# US 12,213,827 B2 · Method and data processing system for providing a stroke information > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,213,827 B2 - Document ID: US12213827B2 - Status: Granted - Kind: B2 - Application number: 17/764,326 - Filed: 2020-09-18 - Published: 2022-12-15 - Granted: 2025-02-04 - Earliest claimed: 2019-09-30 - Assignees: Siemens Healthineers AG - Applicants: Siemens Healthcare GmbH - Inventors: Oliver Taubmann; Hendrik Ditt - Web page: https://patentgrep.com/patent/US12213827B2 ## CPC - A61B 6/5211: {involving processing of medical diagnostic data} - G06T 7/0014: {using an image reference approach} - A61B 6/037: {Emission tomography} - A61B 6/501: {for diagnosis of the head, e.g. neuroimaging or craniography} - G06T 7/0016: {involving temporal comparison} - G06T 7/30: Determination of transform parameters for the alignment of images, i.e. image registration - A61B 6/032: {Transmission computed tomography [CT]} - A61B 6/482: {involving multiple energy imaging} - A61B 6/5205: {involving processing of raw data to produce diagnostic data} - G06T 2207/10081: Computed x-ray tomography [CT] - G06T 2207/20081: Training; Learning - G06T 2207/20084: Artificial neural networks [ANN] - G06T 2207/30016: Brain - G06T 2207/30101: Blood vessel; Artery; Vein; Vascular - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract At least one example embodiment relates to a computer-implemented method for providing stroke information, the method comprising receiving computed tomography imaging data of an examination area of a patient, the examination area of the patient comprising a plurality of brain regions, at least one brain region of the plurality of brain regions being affected by a stroke, receiving brain atlas data, generating registered imaging data based on the computed tomography imaging data and the brain atlas data, the registered imaging data being registered to the brain atlas data, generating the stroke information regarding the stroke based on a set of algorithms and the registered imaging data, and providing the stroke information. ## Claims ### Claim 1 A computer-implemented method for providing stroke information, the method comprising: receiving computed tomography imaging data of an examination area of a patient, the examination area of the patient comprising a plurality of brain regions, at least one brain region of the plurality of brain regions being affected by a stroke; receiving brain atlas data; generating registered imaging data based on the computed tomography imaging data and the brain atlas data, the registered imaging data being registered to the brain atlas data; generating the stroke information regarding the stroke based on a set of algorithms and the registered imaging data; and providing the stroke information. ### Claim 2 (depends on 1) The method according to claim 1, further comprising: calculating, for each brain region of the plurality of brain regions, a local density average of the brain region based on at least one of the computed tomography imaging data or the registered imaging data; calculating, for a reference brain region, a density average of the reference brain region based on at least one of the computed tomography imaging data or the registered imaging data, the examination area of the patient comprising the reference brain region; and calculating, for each brain region of the plurality of brain regions, a density difference information of the brain region based on the local density average of the brain region and the density average of the reference brain region, thereby obtaining a density difference map, wherein the stroke information is generated further based on the density difference map. ### Claim 3 (depends on 2) The method according to claim 2, wherein the reference brain region comprises, for each brain region of the plurality of brain regions, a respective subregion, such that a plane that is perpendicular to a sagittal plane of the patient is located between that brain region and the respective subregion. ### Claim 4 (depends on 2) The method according to claim 2, wherein the reference brain region comprises, for each brain region of the plurality of brain regions, a respective subregion, such that the brain region and the respective subregion are located within the same cerebral hemisphere. ### Claim 5 (depends on 1) The method according to claim 1, further comprising: generating mirrored imaging data based on a left-right mirroring of at least one of the computed tomography imaging data or the registered imaging data; and calculating a left-right difference image based on the computed tomography imaging data and the mirrored imaging data, wherein the stroke information is generated further based on the left-right difference image. ### Claim 6 (depends on 5) The method according to claim 5, wherein the set of algorithms comprises an algorithm for calculating an onset time of the stroke, the onset time of the stroke is calculated by applying the algorithm for calculating the onset time of the stroke onto at least one of the left-right difference image or the density difference map, and the stroke information at least one of comprises the calculated onset time of the stroke or is generated further based on the calculated onset time of the stroke. ### Claim 7 (depends on 1) The method according to claim 1, wherein the set of algorithms comprises an algorithm for generating a representation of a blood clot, the representation of the blood clot is generated by applying the algorithm for generating the representation of the blood clot onto the registered imaging data and the brain atlas data, and the stroke information at least one of