# US 12,224,061 B2 · Determining levels of hypertension from retinal vasculature images > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,224,061 B2 - Document ID: US12224061B2 - Status: Granted - Kind: B2 - Application number: 16/837,733 - Filed: 2020-04-01 - Published: 2020-10-08 - Granted: 2025-02-11 - Earliest claimed: 2019-04-04 - Assignees: Optos PLC - Applicants: Optos PLC - Inventors: Gavin Robertson; Alan Duncan Fleming - Web page: https://patentgrep.com/patent/US12224061B2 ## CPC - G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems - A61B 5/021: Measuring pressure in heart or blood vessels - G06T 5/40: using histogram techniques - G06T 7/0012: {Biomedical image inspection} - G06T 7/11: Region-based segmentation - G06T 7/155: involving morphological operators - G06V 10/764: using classification, e.g. of video objects - G06V 10/82: using neural networks - G06V 40/193: {Preprocessing; Feature extraction} - G16H 30/40: for processing medical images, e.g. editing - G16H 50/30: for calculating health indices; for individual health risk assessment - A61B 3/0025: {characterised by electronic signal processing, e.g. eye models} - A61B 3/12: for looking at the eye fundus, e.g. ophthalmoscopes (A61B3/13 takes precedence) - A61B 2576/02: specially adapted for a particular organ or body part - G06T 2207/20024: Filtering details - G06T 2207/20036: Morphological image processing - G06T 2207/20081: Training; Learning - G06T 2207/20084: Artificial neural networks [ANN] - G06T 2207/30041: Eye; Retina; Ophthalmic - G06T 2207/30101: Blood vessel; Artery; Vein; Vascular - G06V 2201/03: Recognition of patterns in medical or anatomical images - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A computer-implemented method, system, and computer-readable medium, for determining a level of hypertension of a person. Image data defining a retinal image of a retina of the person that has been captured by a retinal imaging system is acquired. Processing the image data is performed to identify a plurality of vessel segments present in the retinal image. Determining is performed of respective widths of vessel segments of at least a subset of the plurality of vessel segments present in the retinal image. Calculating, as a vascular summary metric, at least one average value or median value of the determined widths of the vessel segments present in the retinal image, is performed to determine a level of hypertension by which the person has been affected. ## Claims ### Claim 1 A non-transitory, computer-readable storage medium storing computer program instructions which, when executed by a computer, cause the computer to execute a method for determining a level of hypertension of a person, said method comprising: acquiring image data defining a retinal image of a retina of the person that has been captured by a retinal imaging system; processing the image data based on a classification algorithm, which has been trained on image data defining retinal images of a plurality of subjects, to identify a plurality of vessel segments present in the retinal image, wherein the processing comprises: demarcating a region of interest, which is confined to a portion of the retinal image showing a near periphery of the retina that comprises an annular segment subtending 180 degrees nasal to the optic disc, from a remainder of the retinal image; and processing the image data to identify the plurality of vessel segments in the region of interest of the retinal image; determining respective widths of vessel segments of at least a subset of the plurality of vessel segments present in the region of interest of the retinal image; calculating, as at least one value of a vascular summary metric, at least one of an average value of and a median value of the determined widths of the vessel segments present in the region of interest of the retinal image; and using the at least one value of the vascular summary metric and a relationship between the vascular summary metric and a level of hypertension to determine a level of hypertension by which the person has been affected. ### Claim 2 (depends on 1) The non-transitory, computer-readable storage medium of claim 1, wherein calculating as at least one value of the vascular summary metric comprises: categorizing each of the plurality of the vessel segments as at least one of an artery and a non-artery; and calculating at least one of: an average arterial width, and a median arterial width. ### Claim 3 (depends on 2) The non-transitory, computer-readable storage medium of claim 2, wherein calculating as at least one value of the vascular summary metric further comprises at least one of: calculating a ratio of the average arterial width to an average venule width calculated using the widths of the vessel segments classified as non-artery; and calculating a ratio of the median arterial width to a median venule width calculated using the widths of the vessel segments classified as non-artery. ### Claim 4 (depends on 1) The non-transitory, computer-readable storage medium of claim 1, wherein calculating as at least one value of the vascular