Chest radiography image anomaly detection method and system based on deep learning

Abstract

The invention discloses a chest radiography image anomaly detection method and system based on deep learning. The method comprises the following steps: constructing a basic identification algorithm; obtaining a chest radiography image, learning the chest radiography image according to a basic identification algorithm, and generating a classification model; classifying the chest radiography imagesto be detected according to the classification model to obtain abnormal chest radiography images and normal chest radiography images; and labeling the abnormal chest radiography image to finish detection. A basic identification algorithm is constructed to carry out deep learning on a chest radiography image; therefore, the classification model is generated, the chest radiography images to be detected can be directly classified to obtain normal chest radiography images and abnormal chest radiography images, then the abnormal chest radiography images are sent to doctors, and the normal chest radiography images are not sent to the doctors, so that the working content of the doctors is reduced, and the working efficiency of the doctors is improved; in addition, by labeling the abnormal chest radiography image, a doctor can visually observe the abnormal position and reason of the chest radiography image.

Description

technical field [0001] The present invention relates to the technical field of chest radiograph image detection, in particular to a deep learning-based chest radiograph image abnormality detection method and system. Background technique [0002] Imaging examination is an important part of today's medical field. With the development and popularization of X-ray photography technology, more and more medical institutions have introduced X-ray equipment, especially DR equipment, which is widely used in various medical institutions. A variety of business scenarios, among which, the chest radiograph image is the most common and the largest in number. By analyzing the chest radiograph image, the disease information of the patient can be identified. In the prior art, the patient is generally photographed The chest radiograph images are sent directly to the doctor for analysis. [0003] However, there are the following defects in directly sending the chest radiograph image taken by t...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method and system for abnormality detection of chest radiograph images based on deep learning, which solves the problems of the existing technology that the position of abnormal chest radiograph images cannot be judged and the work efficiency of doctors is low.

Images