WO2022145542A1 - Procédé et dispositif d'aide au diagnostic d'une maladie cardiovasculaire - Google Patents

Procédé et dispositif d'aide au diagnostic d'une maladie cardiovasculaire Download PDF

Info

Publication number
WO2022145542A1
WO2022145542A1 PCT/KR2020/019452 KR2020019452W WO2022145542A1 WO 2022145542 A1 WO2022145542 A1 WO 2022145542A1 KR 2020019452 W KR2020019452 W KR 2020019452W WO 2022145542 A1 WO2022145542 A1 WO 2022145542A1
Authority
WO
WIPO (PCT)
Prior art keywords
diagnosis
information
cardiovascular disease
auxiliary
target
Prior art date
Application number
PCT/KR2020/019452
Other languages
English (en)
Korean (ko)
Inventor
최태근
이근영
Original Assignee
주식회사 메디웨일
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 주식회사 메디웨일 filed Critical 주식회사 메디웨일
Priority to PCT/KR2020/019452 priority Critical patent/WO2022145542A1/fr
Publication of WO2022145542A1 publication Critical patent/WO2022145542A1/fr

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure

Definitions

  • the present specification relates to a method and apparatus for assisting diagnosis of cardiovascular disease, and to a method and apparatus for assisting diagnosis of cardiovascular disease using an artificial neural network model.
  • This specification is derived from research conducted as part of the construction of an open competition platform through the Information and Communication Industry Promotion Agency (NIPA, Korea) created by the Information Promotion Fund of the Ministry of Science and ICT (MSIT, Korea) in 2020. [Task Management No.: A0712-20-1002]
  • Fundus examination can observe abnormalities in the retina, optic nerve, and macula, and is a diagnostic aid frequently used in ophthalmology because the results can be confirmed through relatively simple imaging.
  • the use of the fundus examination is increasing in that it is possible to observe the degree of damage to blood vessels caused by chronic diseases such as high blood pressure and diabetes in a non-invasive way as well as eye diseases.
  • One object of the present invention is to provide a method for assisting diagnosis of cardiovascular disease.
  • Another object of the present invention is to provide a method of assisting in the diagnosis of cardiovascular disease using a neural network model based on a fundus image.
  • a method for assisting in diagnosis of a target cardiovascular disease using a fundus image comprising: acquiring a target fundus image obtained by photographing the fundus of a subject; based on the target fundus image, the fundus image Acquiring the cardiovascular disease diagnosis auxiliary information of the subject according to the target fundus image through the cardiovascular disease diagnosis auxiliary neural network model for obtaining the diagnosis auxiliary information used for the diagnosis of the target cardiovascular disease according to The step of outputting information, wherein the cardiovascular disease diagnosis auxiliary information includes grade information including a grade selected from a plurality of grades indicating the risk of the target cardiovascular disease, score information that is numerical information for determining the risk of the target cardiovascular disease, and blood
  • a method for assisting in diagnosing cardiovascular disease including at least one of risk information indicating whether a sample corresponds to a target cardiovascular disease risk group may be provided.
  • information that can be used for diagnosing cardiovascular disease may be acquired based on the fundus image.
  • various types of information that can be used for diagnosing cardiovascular disease may be acquired based on the fundus image.
  • FIG. 1 is a diagram for explaining an example of a method for assisting diagnosis of cardiovascular disease.
  • FIG. 2 is a diagram for describing a diagnosis unit according to an exemplary embodiment.
  • FIG. 3 is a view for explaining a method of assisting in diagnosing cardiovascular disease according to an exemplary embodiment.
  • FIG. 4 is a view for explaining a method for assisting in diagnosing a cardiovascular disease according to another exemplary embodiment.
  • FIG. 5 is a diagram for explaining a method of assisting in diagnosis of cardiovascular disease according to another exemplary embodiment.
  • FIG. 6 is a diagram for explaining a method for assisting diagnosis of cardiovascular disease according to another exemplary embodiment.
  • a method for assisting in diagnosis of a target cardiovascular disease using a fundus image comprising: acquiring a target fundus image obtained by photographing the fundus of a subject; based on the target fundus image, the fundus image Acquiring the cardiovascular disease diagnosis auxiliary information of the subject according to the target fundus image through the cardiovascular disease diagnosis auxiliary neural network model for obtaining the diagnosis auxiliary information used for the diagnosis of the target cardiovascular disease according to The step of outputting information, wherein the cardiovascular disease diagnosis auxiliary information includes grade information including a grade selected from a plurality of grades indicating the risk of the target cardiovascular disease, score information that is numerical information for determining the risk of the target cardiovascular disease, and blood
  • a method for assisting in diagnosing cardiovascular disease including at least one of risk information indicating whether a sample corresponds to a target cardiovascular disease risk group may be provided.
  • the method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium.
  • the computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination.
  • the program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and available to those skilled in the art of computer software.
  • Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks.
  • - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like.
  • Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.
  • the hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.
  • a system, apparatus, method, and the like for assisting in the diagnosis of cardiovascular disease using a fundus image.
  • a system, apparatus, method, and the like for assisting in the diagnosis of cardiovascular disease by using a neural network model and obtaining diagnostic auxiliary information useful for diagnosis of cardiovascular disease based on a fundus image.
  • biomarkers used directly or indirectly for disease diagnosis may be used.
  • a method of managing the risk of a disease may be used in consideration of values of disease-related indexes, scores, or indicators (hereinafter, scores). With respect to diseases diagnosed in consideration of a value such as a score, it may be more efficient to provide a score, not the presence or absence of a disease, because the clinician may directly consider the score to determine the patient's condition or treatment.
  • the cardiovascular disease described herein may refer to a cerebrovascular disease.
  • Cardiovascular disease includes coronary artery disease such as myocardial infarction or angina, coronary artery disease, ischemic heart disease, congestive heart failure, peripheral vascular disease, heart attack, heart valve disease, cerebrovascular disease (eg, stroke). , cerebral infarction, cerebral hemorrhage or transient ischemic attack) and may refer to diseases related to the brain, heart or blood vessels, including renal and vascular diseases.
  • cardiovascular diseases described herein can be accompanied by complications.
  • cardiovascular disease can be accompanied by complications such as heart attack, heart failure, stroke, aneurysm, peripheral arterial disease, renal failure, dementia or skin ulceration.
  • Cardiovascular diseases described herein may refer to these complications.
  • diagnostic assistance information used for diagnosing a disease may be provided.
  • the diagnostic auxiliary information may include a parameter value related to a cardiovascular disease, a grade indicating a risk of a cardiovascular disease, or information on the presence or absence of a cardiovascular disease.
  • the score assisting in the diagnosis of cardiovascular disease may be a score that can be measured from a subject or a score calculated by combining values measured from the subject and/or personal information of the subject.
  • the score used for the diagnosis of cardiovascular disease may be a score suggested by a known cardiovascular disease prediction model.
  • a score that aids in the diagnosis of cardiovascular disease may be a cardiac calcification index indicating a degree of cardiac calcification.
  • a score that aids in the diagnosis of cardiovascular disease may be a coronary calcium score.
  • the score may be an atherosclerosis risk score.
  • the score supporting the diagnosis of cardiovascular disease may be a Carotid Intima-Media Thickness (CIMT) value.
  • the score may be a Framingham Coronary Risk Score.
  • the score supporting the diagnosis of cardiovascular disease may be a value for at least one factor included in the Framingham risk score.
  • the score may be a QRISK score.
  • the score supporting the diagnosis of cardiovascular disease may be a value according to ASCVD (Atherosclerotic Cardiovascular Disease).
  • the score may be a score according to European Systematic Coronary Risk Evaluation (SCORE).
  • a coronary calcium score it may be used as a determination index for calcification of the coronary artery.
  • the coronary arteries are calcified as plaque accumulates in the blood vessels, the walls of the cardiovascular vessels narrow, causing various heart diseases, such as coronary artery disease, myocardial infarction, angina pectoris, and ischemic heart disease.
