# US 9,775,506 B2 · System and method for detecting and classifying severity of retinal disease > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 9,775,506 B2 - Document ID: US9775506B2 - Status: Granted - Kind: B2 - Application number: 14/437,777 - Filed: 2013-07-10 - Published: 2015-09-24 - Granted: 2017-10-03 - Earliest claimed: 2012-11-08 - Assignees: The Johns Hopkins University - Applicants: The Johns Hopkins University - Inventors: Philippe M. Burlina; David E. Freund; Srihari Kankanahalli; Neil M. Bressler - Web page: https://patentgrep.com/patent/US9775506B2 ## CPC - A61B 3/0025: {characterised by electronic signal processing, e.g. eye models} - A61B 3/10: Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions - A61B 3/102: {for optical coherence tomography [OCT]} - A61B 3/12: for looking at the eye fundus, e.g. ophthalmoscopes (A61B3/13 takes precedence) - A61B 3/14: Arrangements specially adapted for eye photography - A61B 5/7246: {using correlation, e.g. template matching or determination of similarity} - A61B 5/7264: {Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems} - A61B 5/7275: {Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor} - G06T 7/0012: {Biomedical image inspection} - G06T 2207/20076: Probabilistic image processing - G06T 2207/30041: Eye; Retina; Ophthalmic - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract A method of detecting, and classifying severity of, a retinal disease using retinal images includes at least one of receiving, retrieving or generating reference data that includes information concerning occurrences of key image features for each of a plurality of retinal disease and disease severity conditions; receiving a retinal image of an individual; processing the retinal image of the individual to identify occurrences of each of a plurality of distinguishable image features throughout at least a region of interest of the retinal image; identifying which ones of the identified occurrences of the plurality of distinguishable image features of the retinal image of the individual correspond to the key image features of the reference data; calculating, based on the identifying, a number of occurrences of each of the key image features in the retinal image of the individual; and determining at least one of a likelihood of a presence of a retinal disease or a likelihood of developing a retinal disease based on a comparison of the number of occurrences of each of the key image features in the retinal image of the individual to the reference data. ## Claims ### Claim 1 A method of detecting, and classifying severity of, a retinal disease using retinal images, comprising: at least one of receiving, retrieving or generating reference data comprising information concerning occurrences of key image features for each of a plurality of retinal disease and disease severity conditions; receiving a retinal image of an individual; processing said retinal image of said individual to identify occurrences of each of a plurality of distinguishable image features throughout at least a region of interest of said retinal image; identifying which ones of said identified occurrences of said plurality of distinguishable image features of said retinal image of said individual correspond to said key image features of said reference data; calculating, based on said identifying, a number of occurrences of each of said key image features in said retinal image of said individual; and determining at least one of a probability of a presence of a retinal disease or a probability of developing a retinal disease based on a comparison of said number of occurrences of each of said key image features in said retinal image of said individual to said reference data, wherein said processing said retinal image of said individual to identify occurrences of each of said plurality of distinguishable image features comprises applying a plurality of image operators to said retinal image. ### Claim 2 (depends on 1) The method of claim 1, further comprising determining a probability of a severity of said retinal disease based on said comparison of said number of occurrences of each of said key image features in said retinal image of said individual to said reference data. ### Claim 3 (depends on 1) The method of claim 1, wherein said plurality of image operators includes at least one of a Laplacian of Gaussian (LoG), Difference of Gaussian (DoG), Determinant of Hessian (DoH), SURF (Speeded Up Robust Features), SIFT (Scale Invariant Feature Transform), Morphological-based feature detector, Multi-scale feature detector, or Corner detector. ### Claim 4 (depends on 1) The method of claim 1, wherein said retinal image of said individual is an optical image from a fundus camera. ### Claim 5 (depends on 1) The method of claim 1, wherein said retinal image of said individual is an optical coherence tomography image. ### Claim 6 (depends on 1) The method of claim 1, further comprising receiving non-image information regarding said individual, and wherein said determining at least one of said probability of said presence of said retinal disease or said probability of developing said retinal disease is further based on said non-image information regarding said individual. ### Claim 7 (depends on 1) The method of claim 1, wherein said generating said reference data comprises generating said reference data based on processing a plurality of expert-classified retinal images that have been classified according to at least one of risk of developing a retinal disease, retinal