A diagnostic system for multiple types of pneumonia

Abstract

The invention discloses a multi-type pneumonia diagnosis system, which includes a film reading module, a feature acquisition module, a training and learning module, a diagnosis module and a result output module. The film reading module is used to store the diagnostic system on the hard disk of the computer, and input the X-ray chest image of the physical examiner through the view transmission device; the feature acquisition module is used to construct and pre-train the deep convolution block network model as an image feature extractor, The image feature extractor is used to extract the feature-forming samples of the X-ray anterior chest image; the training and learning module is used to build and train multiple dynamic learning network classifiers; the diagnosis module is used to build a two-stage integrated dynamic for each dynamic learning network classifier Learn the network model to diagnose the image and get the diagnosis prediction result; the result output module is used to output the final diagnosis result. The system greatly improves the training speed and diagnosis efficiency by dynamically learning the network model through two-stage integration.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence prediction and evaluation, in particular to a multi-type pneumonia diagnosis system. Background technique [0002] In the prior art, the use of artificial intelligence methods and models to read and diagnose X-ray positive chest images of medical examiners has the following advantages: [0003] 1. The diagnosis result is very intuitive, not only can judge whether there is a disease, but also can judge the type, stage and severity of the disease; [0004] 2. Assist in the verification of diagnostic results to ensure that the diagnostic results are accurate and stable; [0005] 3. The detection equipment is portable, easy to transport, easy to maintain, and consumes less raw materials; [0006] 4. The whole process takes less time (15s). Once the disease is detected, the patient will be sent to the doctor immediately, and the diagnosis efficiency is high; [0007] 5. Low technical...

AI Technical Summary

Problems solved by technology

However, the generalization performance of fully connected layers at the end of ResNet may not be strong enough to be a suitable classifier for classifying deep convolutional features of images (Y. Zeng, X. Xu, D. Shen, Y. Fang, and Z. Xiao, “Traffic sign recognition using kernel extreme learning machines with deep perceptual features,” IEEE Trans.Intell. Transp. Syst., vol. 18, no. 6, pp. 1647–1653, 2017.)

Moreover, in the process of screening and diagnosing various pneumonias, only a single classifier model was used in the past for the deep convolution features of images, and the reliability of the diagnostic results is difficult to guarantee. These defects are greatly affected in the process of intelligent disease diagnosis. restrict

Images