Method for detecting scoliosis, equipment and medium

Abstract

The embodiment of the invention relates to a method for detecting scoliosis, equipment and a medium. According to the method, a plurality of spine segments of a spine are identified from a spine image based on a trained deep learning model; a minimum enclosing rectangle of each spine segment is determined; the slope of the upper and lower edges of each minimum bounding rectangle is determined so as to determine the position of a first minimum bounding rectangle and a second minimum bounding rectangle for each spinal curvature segment of the spine based on the determined slope, the first minimum bounding rectangle being associated with the upper end vertebra of the spinal curvature segment and the second minimum enclosing rectangle is associated with the lower end vertebra of the spine bending section; for each spine bending segment, a lateral bending angle of the spine bending segment is determined based on the upper edge of the corresponding first minimum enclosing rectangle and the lower edge of the corresponding second minimum enclosing rectangle; and based on the determined scoliosis angle, the existence state of the spine with respect to scoliosis is determined. Therefore, the detection efficiency can be improved.

Description

technical field [0001] Embodiments of the present disclosure relate generally to the fields of image processing and recognition, and more particularly to a method, apparatus and medium for detecting scoliosis. Background technique [0002] The normal human spine has four curvatures in the anteroposterior direction. The cervical segment is convex forward, the thoracic segment is convex backward, the lumbar segment is convex forward, and the sacral segment is convex backward. It is similar to an "S" shape and is called the physiological curvature of the spine. Scoliosis refers to the deviation of the spine from the midline to both sides. It is a three-dimensional deformity of the spine, including sequence abnormalities in the coronal, sagittal, and axial positions. Scoliosis can be divided into thoracic scoliosis, lumbar scoliosis, and thoracolumbar scoliosis according to the different parts of the occurrence. Usually, scoliosis occurs in the thoracic spine, the lower lumbar ...

AI Technical Summary

Problems solved by technology

[0004] However, there is currently no mature technology for detecting scoliosis, and the detection of each spine is mainly performed by the inspector by manually drawing detection auxiliary lines on spinal images (for example, X-ray films of the spine, etc.). The existence status of the spine (that is, the problem of detecting whether the spine has scoliosis), and in the case of determining the presence of scoliosis in the spine, it is also necessary to determine the type of scoliosis by looking up the table, so the labor cost is very high and the efficiency is very low , and the examiner must have the relevant professional knowledge to accurately determine the existence of the spine with respect to scoliosis and the type of scoliosis

For example, it usually takes several minutes or tens of minutes for an inspector (e.g., a clinician) to complete the determination of scoliosis on a spine image, and different levels of inspectors have different levels of expertise due to their different levels of expertise. There may also be some differences in the conclusions

Images