disease type and retinal disease severity by an expert. ### Claim 8 (depends on 7) The method of claim 7, wherein said generating said reference data comprises: processing said plurality of expert-classified retinal images to identify occurrences of each of a plurality of distinguishable reference image features, identifying key image features corresponding to said plurality of expert-classified retinal images, and computing a frequency of occurrence of each of said key image features corresponding to each classification of said expert-classified retinal images. ### Claim 9 (depends on 8) The method of claim 8, wherein said identifying key image features uses a statistical clustering process. ### Claim 10 (depends on 9) The method of claim 9, wherein said identifying key image features uses at least one of a K-means, Mean Shift, Expectation maximization or Support Vector Data Description process. ### Claim 11 A computer-readable medium for detecting, and classifying severity of, a retinal disease using retinal images, said computer-readable medium comprising non-transitory computer-executable code which, when executed by a computer, causes the computer to: at least one of receive, retrieve or generate reference data comprising information concerning occurrences of key image features for each of a plurality of retinal disease and disease severity conditions; receive a retinal image of an individual; process said retinal image of said individual to identify occurrences of each of a plurality of distinguishable image features throughout at least a region of interest of said retinal image; identify which ones of said identified occurrences of said plurality of distinguishable image features of said retinal image of said individual correspond to said key image features of said reference data; calculate, based on said identifying, a number of occurrences of each of said key image features in said retinal image of said individual; and determine at least one of a probability of a presence of a retinal disease or a probability of developing a retinal disease based on a comparison of said number of occurrences of each of said key image features in said retinal image of said individual to said reference data, wherein said processing said retinal image of said individual to identify occurrences of each of a plurality of distinguishable image features comprises applying a plurality of image operators to said retinal image. ### Claim 12 (depends on 11) The computer-readable medium of claim 11, wherein said non-transitory computer-executable code which, when executed by a computer, further causes the computer to determine a probability of a severity of said retinal disease based on said comparison of said number of occurrences of each of said key image features in said retinal image of said individual to said reference data. ### Claim 13 (depends on 11) The computer-readable medium of claim 11, wherein said plurality of image operators includes at least one of a Laplacian of Gaussian (LoG), Difference of Gaussian (DoG), Determinant of Hessian (DoH), SURF (Speeded Up Robust Features), SIFT (Scale Invariant Feature Transform), Morphological-based feature detector, Multi-scale feature detector, or Corner detector. ### Claim 14 (depends on 11) The computer-readable medium of claim 11, wherein said retinal image of said individual is an optical image from a fundus camera. ### Claim 15 (depends on 11) The computer-readable medium of claim 11, wherein said retinal image of said individual is an optical coherence tomography image. ### Claim 16 (depends on 11) The computer-readable medium of claim 11, wherein said non-transitory computer-executable code which, when executed by a computer, further causes the computer to receive non-image information regarding said individual, and wherein said determining at least one of a probability of a presence of a retinal disease or a probability of developing a retinal disease is further based on said non-image information regarding said individual. ### Claim 17 (depends on 11) The computer-readable medium of claim 11, wherein said non-transitory computer-executable code which, when executed by a computer, further causes the computer to generate said reference data based on processing a plurality of expert-classified retinal images that have been classified according to at least one of risk of developing a retinal disease, retinal disease type and retinal disease severity by an expert. ### Claim 18 (depends on 17) The computer-readable medium of claim 17, wherein said generating said reference data comprises: processing said plurality of expert-classified retinal images to identify occurrences of each of a plurality of distinguishable reference image features, identifying key image features corresponding to said plurality of expert-classified retinal images, and computing a frequency of occurrence of each of said key image features corresponding to each classification of said expert-classified retinal images. ### Claim 19 (depends on 18) The computer-readable medium of claim 18, wherein said identifying key image features uses a statistical clustering process. ### Claim 20 (depends on 19) The computer-readable medium of claim 19, wherein said identifying key image features uses at least one of a K-means, Mean Shift, Expectation maximization or Support Vector Data Description process. ### Claim 21 A system for detecting, and classifying severity of, a retinal disease, comprising: a retinal scanner constructed to obtain retinal images of an