  • the coronary calcium index can be used as a basis for judging the risk of various heart diseases. For example, when the coronary calcium score value is large, it may be determined that the risk of coronary artery disease is large.
  • the coronary calcium score is directly related to heart disease, especially coronary artery disease (heart calcification), compared to factors indirectly related to heart disease such as smoking status, age, and sex, and is used as a strong biomarker for heart health.
  • coronary artery disease heart calcification
  • factors indirectly related to heart disease such as smoking status, age, and sex
  • diagnosis auxiliary information such as a score supporting the diagnosis of cardiovascular disease may be used as a criterion for selecting a specific medical treatment or prescription target.
  • a coronary calcium score may be used to select subjects for coronary workup.
  • the coronary calcium score may be used to select subjects for taking antihypertensive agents.
  • Coronary artery calcium score can be used as a standard for prescribing antihyperlipidemic agents such as statins.
  • the Framingham risk score value or a value used to calculate the Framingham risk score may be obtained and provided as diagnostic auxiliary information for determining the risk of coronary artery disease. For example, it may be determined that the higher the Framingham risk score, the higher the risk of coronary artery disease.
  • the carotid intima thickness value it may be obtained and provided as diagnostic auxiliary information for determining the risk of cerebral infarction or acute myocardial infarction. For example, the thicker the carotid intima, the greater the risk of cerebral infarction or acute myocardial infarction may be determined.
  • the grade that aids in the diagnosis of cardiovascular disease may be at least one grade indicative of a risk of cardiovascular disease. For example, when a score or the like may be used for diagnosis of a disease, a grade may be used in place of or together with a score or the like.
  • the diagnostic assistance information may include a cardiovascular disease diagnostic assistance score and/or a cardiovascular disease diagnostic assistance grade.
  • the grade may include a normal grade indicating that the subject is normal for the target cardiovascular disease and an abnormal grade indicating that the subject is abnormal for the target cardiovascular disease.
  • the ratings may include a plurality of ratings indicating a subject's degree of risk for a target cardiovascular disease. In this regard, it will be described in more detail in the section on methods for diagnosing cardiovascular diseases below.
  • Scores and the like described herein may be used for diagnosing a disease.
  • the score and the like may be used to diagnose a patient's current condition and/or to manage the prognosis of a disease.
  • the generation and provision of diagnostic auxiliary information including scores and the like will be described in more detail in the following diagnostic auxiliary information section.
  • cardiovascular disease diagnosis assistance information may be acquired prior to diagnosis of various cardiovascular diseases (eg, coronary artery disease).
  • cardiovascular disease diagnosis auxiliary information may be used for preliminary diagnosis for selecting subjects for detailed examination prior to the detailed examination for cardiovascular disease. In this regard, it will be described in more detail in the section on providing diagnostic auxiliary information described below.
  • the diagnostic assistance method, apparatus, and system described below use a neural network model to provide a numerical value as diagnostic assistance information related to the disease from a fundus image. Alternatively, it may be similarly applied to all cases of obtaining a grade.
  • a diagnosis assisting system and apparatus for acquiring the presence or absence of a cardiovascular disease or information serving as a basis for the determination based on the fundus image will be described.
  • deep learning to build a neural network model for predicting cardiovascular disease-related information, learning the built model, and predicting information using the learned model, systems and devices that aid in the diagnosis of cardiovascular disease, etc. about it.
  • a cardiovascular disease diagnosis assistance system that acquires cardiovascular disease diagnosis assistance information based on a fundus image may be provided.
  • a cardiovascular disease diagnosis assistance system or a device constituting the same may perform diagnostic assistance and/or cardiovascular disease diagnosis assistance described throughout this specification.
  • the cardiovascular disease diagnosis assistance system may include a learning device, a diagnosis device, and a client device, and each of the system and device may operate similarly to the diagnosis assistance system described above with reference to FIG. 1 .
  • the training device learns a neural network model for assisting in diagnosing cardiovascular disease
  • the diagnostic device performs diagnosing assist in cardiovascular disease using the learned neural network model
  • the client device acquires a fundus image, and a cardiovascular system generated based on the fundus image
  • the disease diagnosis auxiliary information may be obtained, and the diagnosis auxiliary information may be provided to the user.
  • the cardiovascular disease diagnosis assistance system may include a diagnostic device and a client device.
  • the diagnosis apparatus may perform functions of a learning apparatus and/or a server apparatus.
  • the diagnostic device and/or the client device may perform the cardiovascular disease diagnostic aid described herein.
  • the cardiovascular disease diagnosis assistance system may include a mobile device.
  • the mobile device may perform all or part of the operations of the learning device, the diagnostic device, and/or the client device described above.
  • the mobile device may perform the cardiovascular disease diagnosis assistant described herein.
  • a diagnostic assistance apparatus may be provided.
  • the diagnosis auxiliary device may include a cardiovascular disease diagnosis auxiliary unit.
  • the cardiovascular disease diagnosis assistant may perform the cardiovascular disease diagnosis assistant described herein.
  • the diagnosis assistance device may obtain diagnosis assistance information of a cardiovascular disease based on the fundus image.
  • the diagnostic assistant device may be one or more devices including a cardiovascular disease diagnostic assistant.
  • the diagnosis auxiliary device may be the above-described learning device, diagnosis device, or client device.
  • the diagnostic assistance device may be a mobile device.
  • the diagnostic auxiliary device may be a server device.
  • the diagnosis assistance device may include a cardiovascular disease diagnosis auxiliary unit that assists in the diagnosis of cardiovascular diseases described herein.
  • a diagnostic aid device may be included in any device or component described herein.
  • the learning device may assist in diagnosing a cardiovascular disease.
  • the learning device may include a cardiovascular disease diagnosis auxiliary unit that assists in diagnosing cardiovascular disease.
  • the cardiovascular disease diagnosis auxiliary unit of the learning apparatus may train a neural network model for predicting cardiovascular disease diagnosis auxiliary information based on the fundus image.
  • the cardiovascular disease diagnosis auxiliary unit of the learning apparatus may assist in the diagnosis of cardiovascular disease by learning a neural network model for predicting cardiovascular disease diagnosis auxiliary information based on the fundus image.
  • the learning unit of the learning device may perform a function of the cardiovascular disease diagnosis auxiliary unit supporting cardiovascular disease diagnosis.
  • the learning unit may include a cardiovascular disease diagnosis auxiliary unit.
  • the processor or the learning module of the learning device may perform a function of the cardiovascular disease diagnosis assistant.
  • the processor or the learning module of the learning device may include a cardiovascular disease diagnosis assistant.
  • the control unit of the learning device may assist in diagnosing a cardiovascular disease described herein.
  • the control unit may include a cardiovascular disease diagnosis auxiliary unit.
  • a memory unit (or a volatile memory, a non-volatile memory, or a mass storage device) of the learning device may store a neural network model for assisting cardiovascular disease diagnosis.
  • the communication unit of the learning apparatus may transmit the learned model or information for driving the learned model to an external device. Alternatively, the learning device may acquire information necessary for learning the neural network model from an external device through the communication unit.
  • the diagnostic device may assist in diagnosing a cardiovascular disease.
  • the diagnosis apparatus may include a cardiovascular disease diagnosis auxiliary unit that assists in diagnosing a cardiovascular disease.
  • the cardiovascular disease diagnosis auxiliary unit of the diagnosis apparatus may obtain cardiovascular disease diagnosis auxiliary information on the target fundus image by using the learned neural network model.
  • the cardiovascular disease diagnosis auxiliary unit of the diagnosis apparatus may obtain the cardiovascular disease diagnosis auxiliary information by using a neural network model that outputs the cardiovascular disease diagnosis auxiliary information based on the fundus image.
