| Gastrointestinal cancers such as stomach and esophageal cancers are a serious threat to the health of our citizens.Although most patients with early-stage GI cancers have a survival rate of more than 90% after a five-year surgery,the early diagnosis and treatment of GI cancers in China is relatively lagging behind.At present,most of the patients with GI cancers in China are in the middle and late stages.The early and ealier stage patients account for only about one-tenth,which is one of the root causes of the high incidence and low survival rate of GI cancers in China.Therefore,the early diagnosis and treatment technology for digestive tract cancer is a major need to protect people’s life and health.Gastrointestinal endoscopy can penetrate deep into the human body to obtain real-time image data without dissection,and can also be used to take biopsies of key areas and excise early lesions with appropriate medical devices.Gastrointestinal endoscopy is the most commonly used gastrointestinal detection method,of which endoscopic fluorescence imaging has become the main way to improve the detection and cure rate of early GI cancer and has the greatest potential for development.However,the existing endoscopic fluorescence imaging technologies are difficult to meet the requirements of clinical use.This is highlighted in four aspects: the imaging quality is limited by imaging mechanism and method in principle,the imaging system functions and indicators are difficult to meet clinical needs,the image processing algorithm is not targeted enough,and the effect of fluorescence sensitizer needs to be improved.These key technologies are in urgent need of breakthrough,which are of great significance to improve the level of early diagnosis and treatment technology of gastrointestinal cancers,increase the detection and cure rates of early gastrointestinal cancers,and protect people’s health.In this thesis,we investigate the problems encountered in the early detection of tumors using endofluorescence imaging based on four aspects: imaging mechanism and method,system optimization,image quality improvement,and nano-probe,using the imaging system as the starting point.The relevant research works conducted in this thesis are summarized as follows:1.The Cerenkov luminiscence imaging system can achieve tumor detection,but the location information can only be roughly estimated.The endoscopic Cerenkov luminiscence imaging system can achieve the localization of tumor boundary information in small field of view,but cannot perform rapid determination of tumor area in large field of view.In this thesis,a dual-field Cerenkov luminiscence imaging system was designed and developed to address the problem that the existing systems cannot simultaneously observe large and small fields of view for pathological regions.The proposed system can achieve dual-field in vivo imaging of small animals by switching between wide-angle field of view and macroscopic field of view,while having the advantages of rapid macroscopic localization and microscopic high-resolution detection at the same time.The performance of the proposed system was demonstrated by a series of experimental tests.The results of in vitro and in vivo experiments showed that the system is capable of both rapid localization of tumor regions and in vivo tumor boundary detection,providing a convenient and effective imaging tool for early tumor detection and basic research.2.The appearance of early tumors does not differ significantly from that of normal human tissues,and the lesion areas are mostly focally distributed.This makes it difficult to distinguish the appearance and affects the accuracy of biopsy sampling,and to a certain extent,affects the detection results,thus causing problems such as tumor margins are not easily distinguished and early tumors are not easily detected.The existing confocal microendoscope excitation light is a single wavelength,which limits the selection of fluorescent probes for application and makes it difficult to combine many fluorescent probes with microscopic imaging to achieve immediate immunohistochemical imaging.To solve these problems,a multispectral fluorescence microendoscopic imaging system was constructed.The system can not only enhance the accuracy of early tumor detection and improve the efficiency of early detection,but also achieve the ability to target tumor tissues at the molecular level.In addition,it can meet the imaging requirements of fluorescent probes under different excitation light bands.Corresponding excitation fluorescence can be generated for different fluorescent probes,thereby detecting the optical signals of different bands of fluorescent probes,and realizing the microscopic imaging of multi-spectral and multi-target fluorescent molecules of the same tissue by the same device with different probes.It has the advantages of multispectral imaging and improved image quality,and then the validation results through experiments show that the imaging results are in good agreement with the pathological results,which are initially qualified for clinical application and more suitable for joint molecular imaging techniques for clinical application research.3.Due to the noise caused by the irregular honeycomb structure of the endoscopic fiber bundle itself,it is often referred to as grid-like spatial noise.This spatial noise is unavoidable because it comes from the physical structure.Therefore,it can cause problems for researchers and healthcare workers during use,such as difficulty in diagnosis,and it is also a great obstacle to the subsequent analysis of images.To address these problems,this chapter first preprocesses the endoscopic images taken with high-resolution fiber optic bundles using Gaussian smoothing filtering algorithm,and then removes the grid-like spatial noise in the images using watershed algorithm.On this basis,the images are quantitatively analyzed to distinguish small intestinal cell carcinoma tissue from normal small intestinal tissue.As a result,the edge of the lesion is accurately determined,and the reduction of the medical trauma suffered by tumor patients in the process of treatment is aimed at.Furthermore,the extent of diseased tissues removed by tumor patients during surgery can be minimized.4.Zinc oxide quantum dots are an excellent nanoprobe with not only fluorescence signal enhancement but also some tumor cell inhibition.However,they have their own problems such as unclear effect of biological body application.To address these problems,different polymeric ligands were used in this thesis to coat Zn O quantum dots.In addition,their preparation and characterization were investigated,and their photoluminescence and tumor photodynamic therapeutic properties were explored.The in vitro and in vivo experiments showed that they meet the needs of quantum level nanodrug therapy integration.This not only provides an excellent image-enhanced probe for early tumor detection,but also can have a certain degree of tumor inhibition.The dual function of tumor detection and therapy at the same time is one of the future trends in the development of therapeutic integration. |
PDF Full Text Request