Research On Image Algorithms Of Laser Fluorescence Imaging For Parathyroid Detection

In recent years,it has become a research hotspot to explore the application of near-infrared fluorescence with wavelength range of 700-1300 nm in the field of medical surgery.Near-infrared fluorescence imaging has the advantages of non-invasive and real-time.It is widely used in tumor localization,cancer diagnosis and so on.In near-infrared fluorescence imaging,the energy distribution of the output spot of the light source must be uniform,so as to judge the accurate area of fluorescent substances,so it is necessary to evaluate the excitation light source.In addition,near-infrared fluorescence imaging also has some problems,such as gray image background and fuzzy biological tissue boundary.It is necessary to optimize the fluorescence image and improve the image quality,so as to provide more accurate assistance for doctors.This thesis has carried out relevant research on the above problems.The main research contents and results are as follows:(1)An evaluation method of excitation light source based on spot image is studied.Based on spot mode analysis theory,the output spots of semiconductor laser and fiber laser are compared.Combined with the influence of spot energy distribution uniformity on imaging quality,the feasibility of using LD as the light source of subsequent fluorescence experiments is analyzed.(2)The algorithms for fluorescence image optimization processing are studied,including bilateral filtering,mean filtering,median filtering and Gaussian filtering,as well as brightness,contrast adjustment algorithm and pseudo color algorithm.The basic principles and logical architecture of various algorithms are described,and four filtering algorithms are verified,analyzed and compared through experiments.The results show that the bilateral filtering algorithm has better effect in protecting image edge information.(3)A fluorescent region labeling algorithm based on bilateral filtering algorithm is designed,which can mark the specific fluorescent region in the image with complete contour color.A new near infrared fluorescence imaging system is built to verify the algorithm.The test results show that the algorithm can successfully obtain the fluorescent image with clear overall background and color labeled specific fluorescent areas,and improve the accuracy of doctors’ judgment of parathyroid tissue by 10%.