedic image processing technology. Background technique [0002] X-rays of bones and joints are the main way to detect the condition of bones and joints clinically. At present, X-rays are manually recognized. According to the development of modern medicine, automatic methods are used for grading of bones and joints. It is very necessary to assist doctors in monitoring the conditions of bones and joints. [0003] The field of modern orthopedic medical automatic discrimination is a promising direction, but manual grading by doctors watching bone and joint X-rays is often too subjective, and the grading results of different people will inevitably vary, and the workload is heavy. The automatic recognition of orthopedic X-ray films currently uses image binary processing for correlation recognition. However, for the case where t...

AI Technical Summary

Problems solved by technology

However, for the case where the object cannot be directly binarized, due to the large degree of image distortion, the high precision of the outline extraction of bone joints makes it difficult to identify the X-ray film.

Moreover, because there are many levels, and the difference between levels is not very large, some levels cannot even be distinguished by the naked eye. A new algorithm must be developed to intercept and enlarge the key part of the bone joint to ensure the recognition accuracy. Therefore, At present, it is still difficult to realize the automatic grading of bone and joint X-ray films

Images