46 results about "Optic cup/disc ratio" patented technology

The invention discloses a retinal fundus image cup / disc ratio automatic evaluation method. The method comprises the following steps: A, extracting an optic disk area image from a retinal fundus image;B, establishing and training a optic disc optic cup segmentation network based on the deep convolutional neural network; C, acquiring a to-be-detected optic disc area image from the to-be-detected retina fundus image according to the step A, and inputting the to-be-detected optic disc area image into an optic disc optic cup segmentation network to output a to-be-detected optic disc segmentation mask image and a to-be-detected optic cup segmentation mask image; And step D, calculating the cup-disc ratio of the retinal fundus image according to the to-be-detected optic disc segmentation mask image and the to-be-detected optic cup segmentation mask image. The method is high in operation speed, good in effect, free of manual participation, low in cost and high in universality, and can be widely applied to auxiliary screening of glaucoma.

The invention discloses a fundus image optic disc segmentation method and a system for assisting glaucoma screening, and relates to the technical field of image information processing. The fundus image optic disc segmentation method comprises the steps that a plurality of fundus images are collected and preprocessed, and a training image sample set and a verification image sample set are obtained;training of a constructed W-Net-Mcon full convolutional neural network by using the training image sample set to obtain an optimal W-Net-Mcon full convolutional neural networkis carried out; preprocessing the fundus image to be segmented, and inputting the preprocessed fundus image to be segmented into the optimal W-Net-Mcon full convolutional neural network to obtain a prediction target result image; Processing prediction target result graph by utilizing polar coordinate inverse transformation and ellipse fitting to obtain final segmentation result so as to obtain cup-to-disk ratio and finally obtain glaucoma preliminary screening result. According to the method, image semantic information can be effectively extracted in a multi-size mode, fusion of features of different levels, fusion of global features and detail features and encouragement of feature multiplexing are carried out, gradient back propagation is improved, and the image segmentation precision is improved.

A non-stereo fundus image is used to obtain a plurality of glaucoma indicators. Additionally, genome data for the subject is used to obtain genetic marker data relating to one or more genes and / or SNPs associated with glaucoma. The glaucoma indicators indicators and genetic marker data are input into an adaptive model operative to generate an output indicative of a risk of glaucoma in the subject. In combination, the genetic indicators and genome data are more informative about the risk of glaucoma than either of the two in isolation. The adaptive model may be a two-stage model, having a first stage in which individual genetic indicators are combined with respective portions of the genome data by first adaptive model modules to form respective first outputs, and a second stage in which the first outputs are combined by a second adaptive mode. Texture analysis is performed on the fundus images to classify them based on their quality, and only images which are determined to meet a quality criterion are subjected to an analysis to determine if they exhibit glaucoma indicators. Also, the images are put into a standard format. The system may include estimating the position of the optic cup by combining results from multiple optic cup segmentation techniques. The system may include estimating the position of the optic disc by applying edge detection to the funds image, excluding edge points that are unlikely to be optic disc boundary points, and estimating the position of an optic disc by fitting an ellipse to the remaining edge points.

A glaucoma diagnosis apparatus according to an embodiment includes a fundus image processor configured to receive a fundus image and extract a first region of interest (ROI) and a second ROI from the received fundus image, an image classification neural network configured to learn the extracted first ROI and perform classification into a normal fundus image and a glaucoma fundus image on the basis of the learned first ROI, a vertical cup-to-disc ratio (vCDR) calculator configured to recognize an optic disc (OD) and an optic cup (OC) from the extracted second ROI and calculate a vCDR, and a determinator configured to aggregate a vCDR calculation result and an image classification result of the image classification neural network to determine whether glaucoma is present in the fundus image.