  • the diagnosis unit of the diagnosis apparatus may perform a function of the cardiovascular disease diagnosis auxiliary unit supporting cardiovascular disease diagnosis.
  • the diagnostic unit may include a cardiovascular disease diagnosis assistant.
  • a processor or a diagnosis module of the diagnosis apparatus may perform a function of a cardiovascular disease diagnosis auxiliary unit.
  • a processor or a diagnostic module of the diagnostic apparatus may include a cardiovascular disease diagnostic assistant.
  • the controller of the diagnosis apparatus may assist in diagnosing a cardiovascular disease by using the learned neural network model.
  • the control unit of the diagnosis apparatus may perform a function of the cardiovascular disease diagnosis auxiliary unit.
  • the control unit of the diagnosis apparatus may include a cardiovascular disease diagnosis auxiliary unit.
  • the memory unit of the diagnosis apparatus may store a neural network model learned to assist in diagnosing a cardiovascular disease.
  • the memory unit of the diagnosis apparatus may include a cardiovascular disease diagnosis auxiliary unit.
  • the diagnostic apparatus may communicate with an external device using a communication unit.
  • the diagnosis apparatus may acquire a diagnosis target image from an external device or transmit diagnosis auxiliary information to the external device using the communication unit.
  • the diagnosis apparatus may acquire a neural network model learned from an external device (eg, a learning device) or information necessary for driving the learned neural network model by using the communication unit.
  • the client device may assist in the diagnosis of cardiovascular disease.
  • the client device may include a cardiovascular disease diagnosis assistant to assist in the diagnosis of cardiovascular disease.
  • the cardiovascular disease diagnosis assistant of the client device may learn a neural network model, obtain diagnostic assistance information using the neural network model, or provide data (eg, fundus image) necessary for driving the neural network model.
  • the client device may obtain information necessary for assisting the diagnosis of cardiovascular disease from the user, or may provide information for assisting diagnosis of cardiovascular disease to the user.
  • the control unit of the client device may include a cardiovascular disease diagnosis auxiliary unit.
  • the control unit of the client device may perform a function of the cardiovascular disease diagnosis auxiliary unit.
  • the processor of the client device may include a cardiovascular disease diagnosis assistant or perform cardiovascular disease diagnosis assistance.
  • the server device may assist in the diagnosis of cardiovascular disease.
  • the server device may include a cardiovascular disease diagnosis auxiliary unit that assists in the diagnosis of cardiovascular disease.
  • the cardiovascular disease diagnosis assistant of the server device may store, learn, or drive the neural network model.
  • the server device may store data (eg, fundus image data) necessary for storing, learning, or driving a neural network model supporting cardiovascular disease diagnosis.
  • the server device may store user information used to assist cardiovascular disease diagnosis.
  • Cardiovascular disease diagnosis assistance includes learning the cardiovascular disease diagnosis auxiliary neural network model using the learning target fundus image, and acquiring cardiovascular disease diagnosis auxiliary information based on the diagnosis target fundus image using the learned cardiovascular disease diagnosis auxiliary neural network model.
  • the cardiovascular disease diagnosis auxiliary neural network model may be a multi-layered neural network model that outputs an auxiliary diagnosis result related to cardiovascular disease.
  • the cardiovascular disease diagnosis auxiliary neural network model may be a convolutional neural network model that acquires diagnostic auxiliary information based on a fundus image.
  • the cardiovascular disease diagnosis auxiliary neural network model may be prepared in the form of an ensemble.
  • the cardiovascular disease diagnosis auxiliary neural network model includes a first sub-neural network model that outputs a first result and a second sub-neural network model that outputs a second result, and the obtained diagnostic auxiliary information includes the first result and the second result. It can be decided by considering the results together.
  • Cardiovascular disease diagnosis assistance can be largely divided into training of the cardiovascular disease diagnosis secondary neural network model and diagnosis assistance using the learned cardiovascular disease diagnosis secondary neural network model.
  • Learning the cardiovascular disease diagnosis auxiliary neural network model may include acquiring training data and learning the cardiovascular disease diagnosis auxiliary neural network model based on the acquired training data.
  • Acquiring the training data may include acquiring fundus image training data.
  • the acquired fundus image learning data may be fundus image learning data labeled with cardiovascular disease diagnosis information. In this regard, it will be described in more detail in the fundus image acquisition section below.
  • Training the cardiovascular disease diagnosis auxiliary model may include transforming (or pre-processing) the acquired fundus image in some cases.
  • the cardiovascular disease diagnosis auxiliary neural network model may be trained using the transformed fundus image.
  • the fundus image may be converted or pre-processed into a form more suitable for obtaining diagnostic auxiliary information for cardiovascular disease. In this regard, it will be looked at in more detail in the fundus image pre-processing section below.
  • Training the cardiovascular disease diagnosis auxiliary neural network model may include predicting a result on unit training data, comparing the predicted result with a label, and repeating the steps of updating the neural network model several times.
  • Aiding the diagnosis of cardiovascular disease using the neural network model may include acquiring an image of a fundus to be diagnosed and acquiring aiding information for diagnosis of a cardiovascular disease from the fundus image to be diagnosed using the learned neural network model. have.
  • Acquiring the fundus image to be diagnosed may be performed through an imaging unit or a captured image may be acquired through a separate imaging device.
  • Assisting in the diagnosis of cardiovascular disease may include preprocessing the fundus image to be diagnosed.
  • the acquired fundus image to be diagnosed may be pre-processed.
  • Acquiring the diagnosis auxiliary information of cardiovascular disease may be obtaining the diagnosis auxiliary information on the fundus image preprocessed using a learned neural network model. In this regard, it will be looked at in more detail in the fundus image pre-processing section below.
  • the pre-processing of the fundus image to be diagnosed may include converting or pre-processing the fundus image to be diagnosed into a form more suitable for obtaining diagnostic auxiliary information of cardiovascular disease.
  • Acquiring the diagnosis auxiliary information of cardiovascular disease using the learned neural network model may include acquiring disease information, numerical information, grade level, etc. that can be used for diagnosis of cardiovascular disease.
  • the diagnosis assistance process performs preprocessing on an image (S101), learns a cardiovascular disease diagnosis auxiliary neural network model based on the preprocessed image (S103), and , the learning process of acquiring the parameters of the learned cardiovascular disease diagnosis auxiliary neural network model (S105), and after acquiring the diagnosis target image and preprocessing the diagnosis target image (S201), using the learned cardiovascular disease diagnosis auxiliary neural network model (S203)
  • a diagnosis auxiliary process of obtaining ( S205 ) auxiliary diagnosis auxiliary information for diagnosing cardiovascular disease may be included.
  • the training process of the cardiovascular disease diagnosis auxiliary neural network model includes a preprocessing step of preprocessing the fundus image to improve the prediction accuracy of cardiovascular disease diagnosis information, and learning to train the cardiovascular disease diagnosis auxiliary neural network model using the preprocessed fundus image It can include processes.
  • the training process may be performed by the learning device.
  • the diagnosis auxiliary process using the cardiovascular disease diagnosis auxiliary neural network model may include a preprocessing process for preprocessing the input target fundus image and a diagnosis auxiliary process for supporting the diagnosis of cardiovascular disease using the preprocessed fundus image.
  • the diagnostic auxiliary process may be performed by the aforementioned diagnostic device server device.
  • a cardiovascular disease diagnosis auxiliary neural network model that acquires a score related to a patient's cardiovascular disease based on a fundus image may be trained.
  • the cardiovascular disease diagnosis auxiliary neural network model may be trained to predict a score used for diagnosis of cardiovascular disease based on the fundus image.
  • Learning of the cardiovascular disease diagnosis auxiliary neural network model that acquires a score may be performed by the above-described learning unit.
  • the cardiovascular disease diagnosis assistance neural network model may be trained to predict the corresponding diagnostic assistance score from fundus images.
  • the cardiovascular disease diagnosis adjunct neural network model may be trained to predict a value of a specific parameter related to a target cardiovascular disease of a patient.
