# US 12,002,581 B2 · Diagnostic modelling method and apparatus > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,002,581 B2 - Document ID: US12002581B2 - Status: Granted - Kind: B2 - Application number: 16/772,603 - Filed: 2018-12-12 - Published: 2020-12-10 - Granted: 2024-06-04 - Earliest claimed: 2017-12-13 - Assignees: Oxford University Innovation Limited - Applicants: Oxford University Innovation Limited - Inventors: Ross Upton; Paul Leeson - Web page: https://patentgrep.com/patent/US12002581B2 ## CPC - G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems - A61B 5/7267: {involving training the classification device} - A61B 8/0883: {for diagnosis of the heart} - A61B 8/5223: {for extracting a diagnostic or physiological parameter from medical diagnostic data (for algorithms to analyse biomedical images G06T7/0012)} - G06T 7/0012: {Biomedical image inspection} - G06T 7/0016: {involving temporal comparison} - A61B 8/463: {characterised by displaying multiple images or images and diagnostic data on one display} - G06T 2207/10132: Ultrasound image - G06T 2207/30048: Heart; Cardiac - 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 ## Abstract The present disclosure relates to a system (100) for generating a diagnostic model. The system (100) includes a processor (108) configured to analyse a plurality of reference data sets. The reference data sets each include at least one image (230, 240). The analysis identities at least one feature in each image (230, 240). A metric is calculated in dependence on the at least one identified feature. Outcome data associated with at least some of the reference data sets is acquired. The diagnostic model is compiled in dependence on the at least one calculated metric and the associated outcome data. The present disclosure also relates to a method of generating a diagnostic model; and a non-transitory computer-readable medium. ## Claims ### Claim 1 A system for generating a cardiac diagnostic model, the system comprising a processor configured to: analyse a plurality of reference data sets, each reference data set comprising at least one image, the analysis comprising identifying at least one feature in each image; calculate at least one metric in dependence on the at least one identified feature; acquire outcome data associated with at least some of the reference data sets; and compile the diagnostic model in dependence on the at least one calculated metric and the associated outcome data; wherein each reference data set comprises first and second images, the processor being configured to analyse each reference data set to: identify at least one first feature in the first image, and identify at least one second feature in the second image, each at least one first feature being paired with a corresponding one of the at least one second feature; and compare each pair of corresponding first and second features to identify one or more difference therebetween; wherein the processor is configured to calculate the at least one metric in dependence on the one or more difference identified between each pair of corresponding first and second features; wherein the first image comprises a first end systolic image and the second image comprises a second end diastolic image, the one or more difference identified between each pair of corresponding first and second features represents a cardiac cyclic change, wherein the at least one metric includes principal transformation and/or shear transformation; wherein the outcome data comprises diagnostic information indicating whether the reference data set is from a patient having a positive or negative diagnosis for a cardiac condition, and is associated with a corresponding set of data in the reference data set; and wherein the diagnostic model is updated when new outcome data associated with a corresponding set of data in the reference data set becomes available or when the outcome data associated with a corresponding set of data in the reference data set is updated. ### Claim 2 (depends on 1) A system as claimed in claim 1, wherein the processor is configured to label the images in dependence on the outcome data. ### Claim 3 (depends on 2) A system as claimed in claim 2, wherein the label distinguishes between different classifications. ### Claim 4 (depends on 1) A system as claimed in claim 1, wherein the diagnostic information comprises a record of a cardiac event, such as a myocardial infarction. ### Claim 5 (depends on 1) A system as claimed in claim 1, wherein the outcome data comprises a record of stenosis greater than a threshold value. ### Claim 6 (depends on 1) A system as claimed in claim 1, wherein the processor is configured to determine a weighting for at least some of the calculated metrics, each weighting being determined in dependence on the outcome data associated with a given one of the reference data sets. ### Claim 7 (depends on 6) A system as claimed in claim 6, wherein the diagnostic model is compiled in dependence on the calculated metrics and the associated weightings. ### Claim 8 (depends on 1) A system as claimed in claim 1, wherein the processor is configured to implement a machine learning algorithm to generate the diagnostic model. ### Claim 9 (depends on 1) A system as claimed in claim 1, wherein the processor is configured to analyse further reference data sets; and to update the diagnostic model in dependence on the analysis of the further reference data sets. ### Claim 10 A method of generating a cardiac diagnostic model, the method comprising: analysing a plurality of reference data sets, each reference data set comprising at least one image, the analysis comprising identifying at least one feature in each image; calculating at least one metric in dependence on the at least one identified feature; acquiring outcome data associated with at least some of the reference data sets; and compiling the diagnostic model in dependence on the calculated metrics and the associated outcome data; wherein each reference data set comprises first and second images, the method comprising analysing each reference data set to: identify at least one first feature in the first image, and identify at least one second feature in the second image, each at least one first feature being paired with a corresponding one of the at least one second feature; and comparing each pair of corresponding first and second features to identify one or more difference therebetween; comprising calculating the at least one metric in dependence on the one or more difference identified between each pair of corresponding first and second features; wherein the first image comprises a first end systolic image and the second image comprises a second end diastolic image, the one or more difference identified between each pair of corresponding first and second features represents a cardiac cyclic change, wherein the at least one metric includes principal transformation and/or shear transformation; wherein the outcome data comprises diagnostic information indicating whether the reference data set is from a patient having a positive or negative diagnosis for a cardiac condition, and is associated with a corresponding set of data in the reference data set; and wherein the diagnostic model is updated when new outcome data associated with a corresponding set of data in the reference data set becomes available, or when the outcome data associated with a corresponding set of data in the reference data set is updated. ### Claim 11 (depends on 10) A method as claimed in claim 10, wherein the outcome data is used to label the corresponding reference data sets. ### Claim 12 (depends on 11) A method as claimed in claim 11, wherein the label distinguishes between different classifications. ### Claim 13 (depends on 10) A method as claimed in claim 10, wherein the diagnostic information comprises a record of a cardiac event, such as a myocardial infarction. ### Claim 14 (depends on 10) A method as claimed in claim 10, wherein the outcome data comprises a record of stenosis greater than a threshold value. ### Claim 15 (depends on 10) A method as claimed in claim 10 comprising determining a weighting for at least some of the calculated metrics, each weighting being determined in dependence on the outcome data associated with a given one of the reference data sets. ### Claim 16 (depends on 15) A method as claimed in claim 15, wherein the diagnostic model is compiled in dependence on the calculated metrics and the associated weightings. ### Claim 17 (depends on 10) A method as claimed in claim 10, wherein a machine learning algorithm is implemented to generate the diagnostic model. ### Claim 18 (depends on 10) A method as claimed in claim 10 comprising adding further reference data sets incrementally, the method comprising analysing the further reference data sets; and updating the diagnostic model in dependence on the analysis of the further reference data sets. ### Claim 19 (depends on 10) A non-transitory computer-readable medium having a set of instructions stored therein which, when executed, cause a processor to perform the method claimed in claim 10.