Bone age prediction method

Abstract

The invention discloses a bone age prediction method. A preset training set is trained in advance to form a bone age prediction model, and the bone age prediction model is used for bone age prediction. According to the technical scheme, through a brand-new attention mechanism adding mode, tan attention mechanism is added in a targeted mode according to different convolution layer information coupling degrees in the bone age prediction model, and the problem that a traditional attention mechanism is not easy to converge is solved; gender information is directly added into the neural network model for end-to-end training, and the accuracy of bone age prediction is further improved while the training process is optimized; the bone age prediction model used in the bone age prediction method isvisually verified through the attention mechanism, it is reflected that the model more pays attention to the hand bone part where the development degree of a patient is presented in a centralized mode, and a good prediction effect is achieved.

Description

technical field [0001] The invention relates to the field of bone age prediction, in particular to a bone age prediction method. Background technique [0002] Bone age is a physiological age that characterizes the developmental level of adolescents. It is widely used in height prediction, athlete selection, and medical and health fields. The method of quantitative determination of bone age is called bone age prediction. Bone age prediction is divided into traditional artificial bone age prediction methods, bone age prediction methods based on machine learning, and bone age prediction methods based on deep learning that have emerged recently. [0003] Traditional bone age prediction methods generally include counting methods, atlas methods (such as GP atlas methods) and scoring methods (such as TW, CNH, and Zhonghua 05 methods). The counting method is to count each ossification center of the palmar bone and the epiphyseal forming area. Different values ​​correspond to differ...

AI Technical Summary

Problems solved by technology

Although these methods use deep learning technology in the field of bone age prediction to improve the prediction speed and accuracy of the model, these methods do not make the model focus on the key bone areas for bone age prediction, making the model robust and reliable. The interpretability is poor, and there is still room for optimization in terms of accuracy

Images