Blood vessel detection and registration method, device and application of three-dimensional retinal oct image

Abstract

The invention discloses a blood vessel detection and registration method, device and application of a three-dimensional retinal OCT image. The method includes an image preprocessing step, and generates a vertical projection of the blood vessel on the corresponding position of the image according to the position of the blood vessel in the retina Figure, and image denoising and enhancement; blood vessel skeleton extraction step, extracting the skeleton structure of blood vessels based on the vertical projection image after denoising and enhancement; feature extraction step, find out the matching feature points of the target image and the image to be registered; In the registration step, a rigid registration model between the target image and the image to be registered is generated based on the extracted feature points. This method can well eliminate the influence of speckle noise, realize automatic detection and accurate registration of blood vessels in retinal SD‑OCT images, and play an important role in the evaluation of the course of common ophthalmic diseases and the determination of doctors' preoperative and postoperative treatment plans Supporting role.

Description

technical field [0001] The invention belongs to the field of medical image processing, and in particular relates to a blood vessel detection and registration method, device and application of a three-dimensional retinal OCT image. Background technique [0002] The retina is the photosensitive tissue located at the back of the eyeball, and it is an important part of the human visual system. At present, SD-OCT technology has become a powerful tool for non-destructive evaluation of retinal diseases, which can provide fast, high-resolution, three-dimensional images showing the inner layer of the retina. The registration of the longitudinal retinal data of the same patient can effectively evaluate the patient's disease development, and provide effective help for doctors to determine preoperative and postoperative treatment plans. Due to the strong speckle noise in SD-OCT retinal images and its unique imaging method, the existing mature methods for longitudinal data registration ...

AI Technical Summary

Problems solved by technology

Due to the strong speckle noise in SD-OCT retinal images, as well as its unique imaging method and other factors, the existing mature methods for longitudinal data registration of MRI, CT and other images can be directly used in SD-OCT retinal images It is difficult to achieve ideal results for the registration of

The main technical defects are as follows: (1) the grayscale-based registration method cannot be accurately registered due to the influence of speckle noise; (2) the discontinuity of the B-scan direction data caused by eye movement has caused great difficulties to the existing registration technology (3) More importantly, SD-OCT data has a strong correlation in the A-scan direction, but the current registration methods do not consider this imaging characteristic, resulting in excessive registration freedom

So far, there is no related report on a complete method for vessel extraction and registration in retinal SD-OCT images

Images