A wrist bone interest region cutting method based on machine learning

Abstract

The invention discloses a wrist bone interest region cutting method based on machine learning, which utilizes a machine learning method to segment a medical image, and achieves a better effect throughthe combination of central point cutting and cutting frame size self-adaption. According to the method, on the premise that it is guaranteed that the size of each individual is independent, more accurate cutting and feature extraction are conducted on each wrist bone picture. According to the method, the self-adaptive cutting frame can be generated for any height and age, and data support is provided for more accurate judgment of the wrist bone grade. The method can be applied to feature region cutting in the field of image recognition, and the recognition rate of the deep learning network isgreatly increased.

Description

technical field [0001] The present invention relates to the technical field of self-adaptive cutting of medical pictures, in particular to a method for cutting regions of interest of wrist bones based on machine learning. Background technique [0002] Now deep learning has very broad prospects in image recognition, natural language processing, speech recognition and other fields. With the wide application of deep learning technology in medical images, automatic assessment of bone age has become a hot spot. [0003] Based on variance analysis and automatic weighting, the ensemble average method of maturity, that is, the CHN method, is an evaluation method for Chinese wrist bones. This method makes a corresponding grade evaluation for 14 wrist bones. Combining the deep learning network to learn the data set of 14 wrist bone medical images to achieve the purpose of automatically judging the bone age. Due to the images of wrist bone age films of different heights and ages, the...