Research On High Resolution Photoacoustic Endoscopic And Vascular Quantification Method For Gastrointestinal Imaging

In recent years,gastrointestinal tumors have long occupied the top five incide nces of malignant tumors in China.The structure and function of tumor vessels are closely related to tumor growth,which is significant for the early detection of tumors.At present,the optical penetration is limited by optical diffraction limit,the penetration depth is low,and the ultrasound endoscope has low contrast,which makes it difficult to obtain high-sensitive vessel images.The emerging photoacoustic endoscopic imaging both has high resolution and large penetration depth,showing obvious potential in monitoring the Gastrointestinal vessels.Therefore,a high-resolution photoacoustic endoscopic imaging system suitable for gastrointestinal tumors,combining with appropriate blood vessel segmentation and vessel structure parameter quantification methods will be helpful for the judgment of tumor growth status,which has important basic medical research and clinical application value.A full 3D in vivo imaging and vessel quantification scheme was proposed in this paper,including a high-resolution photoacoustic endoscopic imaging system,a matching 3D blood vessel segmentation algorithm and vessel structure parameters quantization method.This thesis includes following parts:First,the basic principle of photoacoustic imaging was introduced,and the wave equations of photoinduced ultrasound and photoacoustic signals were analyzed in detail.The Hessian matrix-based blood vessel segmentation algorithm was elaborated,then to solve the abnormality of blood vessels caused by the limited receiving angle of the transducer,the intensity and structure information were fused in blood vessels segmentation.Based on the segmentation,the structure parameters of blood vessels were quantified.Then,a high-resolution photoacoustic endoscopic imaging system was designed and set up.To fit different imaging modes,the system timing and host computer control software were designed.The system realized fully automatic 3D scanning at two imaging rates of 10 fps and 30 fps.The precise optical path module ensured the high resolution of the system and provided powerful system support for the imaging of gastrointestinal vessels.Finally,experiments on phantoms and in vivo were carried out.On the one hand,the phantom was used to test the performance of the imaging system.On the other hand,in vivo experiments were supposed to verify the feasibility of blood vessels segmentation algorithm,which has proved the advantages of the algorithm compared with other algorithms.By comparing the vessel structure parameters obtained by the2 D and 3D quantization methods,the reason of different results has been found.This study has completed the whole work from system application,algorithm development and in vivo experiment verification.High-quality 3D blood vessel images were obtained through high-resolution photoacoustic endoscopic imaging system combined with 3D blood vessel segmentation algorithm;based on the 2D quantization method,the full 3D vessel structure parameter quantization method was proposed.The accuracy of the 3D quantization method was verified by experiments,then the structure parameter of the mouse tumor vessel was quantified.Blood vessel segmentation method and vessel structure parameters quantification method,provided more comprehensive information on the structure of blood vessels.