individual; and a data processing system in communication with the retinal scanner, the data processing system configured to: at least one of receive, retrieve or generate reference data comprising information concerning occurrences of key image features for each of a plurality of retinal disease and disease severity conditions; receive a retinal image of said individual from said retinal scanner; process said retinal image of said individual to identify occurrences of each of a plurality of distinguishable image features throughout at least a region of interest of said retinal image; identify which ones of said identified occurrences of said plurality of distinguishable image features of said retinal image of said individual correspond to said key image features of said reference data; calculate, based on said identifying, a number of occurrences of each of said key image features in said retinal image of said individual; and determine at least one of a probability of a presence of a retinal disease or a probability of developing a retinal disease based on a comparison of said number of occurrences of each of said key image features in said retinal image of said individual to said reference data, wherein said processing said retinal image of said individual to identify occurrences of each of a plurality of distinguishable image features comprises applying a plurality of image operators to said retinal image. ### Claim 22 (depends on 21) The system of claim 21, wherein said data processing system is further configured to determine a probability of a severity of said retinal disease based on said comparison of said number of occurrences of each of said key image features in said retinal image of said individual to said reference data. ### Claim 23 (depends on 21) The system of claim 21, wherein said plurality of image operators includes at least one of a Laplacian of Gaussian (LoG), Difference of Gaussian (DoG), Determinant of Hessian (DoH), SURF (Speeded Up Robust Features), SIFT (Scale Invariant Feature Transform), Morphological-based feature detector, Multi-scale feature detector, or Corner detector. ### Claim 24 (depends on 21) The system of claim 21, wherein said retinal scanner is a fundus camera and said retinal image of said individual is an optical image from said fundus camera. ### Claim 25 (depends on 21) The system of claim 21, wherein said retinal scanner is an optical coherence tomography scanner and said retinal image of said individual is an optical coherence tomography image. ### Claim 26 (depends on 21) The system of claim 21, wherein said data processing system is further configured to receive non-image information regarding said individual, and wherein said determining at least one of a likelihood of a presence of a retinal disease or a likelihood of developing a retinal disease is further based on said non-image information regarding said individual. ### Claim 27 (depends on 21) The system of claim 21, wherein said data processing system is further configured to generate said reference data based on processing a plurality of expert-classified retinal images that have been classified according to at least one of risk of developing a retinal disease, retinal disease type and retinal disease severity by an expert. ### Claim 28 (depends on 27) The system of claim 27, wherein said generating said reference data comprises: processing said plurality of expert-classified retinal images to identify occurrences of each of a plurality of distinguishable reference image features, identifying key image features corresponding to said plurality of expert-classified retinal images, and computing a frequency of occurrence of each of said key image features corresponding to each classification of said expert-classified retinal images. ### Claim 29 (depends on 28) The system of claim 28, wherein said identifying key image features uses a statistical clustering process. ### Claim 30 (depends on 29) The system of claim 29, wherein said identifying key image features uses at least one of a K-means, Mean Shift, Expectation maximization or Support Vector Data Description process. ### Claim 31 A method of detecting, and classifying severity of, a disease using physiological information, comprising: at least one of receiving, retrieving or generating reference data comprising information concerning occurrences of key features for each of a plurality of disease and disease severity conditions; receiving physiological information of an individual; processing said physiological information of said individual to identify occurrences of each of a plurality of distinguishable features throughout at least a portion of said physiological information; identifying which ones of said identified occurrences of said plurality of distinguishable features of said physiological information of said individual correspond to said key features of said reference data; calculating, based on said identifying, a number of occurrences of each of said key features in said physiological information of said individual; and determining at least one of a probability of a presence of a disease or a probability of developing a disease based on a comparison of said number of occurrences of each of said key features in said physiological information of said individual to said reference data, wherein said processing said retinal image of said individual to identify occurrences of each of said plurality of distinguishable image features comprises applying a plurality of image operators to said retinal image.