Image preprocessing method and system for reducing interference features of bone age evaluation model

Abstract

An image preprocessing method for reducing interference features of a bone age evaluation model, comprising the following steps: 1) hand region detection; 2) hand shape image segmentation; 3) palm angle adjustment. And provide an image preprocessing system for bone age evaluation model to reduce interference features, including hand region detection module, hand shape image segmentation module and palm angle adjustment module, wherein the hand region detection module is used for positioning of hand bones , remove the interference features caused by different positions of the hand region in the image; the hand image segmentation module is used for the segmentation of the hand bones and the unification of the background, and removes the interference caused by different backgrounds; the palm angle adjustment module is used for the hand bone Unification of angles removes interference caused by different angles. The present invention adjusts the position and angle of the hand bones through preprocessing means, removes the background, and achieves the purpose of data amplification by reducing interference features, so that model training is more effective, and the classification ability and robustness are better.

Description

technical field [0001] The invention relates to the field of bone age evaluation, and is used for preprocessing hand bone X-ray images before training a neural network model, so that the training is more efficient and the model effect is better. Background technique [0002] Bone age is one of the important indicators in the detection of physical development of children and adolescents, the inspection of endocrine diseases, and the selection of sports materials, so bone age evaluation has a wide range of applications. At present, bone age evaluation methods include manual methods, computer-aided methods, and deep learning methods. Due to the complexity of bone age assessment and the excessive human subjective factors, the first two methods cannot accurately assess bone age; with the development of deep learning technology, more and more researchers have begun to use deep learning for bone age assessment. . [0003] Existing deep learning models are mostly data-driven and r...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the shortcomings of the existing bone age evaluation methods, such as uneven bone age X-ray data quality, many interference features, difficult training of deep learning models, and poor classification ability and robustness, the present invention provides a method for reducing interference features. Image preprocessing method and system for bone age evaluation model with better classification ability and robustness to reduce interference features

Images