The invention relates to the field of histopathology, and particularly discloses a preparation method of an animal eyeball pathological section. The preparation method comprises the steps of: (1) eyeball extraction; (2) eyeball fixation, which is implemented by putting an separated eyeball into a stationary liquid which is pre-cooled to 4.0 DEG C, and putting the stationary liquid containing the eyeball into a refrigerator of which the temperature is set to 4.0 DEG C for fixation for 48 hours, wherein the stationary liquid comprises the following components: 4% paraformaldehyde solution, anhydrous acetic acid and absolute ethyl alcohol in a volume ratio of 10:(1.5-2.0):(8.0-8.5); (3) dehydration and transparent processing of an eyeball optic cup; (4) wax dipping and embedding; (5) and slicing. The preparation method is beneficial to preparation of the high-quality eyeball pathological section so as to facilitate improvement of basic application of retina related research.

The invention discloses an automatic segmentation method for an optic disc optic cup of a color fundus image. The method comprises the following steps: acquiring a known color fundus image and a corresponding fundus image data set; Constructing and obtaining an optic disc segmentation model; segmentating and intercepting The data set, and obtaining a screenshot; Constructing and obtaining an optic disc optic cup segmentation model under the Euclidean coordinates; Performing polar coordinate transformation on the screenshot to obtain a polar coordinate screenshot; Constructing and obtaining anoptic disc optic cup segmentation model under polar coordinates; Segmenting the to-be-analyzed data by using two types of models to obtain an Euclidean coordinate optic disc optic cup segmentation result and a polar coordinate optic disc optic cup segmentation result; And fusing the two types of segmentation results to obtain a final optic disc and optic cup segmentation result of the fundus image. According to the method, automatic optic disk and optic cup segmentation can be more accurately carried out on the color fundus image, and the method is simple, reliable and good in applicability.

A method is presented to obtain, from a retinal image, data characterizing the optic cup, such as data indicating the location and / or size of the optic cup in relation to the optic disc. A disc region of the retinal image of an eye, is expressed as a weighted sum of a plurality of pre-existing “reference” retinal images in a library, with the weights being chosen to minimize a cost function. The data characterizing the cup of the eye is obtained from cup data associated with the pre-existing disc images and the corresponding weights. The cost function includes (i) a construction error term indicating a difference between the disc region of the retinal image and a weighted sum of the reference retinal images, and (ii) a cost term, which may be generated using a weighted sum over the reference retinal images of a difference between the reference retinal images and the disc region of the retinal image.

The invention relates to a color fundus image optic cup segmentation method based on deep learning. The method comprises the following steps: 1) inputting a fundus image; 2) segmenting the optic diskby utilizing a Seg-ResNet network; the segmented optic disk area is used as an interested area of optic cup segmentation; then, a Seg-ResNet network is used for carrying out optic cup segmentation onthe optic disk area; the network is based on a residual basic structure, channel weighting is carried out by considering the relationship between feature channels, modeling is carried out on the dependency relationship between the channels, the feature response value of each channel is adaptively adjusted, feature fusion is carried out on multiple layers, and the position information of pixel points is positioned while image semantic information is captured; and 3) outputting an optic cup segmentation result using the Seg-ResNet network. According to the method, the optic cup segmentation testis carried out on the public data sets GlucomaRepo and Driver-GS, and the result shows that the segmentation accuracy and the algorithm robustness are improved by the test result.

The embodiment of the invention provides a fundus image classification system based on integrated deep learning. The fundus image classification system comprises a pre-diagnosis classification network, a segmentation network and a final diagnosis module. The pre-diagnosis classification network obtains an initial diagnosis result based on the global information of a target fundus image; the segmentation network performs image segmentation on the optic disc, optic cup and optic nerve fiber layer states of the target fundus image based on the initial diagnosis result; and the final diagnosis module extracts a vertical cup-to-disk ratio and an ISNT score based on a result of optic disk, optic cup and optic nerve fiber layer state segmentation, and acquires and displays a final category of thetarget eye fundus image based on the vertical cup-to-disk ratio, the ISNT score and an optic nerve fiber layer defect state. According to the embodiment of the invention, firstly, glaucoma pre-diagnosis is carried out on a target fundus image, and a proper target segmentation network is selected, so that the segmentation precision is improved; and when glaucoma judgment is carried out, the accuracy of a classification result is further improved by combining a plurality of quantitative indexes capable of reflecting the disc edge form.