  • the cardiovascular disease diagnosis adjunct neural network model may be trained to predict a score that can be used in diagnosing a target cardiovascular disease for a patient.
  • the adjuvant neural network model for diagnosing cardiovascular disease is the coronary calcium score
  • the score is the arteriosclerosis risk score
  • the carotid Intima-Media Thickness (CIMT) value the ankle-brachial index
  • the vascular stiffness test is the adjuvant neural network model for diagnosing cardiovascular disease.
  • the cardiovascular disease diagnosis auxiliary neural network model may be trained to predict a score based on an input fundus image. Scores can be predicted as real values. The score may be predicted as an integer value. The score may be predicted as a positive value.
  • the cardiovascular disease diagnosis auxiliary neural network model may be trained using a fundus image training dataset including fundus images with score labels.
  • the cardiovascular disease diagnosis auxiliary neural network model may be trained using fundus image learning data including a fundus image labeled with a coronary calcium score.
  • the cardiovascular disease diagnosis auxiliary neural network model may be trained using a fundus image training data set including fundus images to which labels (eg, disease presence labels or grade labels) are assigned rather than score labels.
  • the cardiovascular disease diagnosis auxiliary neural network model may be trained using fundus image learning data including a fundus image to which a score label and a label other than a score label are given together.
  • a neural network model trained to predict a score based on the fundus image may be trained in the form of a linear regression model that outputs continuous values.
  • the cardiovascular disease diagnosis auxiliary neural network model may include a linear regression neural network model for predicting scores and a classifier model for outputting grade or disease information.
  • a score prediction model including a primary neural network model and a secondary neural network model serially connected to each other may be trained.
  • a primary neural network model that acquires primary diagnostic auxiliary information related to a target cardiovascular disease of a subject based on a fundus image of the subject, and secondary diagnostic auxiliary information related to a target cardiovascular disease based on the primary diagnostic auxiliary information
  • a score prediction model including a secondary neural network model for acquiring score information may be trained.
  • the primary diagnostic auxiliary information may be a probability that the subject corresponds to a coronary artery disease or a probability that the subject does not correspond to a coronary artery disease.
  • score information related to a target cardiovascular disease as secondary diagnosis auxiliary information may be a coronary artery calcium score. That is, the lower the probability of being normal for cardiovascular disease (the higher the probability of not being normal), the higher the cardiovascular-related score (cardiovascular risk), and based on this, a neural network model that predicts the degree of cardiovascular risk/calcium score, etc. is constructed and can be used
  • the primary neural network model may be trained based on primary learning data.
  • the primary neural network model may be trained based on primary learning data including a plurality of fundus images to which a label indicating a probability of whether a subject corresponds to a coronary artery disease.
  • the secondary neural network model may be trained based on secondary learning data including a plurality of probability information corresponding to a coronary artery disease in which a coronary calcium score is labeled.
  • the learned cardiovascular disease diagnosis auxiliary neural network model may be used to assist in the diagnosis of cardiovascular disease.
  • the learned cardiovascular disease diagnosis auxiliary neural network model it is possible to assist in the diagnosis of cardiovascular diseases by obtaining diagnostic auxiliary information useful for the diagnosis of cardiovascular diseases.
  • the above-described diagnostic apparatus, client apparatus, mobile apparatus, or server apparatus may obtain diagnostic assistance information based on the fundus image of the patient.
  • the diagnosis unit, control unit, or processor of each device may acquire auxiliary diagnosis information according to the target fundus image by using the cardiovascular disease diagnosis auxiliary neural network model.
  • the diagnosis unit 500 may include a diagnosis request obtaining module 501 , a cardiovascular disease diagnosis auxiliary module 503 , and a diagnosis auxiliary information output module 505 .
  • the diagnosis unit 500 and each module shown in FIG. 2 are illustratively described based on a logical configuration, and the diagnosis unit 500 or each module is at least one of various devices or other devices described in this specification. may be included in the device.
  • the diagnosis request obtaining module 501 may obtain a diagnosis request from an external device or a user.
  • the diagnosis request obtaining module 501 may obtain a diagnosis request including a fundus image to be diagnosed.
  • the diagnosis request obtaining module 501 may obtain a diagnosis request including diagnosis auxiliary information identification information for identifying the requested diagnosis auxiliary information.
  • the diagnosis request obtaining module 501 may obtain a diagnosis request and initiate a diagnosis auxiliary information obtaining process.
  • the diagnosis request obtaining module 501 may obtain a fundus image after obtaining a diagnosis request, obtain a diagnosis request including a fundus image, or obtain a diagnosis request after obtaining a fundus image.
  • the cardiovascular disease diagnosis auxiliary module 503 may obtain diagnosis auxiliary information by using the learned cardiovascular disease diagnosis auxiliary neural network model.
  • the cardiovascular disease diagnosis assistance module 503 may acquire diagnosis assistance information when a diagnosis assistance request is obtained.
  • the cardiovascular disease diagnosis auxiliary module 503 may acquire a target fundus image and acquire cardiovascular disease diagnosis auxiliary information from a neural network model based on the target fundus image.
  • the cardiovascular disease diagnosis auxiliary module 503 may acquire the learned neural network module or parameters of the learned neural network module, and use the acquired parameters to acquire diagnostic assistance information according to the target fundus image.
  • the cardiovascular disease diagnosis auxiliary module 503 may acquire a target fundus image and acquire disease presence information, grade information, or score information for diagnosing cardiovascular disease.
  • the cardiovascular disease diagnosis auxiliary module 503 may further acquire additional information (ie, secondary diagnosis auxiliary information) in addition to the primary cardiovascular disease diagnosis auxiliary information directly output from the neural network model.
  • additional information ie, secondary diagnosis auxiliary information
  • the cardiovascular disease diagnosis auxiliary module 503 may obtain instruction information or prescription information, which will be described later.
  • the cardiovascular disease diagnosis auxiliary module may also acquire a CAM (class activation map) image corresponding to the diagnostic auxiliary information or outputted diagnostic auxiliary information for a disease other than the target disease.
  • the diagnosis auxiliary information output module 505 may obtain the diagnosis auxiliary information from the cardiovascular disease diagnosis auxiliary module.
  • the diagnostic auxiliary information output module 505 may output diagnostic auxiliary information on cardiovascular disease.
  • the diagnostic auxiliary information obtaining module may transmit the diagnostic auxiliary information to an external device or an external module.
  • the diagnosis auxiliary information may be provided to the user through a user interface or the like.
  • the cardiovascular disease diagnosis assistance method includes the steps of obtaining a diagnosis request (S3010), obtaining diagnostic assistance information using a cardiovascular disease diagnosis auxiliary neural network model (S3020), and It may include outputting the diagnosis auxiliary information (S3030).
  • the step of obtaining the diagnosis auxiliary information using the cardiovascular disease diagnosis auxiliary neural network model may be configured differently depending on the type of the target diagnosis auxiliary information (ie, the diagnosis auxiliary information to be acquired).
  • the cardiovascular disease diagnosis auxiliary neural network model used to obtain the diagnosis auxiliary information may be determined according to the type of the target diagnosis auxiliary information.
  • Acquiring the auxiliary diagnosis information using the cardiovascular disease diagnosis auxiliary neural network model may include acquiring and processing the diagnosis auxiliary information obtained through the cardiovascular disease diagnosis auxiliary neural network model.
  • the obtaining of the auxiliary diagnosis information may further include obtaining the secondary diagnosis auxiliary information obtained based on the primary diagnosis auxiliary information and/or the primary diagnosis auxiliary information directly obtained through the neural network model.
  • the step of outputting the auxiliary diagnosis information may include outputting the auxiliary diagnosis information in a form recognizable by the user.
