A method and electronic equipment for detecting bone quality based on CT images

Abstract

The invention discloses a method and electronic equipment for detecting bone quality based on CT images. The method for detecting bone quality includes performing super-resolution reconstruction on lumbar spine CT images, and using an image classification network to classify lumbar spine super-resolution images, etc. Step, by improving the resolution of the lumbar spine CT image and removing a considerable part of the noise in the original CT image, the accuracy of the classification network for the classification of the input image is improved, thereby improving the accuracy and accuracy of bone quality detection. The super-resolution reconstruction network includes a preliminary feature extraction module, a multi-channel feature extraction module, an MF fusion module, and an image reconstruction module. The MF fusion module includes multiple sub-modules, and the sub-modules are set correspondingly to the multi-channel feature extraction module; The way to fuse feature information of different scales, improve the filtering effect of feature fusion module on useless information, and retain more useful information for image reconstruction.

Description

A method and electronic device for detecting bone condition based on CT images technical field The present invention belongs to medical technology field, specifically, relate to a kind of method based on CT image detection bone condition and electronic equipment. Background technique Osteoporosis (OP) is a systemic metabolic bone disease characterized by low bone mass and destruction of bone tissue microarchitecture. The disease is clinically manifested by bone pain and susceptibility to fractures. Since different bone densities have differences in the absorption of X-rays, it is possible to Quantitative diagnosis of osteoporosis by lumbar spine CT images. Studies have shown that (for example, the application number is 201910913768.0 Invention patent), using artificial intelligence technology, can automatically detect the condition of lumbar vertebrae in CT images. However, set Due to the influence of objective factors such as equipment performance and cost, the...

AI Technical Summary

Problems solved by technology

However, affected by objective factors such as equipment performance and cost, the resolution of the original CT image is usually relatively low, which has a great adverse effect on the classification effect of the neural network, thereby limiting the accuracy and accuracy of the final bone condition detection

Images