summary metric comprises: selecting a reduced subset of the subset of the plurality of vessel segments for which the widths have been determined; and calculating the at least one value of the vascular summary metric based on the widths of the selected reduced subset of the plurality of vessel segments. ### Claim 5 (depends on 4) The non-transitory, computer-readable storage medium of claim 4, wherein selecting the reduced subset comprises, in a case where each of the plurality of vessel segments has been categorized as at least one of veins and arteries, selecting, from among the vessel segments categorized as veins, a first predetermined number of widest vessel segments and selecting, from among the vessel segments categorized as arteries, a second predetermined number of widest vessel segments. ### Claim 6 (depends on 1) The non-transitory, computer-readable storage medium of claim 1, wherein the region of interest is an annulus sector of the retinal image, apart from the optic disc in the retinal image, showing a near periphery of the retina nasal to the optic disc. ### Claim 7 (depends on 1) The non-transitory, computer-readable storage medium of claim 1, wherein: processing the image data to identify a plurality of vessel segments further comprises processing the image data to locate retinal landmarks; and the region of interest is demarcated using the located retinal landmarks. ### Claim 8 (depends on 1) The non-transitory, computer-readable storage medium of claim 1, wherein processing the image data to identify the plurality of vessel segments comprises processing the image data to improve recognition of vessel segments by at least one of: pre-processing the image data using at least one of an adaptive histogram equalization technique and a morphological image crossing technique; enhancing at least one of the image data and pre-processed image data in order to aid detection of retinal features using Hessian filtering; creating the vessel segments by splitting a segmented vascular network; and cleaning the plurality of vessel segments by removing disconnected regions, branches, and offshoots that are unlikely to be the vessel segments. ### Claim 9 (depends on 1) The non-transitory, computer-readable storage medium of claim 1, wherein determining the respective widths of the vessel segments of at least a subset of the plurality of vessel segments comprises shortening the vessel segments such that the vessel segments have a more uniform width. ### Claim 10 (depends on 1) The non-transitory, computer-readable storage medium of claim 1, wherein determining the respective widths of the vessel segments of at least a subset of the plurality of vessel segments comprises: measuring a width of each of the vessel segments of the at least a subset of the plurality of vessel segments; and converting each measured width into metric units. ### Claim 11 (depends on 1) The non-transitory, computer-readable storage medium of claim 1, wherein determining a level of hypertension by which the person has been affected includes determining whether a retinal vasculature has been affected by hypertension by comparing the at least one value of the vascular summary metric to a predetermined threshold. ### Claim 12 An apparatus for determining a level of hypertension of a person, said apparatus comprising: an acquiring module configured to acquire image data defining a retinal image of a retina of the person that has been captured by a retinal imaging system; a processing module configured to process the image data based on a classification algorithm, which has been trained on image data defining retinal images of a plurality of subjects, to identify a plurality of vessel segments present in the retinal image, by: demarcating a region of interest, which is confined to a portion of the retinal image showing a near periphery of the retina that comprises an annular segment subtending 180 degrees nasal to the optic disc, from a remainder of the retinal image; and processing the image data to identify the plurality of vessel segments in the region of interest of the retinal image; a determining module configured to determine respective widths of vessel segments of at least a subset of the plurality of vessel segments present in the region of interest of the retinal image; and a calculating module configured to calculate, as at least one value of a vascular summary metric, at least one of an average value of and a median value of the determined widths of the vessel segments present in the region of interest of the retinal image, and to use the at least one value of the vascular summary metric and a relationship between the vascular summary metric and a level of hypertension to determine a level of hypertension by which the person has been affected. ### Claim 13 (depends on 12) The apparatus of claim 12, wherein the calculating module is further configured to calculate the at least one value of the vascular summary metric by categorizing each of the plurality of the vessel segments as at least one of an artery and a non-artery; and calculating at least one of: an average arterial