Research On 3D High Resolution Flow Detection Techniques For Biological Tissues

Optical coherence tomography(OCT)is an optical imaging technology which is noninvasive and has high resolution,and can obtain biological sample's cross section profile.Swept source OCT(SSOCT)has higher imaging speed,higher SNR and sensitivity compare to other OCT form,and thus is the form of FD-OCT predominantly used in experimental research and biomedical image.Flow information in biological sample can be extract by combine OCT with flow detect technology which contain Doppler technology.Many diseases related to blood flow and vascular morphology.Lesion,malformation and burn in skin may lead to capillary morphology change.In respect of ophthalmology,there are more vasculature disease because blood circulation system is sparse while retina has largest oxygen comsumption per weight in human tissues.Hence,high-resolution 3D angiography is the key to prevent,diagnose and treat the vasculature related disease.In this article,we focus on the problems that may affect flow detection and the improvement of flow filed image in a SS-OCT.The main contend is showed as follow:1)Measured and analyzed the instability of two polygon tunable laser in a 1310 nm SSOCT: Output power fluctuation and spectrum misalign in laser led to intensity and phase instability in interference signal,and thus affect the OCT intensity image and flow image.Sweep range fluctuation reduces the interference signal utilization,and degrade the image resolution and SNR.2)Proposed a Mach-Zehnder interferometer-based phase and amplitude correction method that could stabilize phase and eliminate vertical fringe effectly without signal truncation,and improved flow detecting effect,imaging resolution and SNR.The method had higher imaging speed comare to other correction method because it used only one MZI signal's unwrapped phase information during resampling process.3)Implemented 3D angiography for a vascular net model to show the efficacy of proposed correction method.We also extract the flow information from vasculature in a rabbit ear and a human skin in vivo,and showed a good flow imaging result.During the experiment,the correction method was wrote and ran in a GPU,which can achive real-time flow imaging.