  • Outputting the diagnostic auxiliary information may include outputting the diagnostic auxiliary information in the form of visual or auditory data.
  • the outputting of the diagnostic auxiliary information may be configured differently depending on the type of the outputted diagnostic auxiliary information. For example, the diagnostic auxiliary information may be output differently according to the type.
  • a method of assisting the diagnosis of cardiovascular disease by using a cardiovascular disease diagnosis auxiliary neural network model trained to obtain information on the presence or absence of cardiovascular disease (or to select a risk group) based on a fundus image can be provided.
  • the method for assisting diagnosis of cardiovascular disease described below may be executed by the diagnostic unit, controller, or processor described herein.
  • the method for assisting diagnosis of cardiovascular disease includes the steps of obtaining a diagnosis request (S4010), and obtaining information about the presence or absence of disease by using a neural network model auxiliary for diagnosis of cardiovascular disease (S4020) and outputting diagnostic auxiliary information (S4030).
  • Acquiring the diagnosis request may include acquiring a diagnosis target fundus image (hereinafter, a target fundus image).
  • the obtaining of the diagnosis request may include obtaining information on the diagnosis auxiliary information to be obtained (diagnosis auxiliary information to be acquired).
  • the diagnosis request may be data requesting the start of the diagnosis auxiliary process.
  • the diagnosis request may include information on the target cardiovascular disease and may be data for requesting diagnosis auxiliary information related to the target cardiovascular disease.
  • the diagnosis request may be data requesting information on the presence or absence of a disease related to whether a patient corresponds to a risk group for a target disease.
  • the diagnosis request may include identification information of the target fundus image and the target diagnosis auxiliary information.
  • the diagnosis request obtaining module may obtain a diagnosis request (or diagnosis assistance request) requesting a diagnosis (or diagnosis assistance) for coronary artery disease based on the target fundus image and the target fundus image.
  • the diagnosis unit may obtain a diagnosis request from a user through the user input unit or may obtain a diagnosis request from an external device through the communication unit.
  • the diagnosis request acquiring module may acquire a diagnosis request for requesting diagnosis assistance information related to the presence or absence of a target disease according to the target fundus image and/or a need for administration related to the target disease.
  • the step of acquiring disease information using the cardiovascular disease diagnosis auxiliary neural network model may include acquiring disease presence information based on a diagnosis assistance request requesting disease presence information.
  • the acquiring of the disease information may include acquiring the disease information based on a diagnosis assistance request for a diagnosis assistance for a target cardiovascular disease.
  • the cardiovascular disease diagnosis auxiliary module may use a neural network model that acquires disease information indicating whether a patient has coronary artery disease based on a target fundus image to acquire disease presence information or disease presence information and other information .
  • the diagnosis unit may acquire information on the presence or absence of a target cardiovascular disease with respect to the fundus image from the neural network model through the control unit or the processor.
  • the cardiovascular disease diagnosis auxiliary module may further acquire information inferred based on the disease presence information on the target disease.
  • the cardiovascular disease diagnosis auxiliary module may further acquire a risk level for a disease other than the target cardiovascular disease, based on a predetermined correlation or considering input values other than the fundus image.
  • the cardiovascular disease diagnosis auxiliary module includes at least a portion of the primary neural network model for acquiring primary diagnostic auxiliary information (eg, the target cardiovascular disease corresponding probability of the subject) on the target cardiovascular disease and the primary diagnosis auxiliary information. Based on the secondary diagnosis auxiliary information (eg, the probability corresponding to the target cardiovascular disease risk group of the subject) is obtained, the cardiovascular disease diagnosis auxiliary neural network model including the secondary neural network model serially connected to the primary neural network model is used to obtain cardiovascular disease It is possible to obtain disease diagnosis auxiliary information.
  • primary diagnostic auxiliary information eg, the target cardiovascular disease corresponding probability of the subject
  • the step of outputting the diagnosis auxiliary information may further include outputting disease presence or absence information on the target cardiovascular disease.
  • the outputting of the diagnosis auxiliary information may further include outputting disease presence information and other information about the target cardiovascular disease together.
  • the outputting of the diagnosis auxiliary information may output the disease presence information and the inferred information together.
  • the diagnosis auxiliary information output module may output disease presence or absence information for a target cardiovascular disease.
  • the diagnosis auxiliary information output module may output disease presence information and other information about the target cardiovascular disease.
  • the diagnosis auxiliary information output module may transmit disease presence/absence information to an external device or may output disease presence/absence information in a form recognizable by a user.
  • the diagnosis unit may transmit the diagnosis auxiliary information including the disease presence information to the display unit or the output unit so that the disease information is provided to the user.
  • a method of assisting the diagnosis of cardiovascular disease by using a cardiovascular disease diagnosis auxiliary neural network model trained to obtain grade information indicating the risk of cardiovascular disease based on a fundus image.
  • the method for assisting the diagnosis of cardiovascular disease described below may be executed by a diagnosis unit, a controller, or a processor.
  • the method for assisting diagnosis of cardiovascular disease includes the steps of obtaining a diagnosis request (S5010), obtaining grade information using a cardiovascular disease diagnosis auxiliary neural network model (S5020), and It may include outputting the diagnostic auxiliary information (S5030).
  • Acquiring the diagnosis request may include acquiring a target fundus image.
  • the above-described information in relation to risk group selection may be similarly applied.
  • Acquiring the diagnosis request ( S5010 ) may include acquiring a diagnosis request including information about the target cardiovascular disease and the fundus image to be diagnosed.
  • the step of obtaining the diagnosis request ( S5010 ) may include obtaining a diagnosis request for requesting grade information related to the target cardiovascular disease of the patient.
  • Acquiring the diagnosis request ( S5010 ) may include obtaining a diagnosis request requesting grade information and other information.
  • the diagnosis request obtaining module may obtain a target fundus image and a diagnosis request for requesting diagnosis auxiliary information of a coronary artery disease according to the target fundus image.
  • the diagnosis request obtaining module may obtain a diagnosis request requesting prescription information related to cardiovascular disease according to the target fundus image.
  • the diagnosis unit may obtain a diagnosis request for the target fundus image through a user input unit, a user interface, or the like.
  • grade information on the target cardiovascular disease using the cardiovascular disease diagnosis auxiliary neural network model from the target fundus image may include obtaining Acquiring the grade information may include identifying a grade information request included in the diagnosis request, and acquiring grade information on a target cardiovascular disease in response to a diagnosis request requesting grade information.
  • the cardiovascular disease diagnosis auxiliary module may acquire grade information or grade information and other information by using a neural network model that acquires grade information indicating the risk of a patient's coronary artery disease based on the target fundus image.
  • the diagnosis unit may obtain grade information of a target cardiovascular disease with respect to the fundus image from the neural network model through the control unit or the processor.
  • the cardiovascular disease diagnosis auxiliary module may further acquire information inferred based on the grade information on the target disease. For example, the cardiovascular disease diagnosis auxiliary module may further acquire a risk level for a disease other than the target cardiovascular disease by considering an input value other than a predetermined correlation or fundus image.
  • the cardiovascular disease diagnosis auxiliary module may acquire grade information and prescription information for a target disease together. The prescription information may be obtained based on a matching table in which grades and prescriptions are matched and stored.