width, and a median arterial width. ### Claim 14 (depends on 12) The apparatus of claim 12, wherein the calculating module is further configured to calculate the at least one value of the vascular summary metric by at least one of: calculating a ratio of the average arterial width to an average venule width calculated using the widths of the vessel segments classified as non-artery, and calculating a ratio of the median arterial width to a median venule width calculated using the widths of the vessel segments classified as non-artery. ### Claim 15 (depends on 12) The apparatus of claim 12, wherein the calculating module is further configured to calculate the at least one value of the vascular summary metric by selecting a reduced subset of the subset of the plurality of vessel segments for which the widths have been determined; and calculating the at least one value of the vascular summary metric based on the widths of the selected reduced subset of the plurality of vessel segments. ### Claim 16 (depends on 15) The apparatus of claim 15, wherein the calculating module is further configured to select the reduced subset by, in a case where each of the plurality of vessel segments has been categorized as at least one of veins and arteries, selecting, from among the vessel segments categorized as veins, a first predetermined number of widest vessel segments and selecting, from among the vessel segments categorized as arteries, a second predetermined number of widest vessel segments. ### Claim 17 (depends on 12) The apparatus of claim 12, wherein the processing module is further configured to process the image data to identify the plurality of vessel segments based on a classification algorithm trained on image data defining retinal images of a plurality of subjects. ### Claim 18 (depends on 12) The apparatus of claim 12, wherein the classification algorithm comprises a neural network trained on the image data to generate a set of parameters for the neural network. ### Claim 19 A non-transitory, computer-readable storage medium storing computer program instructions which, when executed by a computer, cause the computer to execute a method for determining a level of hypertension of a person, said method comprising: acquiring image data defining a retinal image of a retina of the person that has been captured by a retinal imaging system; processing the image data based on a classification algorithm, which has been trained on image data defining retinal images of a plurality of subjects, to identify a plurality of vessel segments present in the retinal image, wherein the processing comprises: demarcating a region of interest, which is confined to a portion of the retinal image showing a near periphery of the retina that comprises a half-concentric region nasal to the optic disc, from a remainder of the retinal image; and processing the image data to identify the plurality of vessel segments in the region of interest of the retinal image; determining respective widths of vessel segments of at least a subset of the plurality of vessel segments present in the region of interest of the retinal image; calculating, as at least one value of a vascular summary metric, at least one of an average value of and a median value of the determined widths of the vessel segments present in the region of interest of the retinal image; and using the at least one value of the vascular summary metric and a relationship between the vascular summary metric and a level of hypertension to determine a level of hypertension by which the person has been affected. ### Claim 20 An apparatus for determining a level of hypertension of a person, said apparatus comprising: an acquiring module configured to acquire image data defining a retinal image of a retina of the person that has been captured by a retinal imaging system; a processing module configured to process the image data based on a classification algorithm, which has been trained on image data defining retinal images of a plurality of subjects, to identify a plurality of vessel segments present in the retinal image, by: demarcating a region of interest, which is confined to a portion of the retinal image showing a near periphery of the retina that comprises a half-concentric region nasal to the optic disc, from a remainder of the retinal image; and processing the image data to identify the plurality of vessel segments in the region of interest of the retinal image; a determining module configured to determine respective widths of vessel segments of at least a subset of the plurality of vessel segments present in the region of interest of the retinal image; and a calculating module configured to calculate, as at least one value of a vascular summary metric, at least one of an average value of and a median value of the determined widths of the vessel segments present in the region of interest of the retinal image, and to use the at least one value of the vascular summary metric and a relationship between the vascular summary metric and a level of hypertension to determine a level of hypertension by which the person has been affected.