The invention discloses a multi-subgroup hip joint
dysplasia classification method and
system, and belongs to the field of
artificial intelligence. Comprising the following steps: acquiring a
pelvis normal position X-
ray film image of a patient, and preprocessing the image; a rough classification network is constructed, general features in the image are extracted to obtain a rough
classification result for identifying the rough type of the
dysplasia of the hip joint, and the general features comprise contours and textures; a fine classification network is constructed, on the basis of the graph
convolution network, classification results are refined by additionally increasing expression information of anatomical key points on the basis of coarse classification results, and the anatomical key points comprise the highest point of the ilium, the outermost edge of the
acetabulum, the center point of the
femoral head, the teardrop point and the lowest point of the
ischial tuberosity; and outputting a fine classification network result. According to the method, the
system generalization can be improved while the classification efficiency and accuracy are improved.