  • the cardiovascular disease diagnosis auxiliary module may acquire prescription information using a neural network model that acquires prescription information related to a patient's cardiovascular disease state based on the target fundus image.
  • the cardiovascular disease diagnosis auxiliary module uses a neural network model to obtain prescription information related to the need for a specific medical action, for example, a statin dosage prescription, based on a target fundus image (input fundus image) Prescribing information for essential parts can be obtained.
  • the cardiovascular disease diagnosis auxiliary module includes at least a portion of the primary neural network model for acquiring primary diagnostic auxiliary information (eg, the target cardiovascular disease corresponding probability of the subject) on the target cardiovascular disease and the primary diagnosis auxiliary information.
  • primary diagnostic auxiliary information eg, the target cardiovascular disease corresponding probability of the subject
  • secondary diagnostic auxiliary information eg, grade information related to the subject's target cardiovascular disease or prescription information indicating whether the subject should take statins
  • Cardiovascular disease diagnosis auxiliary information may be obtained using the included cardiovascular disease diagnosis auxiliary neural network model.
  • the step of outputting the diagnostic auxiliary information ( S5030 ) may further include outputting grade information on the target cardiovascular disease.
  • the step of outputting the diagnosis auxiliary information ( S5030 ) may further include outputting grade information and other information about the target cardiovascular disease together.
  • the outputting of the diagnosis auxiliary information may output the grade information and information inferred using the grade information together.
  • the diagnosis auxiliary information output module may output grade information on the target cardiovascular disease.
  • the diagnostic auxiliary information output module may output grade information and other information about the target cardiovascular disease.
  • the diagnostic auxiliary information output module may transmit grade information to an external device or output it in a form recognizable by a user.
  • the diagnosis unit may transmit the diagnosis auxiliary information including the rating information to the display unit or the output unit so that the rating information is provided to the user.
  • a method of assisting in the diagnosis of a cardiovascular disease by using a cardiovascular disease diagnosis auxiliary neural network model trained to obtain a score related to a cardiovascular disease based on a fundus image.
  • the method for assisting the diagnosis of cardiovascular disease described below may be performed by a diagnosis unit, a controller, or a processor.
  • the method for assisting diagnosis of cardiovascular disease includes the steps of obtaining a diagnosis request (S6010), obtaining score information using a cardiovascular disease diagnosis auxiliary neural network model (S6020), and It may include outputting the diagnostic auxiliary information (S6030).
  • Acquiring the diagnosis request may include acquiring a target fundus image.
  • the above-described information in relation to risk group selection may be similarly applied.
  • the step of obtaining a diagnosis request includes a diagnosis request including a fundus image to be diagnosed and information on a target cardiovascular disease, a diagnosis request requesting score information related to a target cardiovascular disease of the patient, or requesting score information and other information. obtaining a diagnostic request.
  • the diagnosis request acquiring module may acquire a target fundus image and a diagnosis request requesting a coronary calcium score according to the target fundus image.
  • score information on the target cardiovascular disease using the cardiovascular disease diagnosis auxiliary neural network model from the target fundus image may include obtaining Acquiring the score information may include acquiring a score information request included in the diagnosis request, and acquiring the requested score information.
  • the cardiovascular disease diagnosis auxiliary module may use the cardiovascular disease diagnosis auxiliary neural network model and obtain a coronary calcium score for determining the risk of coronary artery disease of a patient based on the target fundus image.
  • the diagnosis unit may obtain cardiovascular disease diagnosis auxiliary score information related to the target cardiovascular disease for the fundus image from the neural network model through the control unit or the processor.
  • the cardiovascular disease diagnosis auxiliary module may acquire additional diagnosis auxiliary information obtained by considering additional information related to score information or an input value other than the fundus image.
  • the cardiovascular disease diagnosis auxiliary module uses the cardiovascular disease diagnosis auxiliary neural network model, as a score related to cardiovascular disease according to the target fundus image, the coronary calcium score, the score is the arteriosclerosis risk score, the carotid intima Carotid Intima-Media Thickness (CIMT) value, Framingham coronary risk score, QRISK score, value according to ASCVD (Atherosclerotic Cardiovascular Disease) risk, European Systematic Coronary Risk Evaluation (SCORE), ASSIGN (Score from Scottish) At least one of a score according to the Intercollegiate Guidelines Network) score and a value of at least one factor for calculating any one of the above-listed scores may be obtained.
  • ASCVD Anatherosclerotic Cardiovascular Disease
  • SCORE European Systematic Coronary Risk Evaluation
  • ASSIGN Score from Scottish
  • the cardiovascular disease diagnosis auxiliary module may obtain a score value for determining the necessity of a predetermined medical prescription for the treatment of a target disease by using the cardiovascular disease diagnosis auxiliary neural network model.
  • the cardiovascular disease diagnosis auxiliary module may use the cardiovascular disease diagnosis auxiliary neural network model to obtain a score value (eg, an ASCVD risk score value) for determining the need for prescription of a statin agent for a subject.
  • the cardiovascular disease diagnosis auxiliary module may obtain the cardiovascular disease diagnosis auxiliary information by using a diagnostic value of the subject other than the fundus image as an input value.
  • the cardiovascular disease diagnosis auxiliary module uses, in addition to the fundus image, the cholesterol level, triglyceride level, low-density lipoprotein cholesterol level, high-density lipoprotein cholesterol level, and/or ultra-low-density lipoprotein cholesterol level, of the subject as input data together with the fundus image, It is possible to obtain disease diagnosis auxiliary information.
  • the cardiovascular disease diagnosis auxiliary module includes at least a portion of the primary neural network model for acquiring primary diagnostic auxiliary information (eg, the corresponding probability of coronary artery disease in the subject) on the target cardiovascular disease and the primary diagnosis auxiliary information.
  • primary diagnostic auxiliary information eg, the corresponding probability of coronary artery disease in the subject
  • secondary diagnostic auxiliary information eg, coronary calcium score of the subject
  • the step of outputting the diagnostic auxiliary information includes outputting score information and/or other information about the target cardiovascular disease (eg, diagnostic auxiliary information for other diseases or grade or disease presence information for the target cardiovascular disease). can do.
  • the diagnosis auxiliary information output module or the diagnosis unit may output score information and/or other information about the target cardiovascular disease.
  • the cardiovascular disease diagnosis auxiliary information obtained through the diagnosis auxiliary neural network model may be output.
  • Diagnosis assistance information for cardiovascular disease may be provided to the user.
  • the client device, the mobile device, the diagnostic device, the output unit, etc. described herein may output the cardiovascular disease diagnosis auxiliary information in a form recognizable by a user.
  • the cardiovascular disease diagnosis auxiliary information may be output in the form of visual and/or auditory data.
  • the cardiovascular disease diagnosis auxiliary information may be output through the user interface.
  • the cardiovascular disease diagnosis auxiliary information may be output to an external device.
  • the diagnosis device, the server device, the diagnosis unit, etc. may transmit the acquired cardiovascular disease diagnosis auxiliary information to an external device through wired or wireless communication.
  • the cardiovascular disease diagnosis auxiliary information may be stored in a memory or a server.
  • Secondary information obtained from the diagnostic auxiliary information obtained through the neural network model may be output.
  • the secondary information obtained from the diagnosis assistance information may be information for more specifically supporting a diagnosis of cardiovascular disease or the like.
  • the secondary information obtained from the diagnosis auxiliary information may be implemented as prescription information, instruction information, prediction information, etc. described below.
