11 results about "Retinal vessel" patented technology

The invention relates to the field of fundus photograph enhancement and segmentation, and provides a fundus photograph enhancement segmentation method and device based on Diffusion, and the method comprises the steps: obtaining a fundus photograph, inputting the fundus photograph into a pre-trained Diffusion-based combined image enhancement segmentation model, and obtaining a fundus photograph enhancement image and a fundus photograph retinal blood vessel segmentation image; the model comprises a Diffusion-based image enhancement branch, an image segmentation branch and a symbiotic information interaction module, and the symbiotic information interaction module is used for realizing information flow between the image enhancement branch and the image segmentation branch, so that the image enhancement branch generates a fundus color photograph enhancement image based on a fundus color photograph and information from the image segmentation branch. Meanwhile, the image segmentation branch generates a fundus color photoretinal blood vessel segmentation image based on the fundus color photograph and information from the image enhancement branch. According to the invention, the understanding and analysis of the model on the image can be improved, and accurate blood vessel segmentation is provided while the image quality is improved.

Disclosed are methods and devices useful for imaging blood-containing tissue, for example for angiography, especially retinal angiography, whereby an image is generated by dividing pixels of an image acquired at some wavelength range by corresponding pixels of an image acquired at a different wavelength range. In some embodiments, a first wavelength range includes predominantly light having wavelengths between about 400 nm and about 620 nm and a second wavelength range includes predominantly light having wavelengths between about 620 nm and about 800 nm.

The present application belongs to the field of medical image segmentation of computer vision, and particularly relates to a retinal blood vessel segmentation method based on UNet and Transformer fusion, which has the following characteristics: step 1, pre-processing the image to be trained to obtain a pre-processed image; step 2, inputting the pre-processed image into a retinal blood vessel segmentation model based on UNet and Transformer fusion to obtain a weight file, the model comprising an encoder, a decoder and a fusion attention mechanism, the encoder comprising a multi-flow cascaded convolution layer, a plurality of pooling layers and a plurality of residual modules, each convolution layer using a residual module, and the pooling layer being arranged between two convolution layer units; the decoder comprising a plurality of improved residual modules based on extended convolution, a plurality of up-sampling modules and a deconvolution layer, the up-sampling module being arranged between two adjacent improved residual modules based on extended convolution; the fusion attention mechanism taking the output of the pooling layer in the encoder and the output of the adjacent flow pooling layer as low-level feature input and high-level feature input, respectively, and the high-level feature input of the fusion attention mechanism at the third layer being a feature map formed by the Transformer module corresponding to the encoder pooling layer; step 3, loading the weight file and inputting a test fundus image into the model to obtain a retinal blood vessel segmentation result. In addition, the retinal blood vessel segmentation model of the present application is more sensitive to small blood vessels, and the segmentation accuracy is improved.

The present invention provides a method, device, and electronic device for calculating retinal vessel branching angles. The method comprises: obtaining a retinal fundus image and binarizing the image to obtain a black-and-white image; determining the optic disc center point based on the black-and-white image; and determining two target retinal branching vessels based on the black-and-white image; and determining the retinal vessel branching angles based on the optic disc center point and the two target branching vessels. The method, device, and electronic device for calculating retinal vessel branching angles provided by the present invention can accurately calculate the retinal vessel bifurcation angles of human eyes.

The invention discloses a lightweight OCTA retinal vessel segmentation method based on a full-convolution full-resolution network. According to the method, an encoder and two decoders are adopted; the encoder is combined with the thought of cyclic convolution and a ConvNeXt convolution module of a full-resolution design to extract a full-resolution feature map under the condition of not carrying out down-sampling so as to reserve information of small blood vessels to the maximum extent; dynamic snakelike convolution is introduced into the main decoder for self-adaptive fitting of the slender tubular shape of the blood vessel; the secondary decoder is mainly composed of a depth separable convolution and a point convolution so as to provide an additional supervision signal for the encoder in a training stage; according to the method, good balance between the segmentation performance and the calculation efficiency can be realized, the segmentation performance equivalent to the segmentation performance in the prior art can be realized only by using lightweight calculation cost, and the method has application potential in a computing resource limited scene.