comprises the representation of the blood clot or (ii) is generated further based on the representation of the blood clot. ### Claim 8 (depends on 7) The method according to claim 7, wherein the set of algorithms comprises an algorithm for generating clot composition information, the clot composition information is generated by applying the algorithm for generating the clot composition information onto the registered imaging data and the representation of the blood clot, and the stroke information at least one of comprises the clot composition information or (ii) is generated further based on the clot composition information. ### Claim 9 (depends on 1) The method according to claim 1, further comprising: generating an output-output consistency information based on an output of an algorithm of the set of algorithms and an output selected from a group consisting of a left-right difference image, a density difference map and an output of at least one other algorithm of the set of algorithms, wherein the stroke information is generated based on the output-output consistency information. ### Claim 10 (depends on 1) The method according to claim 1, further comprising: receiving prior information regarding the stroke; and generating an input-output consistency information based on the prior information and an output selected from a group consisting of a left-right difference image, a density difference map and an output of an algorithm of the set of algorithms, wherein the stroke information is generated based on the input-output consistency information. ### Claim 11 (depends on 1) The method according to claim 1, wherein at least one algorithm of the set of algorithms is based on a trained machine learning model, and the generating of the stroke information regarding the stroke is based on the trained machine learning model and the registered imaging data. ### Claim 12 (depends on 11) The method according to claim 11, wherein the trained machine learning model is based on a plurality of training datasets, each training dataset of the plurality of training datasets comprising a training input in form of training computed tomography imaging data and a corresponding training output in form of a training stroke information. ### Claim 13 (depends on 12) The method according to claim 12, wherein one of left and right cerebral hemispheres is selected, and onto each of those training datasets of the plurality of training datasets, for which the cerebral hemisphere that is affected by a respective stroke does not correspond to the selected cerebral hemisphere, a left-right mirroring is applied. ### Claim 14 (depends on 12) The method according to claim 12, wherein for each of the training datasets of the plurality of training datasets: a normalization information is calculated based on the respective training input and the brain atlas data and a normalization is applied onto the respective training input based on the normalization information, thereby obtaining a respective normalized training input, and a normalization information for the registered imaging data is calculated based on the registered imaging data and the brain atlas data, the generating of the stroke information regarding the stroke is further based on the normalization information for the registered imaging data. ### Claim 15 (depends on 1) A data processing system configured to implement the method of claim 1. ### Claim 16 (depends on 1) A non-transitory computer program product comprising program elements that, when executed by a data processing system, cause the data processing system to perform the method of claim 1. ### Claim 17 (depends on 1) A non-transitory computer-readable medium storing program elements that can be read and executed by a data processing system, in order to perform the method of claim 1, when the program elements are executed by the data processing system. ### Claim 18 A data processing system configured to provide stroke information, the data processing system comprising: an imaging data receiver configured to receive computed tomography imaging data of an examination area of a patient, the examination area of the patient comprising a plurality of brain regions, at least one brain region of the plurality of brain regions being affected by a stroke; a brain atlas receiver configured to receive brain atlas data; a registered-image generator configured to generate registered imaging data based on the computed tomography imaging data and the brain atlas data, the registered imaging data being registered to the brain atlas data; a stroke-information generator configured to generate the stroke information regarding the stroke based on a set of algorithms and the registered imaging data; and a stroke-information provider configured to provide the stroke information. ### Claim 19 A data processing system configured to provide stroke information, the data processing system comprising: a memory storing computer-readable instructions; and at least one processor configured to execute the computer-readable instructions to cause the data processing system to, receive computed tomography imaging data of an examination area of a patient, the examination area of the patient comprising a plurality of brain regions, at least one brain region of the plurality of brain regions being affected by a stroke, receive brain atlas data, generate registered imaging data based on the computed tomography imaging data and the brain atlas data, the registered imaging data being registered to the brain atlas data, generate the stroke information regarding the stroke based on a set of algorithms and the registered imaging data, and provide the stroke information.