Study On Classification And Management Methods Of Lung Tissue Images From Patients With Cystic Fibrosis

Cystic fibrosis(CF)is a disease with high risk of death and high mortality.The disease can cause repeated infections in the patient's respiratory system and even lead to death.Quantitative analysis of the patient's lung tissue Computed Tomography(CT)images helps doctors to accurately understand the patient's condition and develop a personalized treatment plan.Quantitative analysis of CT images of lung tissue in patients with CF requires professional imaging knowledge and clinical experience.Currently,this work is mainly done manually by a professional imaging physician.With the increase in the number of CT images,manual processing has problems such as large workload,easy missing labels and mislabeling.Therefore,it is necessary to design an automatic classification method to automatically classify lung CT images of CF patients.At the same time,in order to meet the needs of clinical applications,it is necessary to develop an application labeling classification result and display of the original image.Aiming at the above problems,focusing on the classification and management of lung tissue images in CF patients,this paper studied the automatic classification method of lung tissue in CF patients,the organization and management methods of annotation,and the display of lung CT images and annotation results for portable intelligent terminals.method.Firstly,in the automatic classification of lung tissue of CF patients,this paper introduces relevant feedback on the basis of supervised kernel hash classification method for the problems of multiple sample categories,the difference of visual features between categories and the high initial dimension of samples.thought.This idea strengthens the discriminant features between sample categories by processing positive and negative feedback samples.Experiments show that the proposed method can effectively improve the classification accuracy of lung tissue images in CF patients.Secondly,in the organization and management of the annotation results,the paper organizes the image information and classification information according to the category of the lesion.The XML standard is used to organize and manage the annotation results,providing technical interfaces for the operations of querying,inserting,deleting,and visualizing annotations.Finally,in the display of lung CT images and annotation information for portable intelligent terminals,in view of the increasing popularity of current Android mobile devices,a DICOM image analysis and display program was developed using Android Studio.The prototype system realizes the analysis and display of the DICOM format image,and achieves the purpose of better serving clinical medical treatment.This paper achieves the goal of automatic classification,labeling management and image display of CT images of lung tissue in CF patients.Next,we will explore further to improve the classification accuracy,and expand the function of the program in combination with clinical needs.