A Method for Automatic Segmentation and Labeling of X-ray Spine

Abstract

The invention belongs to the technical field of medical image segmentation, and relates to an automatic spine segmentation and recognition method for X-ray films. The segmentation strategy is thick first and then thin, and the spinal column area can be quickly located by using the constructed deep neural network, and then the subsequent fine segmentation of the vertebral body can be realized. ; The segmentation accuracy is high, and the neural network constructed takes into account the semantic and edge characteristics of the spine. On this basis, the image morphology operation is used to optimize the processing, so that the segmented vertebral body is independent and can maintain a complete edge. Smart measurement lays the foundation.

Description

Technical field: [0001] The invention belongs to the technical field of medical image segmentation, and relates to an X-ray spine automatic segmentation and recognition method based on a deep neural network. Background technique: [0002] The spine is an important part of the human body. Its anatomical structure is complex, mainly including three important parts: vertebral body, intervertebral disc and spinal cord. It is the structural basis of many spinal diseases, such as juvenile idiopathic scoliosis and lumbar degenerative scoliosis. , Lumbar disc herniation, lumbar spinal stenosis, osteoporosis, hyperosteogeny, spinal tuberculosis, spinal tumors, etc. Spinal disease has become one of several persistent diseases that affect public health and brings a huge economic burden to society. In the traditional diagnosis of spinal diseases, it is necessary to comprehensively consider the patient's symptoms and imaging tests to make a diagnosis. For different diseases, it is neces...

AI Technical Summary

Problems solved by technology

[0003] At present, spine X-ray films mainly include 24 vertebral bodies (cervical vertebrae 1-7, thoracic vertebrae 1-12, lumbar vertebrae 1-5), sacrum and ilium. The medical parameters of each part are still manually measured and deduced and calculated. However, manual measurement The following problems exist: 1) The diagnosis of spine X-ray film involves the measurement and derivation and calculation of a large number of medical parameters, the process is complicated, and the reading time is long; 2) Compared with CT and MR images, the imaging clarity of X-ray film is poor, and the edge of the spine is easy to blur , and there are many interference components, such as ribs and organ soft tissue, etc., and manual measurement has unavoidable errors; 3) strong professionalism, difficult learning and long cycle, very few spine surgeons who can master standardized measurement and diagnosis techniques, However, the incidence of spinal deformity is usually widespread, it is difficult to give correct diagnosis guidance, and the disease is delayed; 4) The repeatability is poor, the symptoms are different, the repetitive labor such as manual measurement and calculation is heavy, and measurement errors are often caused by forgetting or negligence, which affects the follow-up treat

At present, there are few relevant reports on the automatic and precise segmentation technology of vertebral bodies for spinal X-ray films

Images