Medical Image Management And Analysis System Based On Web

The number of images taken in a hospital every day is a very large number.How to manage these images is a very important task.It also means that more and more medical images require doctors to diagnose.This rapidly growing huge image data is extremely scarce.Secondly,most medical image management systems at home and abroad currently require hospitals to install clients.Web-based software can effectively solve this problem.At the same time,Medical Image Management Systems are rarely combined with Computer-aided Diagnosis Systems.These two systems are independent,This reduces the efficiency of the hospital.The medical image management and analysis system proposed in this article is a medical software developed in B/S mode after investigating the workflow of relevant departments of the hospital.This software can realize the entire process from patient visit to printing of graphic reports for convenience Manage medical images and online readings,and add computer-aided diagnosis on this basis.Due to the difficulty of collecting medical image data and the limited space of the paper,in the computer-aided diagnosis module,this article only implements online The automatic diagnosis of confocal microscope images of fungal keratitis is provided to ophthalmologists for reference.Fungal keratitis is a disease of ophthalmology.The disease is caused by trauma to the eyes of the patient and infection by external bacteria.Its blinding rate is second only to cataracts.There are many methods for checking fungal keratitis.Confocal microscopy is a biopsy technique that can perform a non-invasive rapid examination of the live cornea.It is a powerful tool for diagnosing fungal keratitis.The image data used for the experiment in this article is positive It was taken with a confocal microscope.The specific work of this paper is as follows:(1)This article firstly obtains the work flow of the relevant departments of the hospital through investigation and data review,then conducts demand analysis and design of the system,and uses SpringBoot+MybatisPlus+LayUI+Mysql to build a B/S mode Medical Image Management and Analysis System;In the architecture,the user layer,application layer,data layer and other layered settings are used.The user layer divides the system into medical staff and background management personnel interface,and the data layer uses Mysql database server.After the system development was completed,the system was further tested and found to be stable.Eventually,the entire workflow of patients from printing appointments to printing graphic reports after diagnosis is completed,which can improve the efficiency of the hospital.(2)In the computer-aided diagnosis module of the system developed in this paper,this paper studies the classification of confocal microscopy images of fungal keratitis in order to obtain a neural network model for the classification of this image.Firstly,through the experimental comparison of different initial learning rates using AlexNet,the initial learning rate of 0.0001 was used to train the three networks of AlexNet,ZFNet,and VGG16 to obtain three different classification models.Secondly,the three networks were integrated through the method of relative majority voting and weighted average method through ensemble learning.Finally,it was found that the weighted average method improved the performance of the previous three networks and achieved a classification accuracy of 0.9964.rate.Finally,the network model is integrated into the developed system to realize the online automatic diagnosis of confocal microscopy images of fungal keratitis for reference by medical staff.