Rapid Detection of Scoliosis Based on Digital Image Processing

OBJECTIVE: To design a rapid scoliosis detection system for general survey. METHODS: The camera was used to take the upright image of human back, and then the region of interest was extracted. After image preprocessing, the feature points of human back spine were extracted. The feature points were fitted into the spine contour curve. Finally, the Cobb angle of scoliosis was calculated according to the contour curve, and the scoliosis degree was judged. RESULTS: The outline curve of the spine can be obtained by this method, and the Cobb angle of scoliosis can be calculated. It can detect scoliosis quickly, effectively and accurately. CONCLUSIONS: Compared with the traditional methods, the digital image processing method can achieve rapid and non-destructive detection of scoliosis, save a lot of manpower and material resources, and is of great significance to the national survey of adolescent scoliosis.