Quantitative Technique And Application Of Vascular Morphology With OCTA

The appearance and development of various human diseases are closely related to the shape of blood vessels.Optical coherence tomography(OCT)is a non-invasive three-dimensional optical imaging technology.Due to its non-contact,high resolution and high detection sensitivity,OCT is widely used in biomedical and other fields.Based on the three-dimensional spatial resolution of OCT technology,optical coherence tomography angiography(OCTA)technology uses the different characteristics of OCT signals of blood flow and static tissue in time dimension to distinguish them.And obtain three-dimensional optical perfusion imaging of biological tissue.OCTA technology has the advantages of noninvasive and no need of exogenous fluorescent markers,which has important application prospects in the fields of skin diseases and ophthalmic diseases.In this paper,the automatic segmentation of 3D angiographic images and the quantification of vascular morphology are studied based on the three-dimensional flow motion imaging technology of OCTA,and its application value in disease diagnosis is explored.It includes:Based on the linear dependence between the inverse signal to noise ratio(iSNR)and the decorrelation coefficient of the OCTA image,an automatic blood flow segmentation method using binary image structure similarity as feedback is proposed.Because the blood flow segmentation threshold of this method is related to the SNR of blood flow image,the blood flow signal in the low SNR region of blood flow image can be retained to the maximum extent,and the static tissue background can be removed at the same time.Compared with the widely used binarization method of OCTA image,the advantages of this algorithm are verified by skin OCTA image.On the basis of accurate segmentation of blood flow in OCTA images,further quantitative study of vascular morphology was carried out.Using the above blood flow segmentation method to binary 3D OCTA image can provide the ability to distinguish overlapping vessels in the depth direction.On the basis of the binary image,the area density of blood vessels,the average length of blood vessels per unit area,the average perimeter of blood vessels per unit area,the average diameter of blood vessels and the complexity of blood vessels were quantified.The above parameters were used to analyze the vascular morphology of the skin of patients with port wine stains,and the response rules of different morphological parameters to the diseased skin and normal skin of patients with port wine stains were summarized.The above morphological parameters were applied to the morphological analysis of human retinal vessels.The retina of human eye is stratified according to the physiological structure.Using graph search method,the boundary of OCT retina structure map is searched,and the boundary of each layer of retina is obtained.The vascular morphological parameters of superficial and deep layers were calculated respectively.This method provides an effective detection method for diseases reflected by retinal vascular morphological abnormalities.