  • the diagnosis auxiliary information described herein may be understood to include primary diagnosis auxiliary information obtained through a neural network model and/or secondary diagnosis auxiliary information obtained from primary diagnosis auxiliary information.
  • prescription information may be obtained.
  • prescription information including a type of a drug to be administered to a user, an administration time, and a dosage of the drug may be included.
  • the prescribing information may include prescribing information for an antihyperlipidemic agent.
  • the prescription information may include drug information that may be prescribed to a subject in relation to a target cardiovascular disease, such as a hyperlipidemia drug, a high blood pressure drug, and an antithrombotic drug.
  • prescribing information includes drugs of the HMG-CoA reductase inhibitor statin (including various agents such as simvastatin, atorvastatin, and rosuvastatin), bile acid sequestrant, and nicotinic acid. ), etc., may include dosing information regarding the summary of administration/dosage amount/administration time for the subject of the preparation.
  • the prescription information may be stored in advance to match the diagnosis auxiliary information.
  • the prescription information may be determined using a database in which a user's prescription action according to the diagnosis auxiliary information is stored.
  • prescription information related to statin administration may be obtained using a database that matches the risk grade for hyperlipidemia and other dyslipidemia and the need for statin administration according to each grade.
  • prescription information when the diagnosis auxiliary information is score information for determining the risk of cardiovascular disease, prescription information may be acquired as secondary information on the score information.
  • the diagnostic auxiliary information is an ASCVD risk or SCORE score for the determination of dyslipidemia
  • the first prescription information indicating that the subject's need for taking a statin when the obtained score is less than or equal to the reference value, the first prescription information indicating that the subject's need for taking a statin is low,
  • second prescription information indicating that the subject's need for taking a statin is significant when the obtained score is equal to or less than the reference value.
  • the acquired diagnostic information is coronary artery calcification scores (CACs)
  • CACs coronary artery calcification scores
  • a reference value eg, 100
  • Prescribing information recommending that the subject take the statin agent is obtained, and when the coronary artery calcification score is less than the reference value, prescribing information for withholding the statin agent may be obtained.
  • Prescribing information may be obtained through a neural network model learned using learning data including information on a user's prescription behavior according to the diagnosis auxiliary information.
  • the prescription data input from the user is obtained in response to the output of the diagnosis auxiliary information, the prescription information learning data in which the prescription data label is attached to the diagnosis auxiliary information is obtained, and the diagnosis auxiliary information is obtained using the obtained prescription information learning data.
  • the prescription information obtained using the prescription assistance neural network model may be provided to the user together with or separately from the diagnosis assistance information.
  • the diagnostic apparatus may obtain prescription data for a prescribed medicament (eg, statin) provided by the user in response to acquiring diagnostic assistance information (eg, grade information or score information) through the diagnostic assistance information acquisition module, Learning data including an input fundus image labeled with prescription data may be acquired.
  • a prescribed medicament eg, statin
  • diagnostic assistance information eg, grade information or score information
  • Instruction information obtained based on the diagnosis auxiliary information may be output.
  • the indication information may include information about a medical treatment method. For example, based on the diagnostic assistance information, instructional information for providing the user with at least one candidate action expected to be appropriate for the patient may be obtained.
  • the instruction information may instruct additionally required examination, next visit time, suggestion of a hospital to be transferred, and measures such as recommended surgery or treatment.
  • the indication information may be stored in advance to match the diagnostic auxiliary information.
  • the instruction information may be determined using a database in which the instruction action of the user according to the diagnosis auxiliary information is stored.
  • the instruction information may include management guideline information related to a target cardiovascular disease, such as lifestyle and exercise prescription recommended to the subject.
  • the instruction information may include additional examination information indicating a recommended type of detailed examination.
  • additional examination information indicating a recommended type of detailed examination.
  • the diagnostic device is CT scan of coronary artery (or Ankle-Brachial Index), vascular stiffness test (pulse wave velocity analysis, Pulse Wave Velocity), 24 hours Holter Monitoring ), etc.) may be acquired and/or outputted.
  • the instruction information may be obtained through a neural network model learned using learning data including instruction action information according to the diagnosis auxiliary information.
  • the instruction data input from the user is obtained, the instruction information learning data to which the instruction data label according to the diagnosis auxiliary information is given is obtained, and the obtained instruction information database is used.
  • the instructional auxiliary neural network model trained to output instructional information by inputting the diagnostic auxiliary information may be provided.
  • the instructional information obtained using the instructional assistance neural network model may be provided to the user together with or separately from the diagnostic assistance information.
  • Prediction information obtained based on the diagnosis auxiliary information may be output.
  • the prediction information may include information on a prognosis related to a target cardiovascular disease of the subject.
  • the prediction information may include death probability information indicating a death probability within 5 years or a death probability within 10 years in relation to a target cardiovascular disease of the subject.
  • prediction information and indication information or prescription information may be provided together.
  • the secondary information may include specific instruction information and prescription information, and prediction information when a subsequent procedure indicated by the corresponding information is performed.
  • the secondary information may include the first predictive information including the probability of death of the subject when the subject does not take the drug, and the recommended amount of the drug administered according to the prescribing information determined according to the obtained cardiovascular disease diagnosis auxiliary information
  • the second prediction information including the death probability of the subject may be included.
  • the secondary information may include predictive information on the probability of death or reduction in the probability of death when the subject complies with a guideline according to instruction information determined according to the acquired cardiovascular disease diagnosis auxiliary information.
  • the diagnosis auxiliary information may include a CAM related to the output diagnosis auxiliary information.
  • a class activation map may be obtained from the neural network model together with or as primary diagnostic auxiliary information.
  • a visualized image of the CAM may be output.
  • the CAM may be provided to the user through the user interface described above.
  • CAM may be provided according to user selection.
  • the CAM image may be provided along with the fundus image.
  • the CAM image may be provided overlaid with the fundus image.
  • the diagnosis assistance system for assisting the diagnosis of cardiovascular disease based on the fundus image includes a fundus image acquisition unit for acquiring a target fundus image, a preprocessing unit for processing the target fundus image so that blood vessels are emphasized, and based on the preprocessed image and an output unit for outputting the auxiliary information for diagnosing cardiovascular disease and obtaining auxiliary information for diagnosing cardiovascular disease, the auxiliary diagnosis unit acquires the CAM related to the auxiliary unit for diagnosing cardiovascular disease, and the output unit uses the obtained CAM The output can be overlaid on the target fundus image.
  • the first image corresponding to the captured image (ie, the original image) and the first image are reconstructed so that the target element (eg, blood vessel) included in the first image is emphasized.
  • to acquire a third image that is a CAM image obtained through an auxiliary neural network model for cardiovascular disease diagnosis based on the obtained second image (eg, a blood vessel-emphasized image or a blood vessel extraction image), and superimpose the first image and the third image may include displaying.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Physics & Mathematics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Cardiology (AREA)
  • Physiology (AREA)
  • Eye Examination Apparatus (AREA)

