Bone age evaluation method and system based on graph convolutional neural network, terminal and storage medium

Abstract

The invention provides a bone age evaluation method and system based on a graph convolutional neural network, a terminal and a storage medium. The method comprises the steps: obtaining N target epiphysis of a wrist bone image; adopting a feature extraction network model to extract initial features X of N target epiphysis and nearby areas of the wrist bone image; adopting a context feature fusion network model based on a graph convolution network, adding the initial feature X and the context feature subjected to multilayer graph convolution to obtain a final fusion feature, inputting the fusionfeature into a prediction distribution network model, and predicting to obtain development level distribution of N target epiphysis, correspondingly calculating development grade expectation and variance, development score expectation and bone maturity total score; and performing calculating to obtain a final bone age prediction value according to a bone age standard by utilizing the predicted development grade and the calculated bone maturity total score. Context feature fusion is realized through the graph convolution network to promote information communication among different local bone regions, accumulated bone age errors are avoided through a bone maturity total score loss function, and bone age evaluation accuracy and robustness are realized.

Description

technical field [0001] The present application relates to the technical field of bone age evaluation, in particular to a method, system, terminal and storage medium for bone age evaluation based on a graph convolutional neural network. Background technique [0002] Bone age is the main method to evaluate the biological age of adolescents and children, and it is used in the fields of clinical medicine, forensic science and kinesiology. Scored bone age standards (such as "Zhonghua-05" bone age standard, TW3 bone age standard) stipulate that the radius, ulna and other short phalanges of the wrist bone should be assessed separately for developmental grades. The quality control of the bone age film shooting requires the left hand to be upright, the five fingers stretched, and the joints to be avoided. After shooting, the developmental grade of each ossification center is scored separately, and the sum of the corresponding scores is calculated, and the final bone age is obtained b...

AI Technical Summary

Problems solved by technology

[0009] Aiming at the deficiencies of the prior art, this application provides a bone age evaluation method, system, terminal and storage medium based on a graph convolutional neural network, which solves the common problems in the labeling tasks of the bone age evaluation in the prior art, such as irregular palm postures or Wearing accessories leads to non-standard bone photographs, and the degree of epiphyseal development under the bone age evaluation standard can only be classified into a limited level, resulting in ambiguous labeling and other issues

Images