Device for detecting spinal osteoporotic compression fractures using chest and abdominal frontal x-ray

By processing chest and abdominal X-ray images using convolutional neural networks, osteoporotic compression fractures of the spine can be automatically detected, solving the problems of missed diagnosis and misdiagnosis in traditional methods, and achieving efficient fracture location detection and timely treatment reminders.

CN115701344BActive Publication Date: 2026-07-21THE CHINESE UNIVERSITY OF HONG KONG
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
THE CHINESE UNIVERSITY OF HONG KONG
Filing Date
2021-08-02
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

Existing chest and abdominal X-ray images are prone to missing diagnoses of osteoporotic compression fractures of the spine, especially mild fractures. Furthermore, insufficient medical expertise can lead to misdiagnosis. Traditional methods are insufficient to effectively detect and alert patients to early treatment.

Method used

Convolutional neural networks (CNNs) were used to process anteroposterior X-ray images of the chest and abdomen. Residual modules and pyramid-like structures were used to extract features. Combining classification and regression branches, the location of osteoporotic compression fractures of the spine was automatically detected and annotated in the X-ray images.

Benefits of technology

It has improved the accuracy of detecting osteoporotic compression fractures of the spine, reduced the workload of doctors, avoided missed diagnoses, promptly reminded patients to undergo examinations and treatment, and reduced the risk of the condition worsening.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115701344B_ABST
    Figure CN115701344B_ABST
Patent Text Reader

Abstract

A device for detecting spinal osteoporotic compression fractures using chest and abdomen frontal X-ray images, comprising a processor, which executes the following steps when running a computer program stored by a computer readable storage medium: S1, image preprocessing is performed on the original chest and abdomen frontal X-ray image; S2, the preprocessed X-ray image is input into a convolutional neural network for processing, wherein the convolutional neural network is configured to extract image features using a residual module and extract semantic features using a network module with a class pyramid structure, so as to realize the extraction of osteoporotic compression fracture position information in the X-ray image; S3, the extracted features are classified and regressed through a prediction output part, and the detection result of the osteoporotic compression fracture is output. Based on the chest and abdomen frontal X-ray image which is very common in clinical diagnosis, the doctor can find the position where the spinal osteoporotic compression fracture is likely to occur, and timely remind the doctor and the patient.
Need to check novelty before this filing date? Find Prior Art