The invention discloses a joint segmentation method for optic cups and optic disks in fundus medical images, which comprises the following steps: firstly, constructing a joint segmentation network based on a UNet network, and introducing a global information extraction module and a multi-path hole convolution module into the joint segmentation network; and inputting a fundus medical image to be processed into the joint segmentation network for joint segmentation of the optic cups and the optic disks. According to the method, global context information and multi-scale context information in thefundus medical image can be fully extracted, and the joint segmentation effect of optic cups and optic disks in the fundus medical image is improved.

The invention discloses a training device for glaucoma recognition. The training device comprises an acquisition module; the image segmentation network is an artificial neural network based on deep learning and is trained through the preprocessed fundus image, the annotated image and the spatial weighted graph so as to output the probability that each pixel point in the preprocessed fundus image belongs to an optic disk and the probability that each pixel point belongs to an optic cup; generating an optic disc area image and an optic cup area image based on the probability that each pixel point in the preprocessed fundus image belongs to the optic disc and the probability that each pixel point belongs to the optic cup; the feature extraction module is used for acquiring glaucoma features based on the optic disc region image and the optic cup region image; and a classifier trained by feature information including glaucoma features and a glaucoma classification tag based on machine learning to output a probability belonging to glaucoma. According to the scheme, the accuracy of glaucoma recognition can be improved.

The invention belongs to the technical field of artificial intelligence, and particularly relates to a medical fundus image data set, in particular to a optic cup and optic disk segmentation method for fundus image data set transfer learning. According to the method, through adversarial training of a backbone segmentation network and two domain discriminators, universal features between fundus image data sets are extracted, and the features are weighted by using an attention module, so that the problem of fuzzy optic disc boundary of the optic cup is solved, and the interference of other fundus lesions on a segmentation task is eliminated. On the premise that target data set labeling information is not used, the algorithm keeps high optic cup and optic disc segmentation precision in the fundus image data set migration process, and the limitation of insufficient labeled fundus data on the performance of a traditional automatic glaucoma screening method is effectively eliminated.

The invention discloses a multi-mechanism adaptive eye fundus image segmentation method, which belongs to the field of artificial intelligence medical image processing, constructs a multi-mechanism adaptive Faster R-CNN eye fundus image segmentation network, and adopts a network global loss function to guide optic disk and optic cup segmentation. The method specifically comprises the following steps: collecting an eye fundus image data set, and performing preprocessing; the preprocessed data set is input into the constructed multi-mechanism adaptive Faster R-CNN eye fundus image segmentation network; a network global loss function is adopted to guide accurate segmentation of an optic disk and an optic cup; eye fundus images are collected in real time and input into a multi-mechanism adaptive Faster R-CNN eye fundus image segmentation network, an optic disc and an optic cup are accurately segmented by using a global loss function in the network, and clinical diagnosis of doctors is assisted. According to the method, accurate segmentation of the optic disc and the optic cup in the eye fundus image is realized, and the method has relatively good generalization and robustness.

The invention relates to the field of artificial intelligence, specifically uses a neural network, and discloses an optic cup and optic disc segmentation model training method, a cup-to-disc ratio determination method, device and equipment based on the neural network, and a storage medium, and the optic cup and optic disc segmentation model training method comprises the following steps: obtaininga sample image and an image label corresponding to the sample image; constructing sample data; inputting the sample data into a preset neural network to obtain a predicted optic cup and optic disc segmentation image; respectively projecting the image label and the predicted optic cup and optic disc segmentation image to obtain a label projection value corresponding to the image label and an imageprojection value of the predicted optic cup and optic disc segmentation image; respectively calculating the numerical value of the segmentation loss function and the numerical value of the projectionloss function to obtain the numerical value of the network loss function; and training the preset neural network according to the numerical value of the network loss function to obtain an optic cup and optic disc segmentation model. The invention is applicable to the field of smart medical treatment.

