Scoliosis screening method based on deep learning

Abstract

The invention discloses a scoliosis screening method based on deep learning. According to the method, the screening process can be simplified technically, the screening accuracy is improved, the manpower and material resource cost is reduced, and powerful technical support is provided for comprehensively achieving scoliosis screening. Meanwhile, an X-ray film is replaced to a certain degree to carry out progress monitoring and treatment effect evaluation on scoliosis, X-ray radiation is reduced, and the method has wide clinical application prospects and social and economic values. The method can directly input spinal images obtained from any terminal into a model for preprocessing, whether spinal deformity exists or not is judged through a deep learning model, the severity degree of the spinal deformity is evaluated, finally risk reports and medical advices are given, and the method is applied to the field of scoliosis screening.

Description

technical field [0001] The invention belongs to the field of scoliosis screening, and more specifically relates to a scoliosis screening method based on deep learning. Background technique [0002] Scoliosis is a disabling and fatal deformity, the incidence rate of which is reported in the literature is 2% to 4%, among which idiopathic scoliosis (IS) is the most common, and 70% to 80% are prone to occur Primary and middle school students in adolescence. [0003] Due to the unclear etiology and insidious onset of the disease, there are often no obvious clinical symptoms in the early stage, and now most middle school students live independently or live in school, parents, schools and primary hospitals are often difficult to detect in time due to lack of understanding, causing many students to miss the precious gold. Timing of treatment. At the same time, because adolescents are at the peak of growth and development, deformity is easy to progress, causing appearance deformity...

AI Technical Summary

Problems solved by technology

[0005] However, traditional scoliosis screening requires a lot of manpower and material resources, and the positive predictive rate of screening is low, resulting in a large number of non-scoliotic patients receiving unnecessary radiation radiation, and it is difficult to seek consultation and conservative treatment after large-scale screening. Screening methods at home and abroad can no longer meet the needs of large-scale scoliosis screening

[0006] At present, it has been reported in the literature that using infrared thermal imaging body position examination and ultrasound examination for the screening of idiopathic scoliosis can effectively reduce the radiation of X-ray films, but these two methods still have the disadvantages of expensive equipment, low accuracy, It is difficult to carry out large-scale screening due to the high technical requirements of inspectors and the difficulty of using equipment in large-scale screening sites.

There is a certain correlation between ultrasonic measurement and X-ray film Cobb angle measurement, but compared with the traditional X-ray film Cobb angle measurement method, the ultrasonic angle is based on different landmarks and cannot accurately reflect the Cobb angle. The former is 15% smaller than the latter 37%, while the equipment is expensive and difficult to use in large-scale scoliosis screening sites

Infrared thermal imaging body position examination is based on the asymmetric activity of paraspinal muscles in patients with scoliosis and non-scoliotic patients, and distinguishes the temperature difference of paraspinal muscle activities. It is only suitable for qualitative research on scoliosis, and there are high false positives. Negative and false positive rates still cannot meet the requirements of large-scale screening

Images