Design And Implementation Of Auxiliary Diagnosis And Treatment System For Mammography Based On Convolutional Neural Network

In recent years,the incidence of breast diseases has been increasing year by year,and Mammary gland molybdenum target X-ray detection of breast is an effective method.With the wide application of medical images in focus detection and image processing technology,it is of great significance to construct an auxiliary diagnosis and treatment system using image processing algorithms for the identification of lesions in medical images.Meanwhile,the remote access capability provided by the system can enable hospitals with excellent medical resources to provide remote consultation services for patients,thus alleviating the problem of unbalanced regional development of medical resources.The main research objective of this paper is to build an auxiliary diagnosis and treatment system of mammography based on convolutional neural network,and the main content includes image processing and the establishment of website service.The purpose of image processing service is to obtain the annotated mammogram image using the trained model and conduct data interaction with the website service.In the current studies on mammography image processing,there is relatively little preprocessing work related to sample expansion of small data sets.Therefore,in this paper,the data of the MIAS mammography image public data set is firstly expanded by fliping image,adjusting image contrast,adjusting image brightness and other methods.After that,the data set is randomly divided into the training set and test set in a ratio of 9 to 1.Then,the Faster R-CNN algorithm based on convolutional neural network and the training set are used to train the target detection model of breast lesions.Finally,the test set is used to verify the model.The results of verification show that the average accuracy of the model is 0.9098,which can be integrated with the website service.Website services are built based on the relevant technologies of SpringBoot framework system.According to the standard of software engineering,a powerful and easy-to-use assistant diagnosis system is finally built through system requirement analysis,design process,implementation process and the test process.With the advantages of facilitating patients making an appointment,assisting doctors to image diagnosis and giving treatment,the system will have a certain practical significance and application prospect.