According to the fundus photo classification method and the fundus image processing method and system, the technical problems that an existing fundus photo classification method is low in efficiency and large in error can be solved. Comprising the steps of obtaining and preprocessing a fundus photo to generate a standardized fundus picture; dividing the processed fundus photos into a training set and a test set, and performing data amplification on the training set; performing data preprocessing on the training set and the test set, and ensuring that each training data set (batch) input into the model in the training stage is uniform in category; integrating learning strategies, and training a branch neural network model; and carrying out model fusion to obtain a final detection model so as to realize classification of the fundus photos. The device is high in effect and excellent in automation degree; according to the method, features such as the optic cup / optic disk ratio and the arteriovenous ratio of the fundus photo are extracted, classification is carried out in combination with a machine learning method, the calculation speed is high, and few computer resources are occupied during operation.

The invention discloses an optic cup and optic disk segmentation method based on a rich context network. The method comprises the following steps: acquiring and processing existing color eye fundus image data to obtain a training data set; constructing an original optic cup and optic disc segmentation model, and training to obtain an optic cup and optic disc segmentation model; and segmenting the target color eye fundus image by using the optic cup and optic disc segmentation model to obtain a final optic cup and optic disc segmentation result. The invention further discloses an imaging method adopting the optic cup and optic disk segmentation method based on the rich context network. The invention provides a segmentation structure which is based on a convolutional neural network and can obtain sufficient context information to carry out optic disc and optic cup segmentation. Therefore, the method can improve the segmentation performance of the optic disc optic cup, solves the problem that the edge segmentation of the optic cup is not smooth enough, and is high in precision, good in reliability and good in segmentation effect.

The invention relates to the field of medical image processing, and provides a fundus image segmentation method and system of a full convolutional neural network based on an Attention mechanism, and acomputer readable storage medium, the method comprising: selecting fundus retina image data as a training set and a test set; preprocessing the fundus retina images in the training set; constructinga full convolutional neural network model on the TensorFlow; and segmenting the test set by using the trained full convolutional neural network model to obtain a final segmentation result. The systemcomprises a data acquisition module, a preprocessing module, a full convolutional neural network construction module and an image segmentation module. According to the multi-connection complete convolutional neural network model based on the Attention mechanism, the optic disc of the optic cup is automatically segmented from the fundus image, various limitations of a traditional method are overcome, the learning ability of the model is improved by fusing multi-level features in the neural network, and the accuracy of cup disc segmentation is improved.

The invention discloses an eye fundus image optic cup and optic disk segmentation method under a unified framework, and the method comprises the steps: obtaining an eye fundus image before segmentation, and carrying out the image preprocessing operation, such as cutting and rotating; generating a corresponding mask image according to an optic cup and optic disk area marked on the fundus color photo by an ophthalmologist; constructing a deep network for segmenting an optic cup and an optic disk; iteratively training the deep segmentation network by using the mask image and the fundus image, and optimizing network parameters; and segmenting the optic cup and the optic disc, and obtaining the segmentation results of the optic cup and the optic disc by using the trained segmentation network model. The invention provides a deep neural network for optic cup and optic disc segmentation. The deep neural network comprises a multi-scale feature extractor, a multi-scale feature transition and an attention pyramid structure. According to the method, the optic cup and the optic disc can be effectively segmented, the segmentation precision is high, and meanwhile, a new thought is provided for segmentation of fundus images and segmentation of other medical images.