The invention discloses a retinal vessel segmentation method, system and device based on improved U-net, and a medium, and relates to the technical field of medical image segmentation. The method comprises the following steps: acquiring a retina image to be measured; constructing a retinal vessel segmentation model; the retinal blood vessel segmentation model comprises an encoder structure and a decoder structure which are connected in sequence; the encoder structure comprises a deformable convolution module and a fractal pooling operation module. The decoder structure comprises a dual-path efficient local attention module and an up-sampling process module; and inputting the to-be-detected retina image into the retina vessel segmentation model for processing to obtain a fundus retina vessel data segmentation result. According to the invention, sufficient texture and edge detail features of the blood vessel can be extracted from the fundus image with a complex blood vessel structure, so that the blood vessel can be segmented more accurately.

The application discloses an OCTA image retinal blood vessel segmentation method based on deep learning, and comprises the following steps: building a double-branch axial compression convolutional neural network segmentation model; training and optimizing parameters of the built model; using the trained model, quickly positioning and accurately segmenting the retinal blood vessel structure based on OCTA to obtain a retinal blood vessel segmentation result binary graph. The OCTA image retinal blood vessel segmentation method based on deep learning has the characteristics of high accuracy and fast processing speed.

This invention relates to a method for calculating retinal vessel diameters in fundus images based on region segmentation. The method includes: acquiring a fundus image to be processed; locating the center point of the fundus image to obtain a fundus image with a center point marker; performing region segmentation on the marked fundus image based on an automatic segmentation model to obtain segmented vessel regions; extracting the initial centerline of each vessel region to obtain a fundus image with vessel centerlines; performing secondary segmentation on the fundus image with vessel centerlines to obtain a fundus image with a tree structure and single-branch vessels; and then performing interpolation and smoothing operations pixel by pixel to obtain the average, maximum, and minimum vessel diameters of each vessel region. This invention aims to divide the vessels into different regions based on an automatic segmentation model and simultaneously calculate various vessel parameters within each region, exploring its segmentation capabilities and potential application value.

This invention discloses a method, apparatus, storage medium, and device for retinal vessel image segmentation, belonging to the field of image segmentation technology. The method includes acquiring retinal vessel images and inputting them into a trained retinal vessel segmentation model. Accurate segmentation of retinal vessels is achieved through newly introduced topological feature optimization and local and global feature optimization. This invention improves the modeling effect of vessel topology by introducing topological feature optimization into the encoder, and achieves accurate extraction of vessel details from different scales to the global level by introducing local and global feature optimization into the encoder, reducing the risk of vessel discontinuities and missed detections in the segmentation results.

The invention discloses a multi-resolution retinal blood vessel image segmentation method based on a dynamic strategy. A used model comprises an encoder and a decoder, the encoder comprises a plurality of encoding layers, the decoder comprises a plurality of decoding layers, and an edge perception multi-scale enhancer and a dynamic multi-scale feature fusion module are embedded in jump connection between each encoding layer and the corresponding decoding layer; in the edge perception multi-scale enhancer, the input features are subjected to 3 * 3 convolution to extract local features; the local features are subjected to 1 * 1 convolution to extract initial features, the initial features are sequentially subjected to three times of convolution and average pooling to obtain multi-scale features, the multi-scale features are subjected to edge feature information enhancement through an edge enhancement module, and after the multi-scale features subjected to edge enhancement and the initial features are spliced, convolution is carried out to obtain multi-scale feature information; and obtaining the output characteristics of the edge perception multi-scale enhancer. The blood vessel feature information with large morphological difference is extracted and fused, so that the model learns more effective blood vessel feature information, and the segmentation precision is improved.

This item is an image that displays information using a GUI. This GUI is an image that has the function of displaying the results of eye examinations in ophthalmology. Specifically, the image is divided into upper and lower sections, and the upper and lower sections display the results of the examination, such as a fundus photograph, a circular map showing the thickness of the retina, a cross-sectional image of the retina, and the reticular structure of the retina's blood vessels, respectively, and the image has the function of displaying the examination results in a way that allows comparison between the upper and lower sections. The roughly circular area on the left side of the image displays information related to the fundus photograph. The roughly square area on the left side of the center of the image displays information related to the circular map showing the thickness of the retina. Furthermore, the horizontal rectangle on the right side of the center of the image displays information related to the cross-sectional image of the retina. The roughly square area on the right side of the image displays information related to the reticular structure of the retina's blood vessels. When in use, it is displayed as shown in "Reference Figure 1 showing the usage state," and when it is a monochrome image in use, it becomes "Reference Figure 2 showing the usage state."