The invention provides an image evaluation method and device, a computer readable storage medium and electronic equipment, and relates to the technical field of image processing. The image evaluation method comprises the following steps: determining optic cup and optic disc feature information corresponding to a fundus image to be evaluated; determining nerve fiber injury information corresponding to the fundus image to be evaluated; and determining glaucoma assessment information corresponding to the fundus image to be assessed based on the optic cup and optic disc feature information and the nerve fiber injury information. Intelligent evaluation is carried out on the early risk of glaucoma, the severity and the change degree of glaucoma and the like according to the optic cup and optic disc feature information in combination with the nerve fiber injury area, so that doctors are helped to carry out glaucoma diagnosis more accurately.

The invention discloses a fundus retina image segmentation method based on a deep convolutional neural network. The method employs the deep convolutional neural network to map features of a vascular tissue, an optic disc optic cup tissue and a lesion tissue in a medical image, and employs the convolutional network to segment the image. In addition, in order to increase segmentation accuracy, a new data preprocessing method of the fundus retina image is used for enhancing the image; an end-to-end deep convolutional network is used for solving a problem of small blood vessel segmentation, and deep significant features of a lesion area are obtained and visualized; a series of problems caused by large pixels of various medical images are solved by using a method of combining multiple deep neural networks.

A two-way cross connection convolutional neural network for image segmentation performs accurate segmentation while executing different interest targets in a multi-modal medical image, and achieves effective extraction of different interest targets by introducing two different network branches and a new cross jump connection in an existing BiO-Net segmentation network. The segmentation experiment based on the disclosed eye fundus image shows that the optic disc and optic cup areas in the eye fundus image can be effectively extracted, and the segmentation performance superior to that of existing networks such as U-Net and BiO-Net is obtained.

The invention relates to the field of artificial intelligence, and discloses a cup-to-disk ratio determination method, device and equipment, and a storage medium. The method comprises the following steps: obtaining and detecting an eye fundus image, and obtaining an optic disk region; inputting the optic disk region into a coding network of an image segmentation model, and extracting image features to obtain a first feature map; inputting the first feature map into a position correction network of the decoding network to obtain a position-corrected second feature map; performing convolution operation on the first feature map through a segmentation network in the decoding network to obtain a third feature map; splicing the second feature map and the third feature map through a connection layer to obtain an image segmentation result; according to an image segmentation result, calculating to obtain an optic cup diameter and an optic disc diameter; and calculating the cup-to-disc ratio of the fundus image according to the optic cup diameter and the optic disc diameter. According to the method, the accuracy of the optic cup and optic disc image obtained through segmentation is improved, multi-screening and screening missing conditions in the disease screening process are reduced, in addition, the invention further relates to a block chain technology, and the eye fundus image can be stored in a block chain.

The invention provides an image processing method and device, a computer readable storage medium and electronic equipment, and relates to the technical field of image processing. The image processing method comprises the following steps: determining optic disc position information corresponding to a fundus image to be processed based on a preset target detection algorithm; determining an optic disc segmentation image and optic disc boundary information corresponding to the fundus image to be processed based on the optic disc position information and a preset edge detection algorithm; and determining an optic cup segmentation image and optic cup boundary information corresponding to the fundus image to be processed based on the optic disc position information. According to the image processing method provided by the embodiment of the invention, the eye fundus image optic disk and optic cup identification and extraction can be realized, and the segmentation precision reaches a sub-pixel level or above.

The invention discloses a glaucoma medical image classification method based on the capsule theory, including image preprocessing operations, constructing a capsule-based convolutional neural network and constructing an image reconstruction fully connected network; the image preprocessing operation first utilizes the existing glaucoma medical The image disc detection model cuts out the region of interest to reduce the interference information, then fuses the optic disc and cup semantic segmentation map information, and finally adopts the adaptive histogram enhancement with limited contrast to improve the overall or local contrast of the image. The capsule-based convolutional neural network can not only automatically learn the features of glaucoma medical images, but also discover the position and direction information between features, so as to identify glaucoma more accurately. Image Reconstruction Fully Connected Network uses category capsules to restore original glaucoma medical images, with the aim of improving the generalization ability of capsule-based convolutional neural networks.