Abstract

L'invention concerne un procédé et un dispositif d'aide au diagnostic d'une maladie cardiovasculaire. Le procédé d'aide au diagnostic d'une maladie cardiovasculaire d'un sujet à l'aide d'une image de fond d'œil selon un mode de réalisation peut comprendre les étapes consistant à : obtenir une image de fond d'œil du sujet qui est une image capturée du fond d'œil d'un sujet ; obtenir des informations d'aide au diagnostic d'une maladie cardiovasculaire du sujet conformément à l'image du fond de l'œil du sujet, au moyen d'un modèle de réseau neuronal d'aide au diagnostic de maladie cardiovasculaire obtenant des informations d'aide au diagnostic destinées à être utilisées dans le diagnostic de la maladie cardiovasculaire du sujet conformément à l'image de fond d'œil, sur la base de l'image du fond de l'œil du sujet ; et délivrer en sortie les informations d'aide au diagnostic de maladie cardiovasculaire du sujet.
PCT/KR2020/019452 2020-12-30 2020-12-30 Procédé et dispositif d'aide au diagnostic d'une maladie cardiovasculaire WO2022145542A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/KR2020/019452 WO2022145542A1 (fr) 2020-12-30 2020-12-30 Procédé et dispositif d'aide au diagnostic d'une maladie cardiovasculaire

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/KR2020/019452 WO2022145542A1 (fr) 2020-12-30 2020-12-30 Procédé et dispositif d'aide au diagnostic d'une maladie cardiovasculaire

Publications (1)

Publication Number Publication Date
WO2022145542A1 true WO2022145542A1 (fr) 2022-07-07

Family

ID=82259465

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2020/019452 WO2022145542A1 (fr) 2020-12-30 2020-12-30 Procédé et dispositif d'aide au diagnostic d'une maladie cardiovasculaire

Country Status (1)

Country Link
WO (1) WO2022145542A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116433644A (zh) * 2023-04-22 2023-07-14 深圳市江机实业有限公司 一种基于识别模型的眼部图像动态诊断方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017031099A1 (fr) * 2015-08-20 2017-02-23 Ohio University Dispositifs et procédés de classification de la dégénérescence diabétique et maculaire
WO2018035473A2 (fr) * 2016-08-18 2018-02-22 Google Llc Traitement d'images du fond de l'œil à l'aide de modèles d'apprentissage machine
KR102058883B1 (ko) * 2019-04-11 2019-12-24 주식회사 홍복 당뇨병 및 전조 증상을 진단하기 위해 홍채 영상 및 망막 영상을 인공지능으로 분석하는 방법
KR20200005412A (ko) * 2018-07-06 2020-01-15 연세대학교 산학협력단 심혈관 질병 진단 보조 방법 및 장치
KR20200011033A (ko) * 2017-10-27 2020-01-31 메디비콘 아이엔씨. 신장 기능 측정 방법

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017031099A1 (fr) * 2015-08-20 2017-02-23 Ohio University Dispositifs et procédés de classification de la dégénérescence diabétique et maculaire
WO2018035473A2 (fr) * 2016-08-18 2018-02-22 Google Llc Traitement d'images du fond de l'œil à l'aide de modèles d'apprentissage machine
KR20200011033A (ko) * 2017-10-27 2020-01-31 메디비콘 아이엔씨. 신장 기능 측정 방법
KR20200005412A (ko) * 2018-07-06 2020-01-15 연세대학교 산학협력단 심혈관 질병 진단 보조 방법 및 장치
KR102058883B1 (ko) * 2019-04-11 2019-12-24 주식회사 홍복 당뇨병 및 전조 증상을 진단하기 위해 홍채 영상 및 망막 영상을 인공지능으로 분석하는 방법

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116433644A (zh) * 2023-04-22 2023-07-14 深圳市江机实业有限公司 一种基于识别模型的眼部图像动态诊断方法
CN116433644B (zh) * 2023-04-22 2024-03-08 深圳市江机实业有限公司 一种基于识别模型的眼部图像动态诊断方法

Similar Documents

Publication Publication Date Title
Maehara et al. Definitions and methodology for the grayscale and radiofrequency intravascular ultrasound and coronary angiographic analyses
Madder et al. Detection by near-infrared spectroscopy of large lipid core plaques at culprit sites in patients with acute ST-segment elevation myocardial infarction
CN102959579B (zh) 医疗信息显示装置、操作方法及程序
Tell et al. Relationship between balance and abnormalities in cerebral magnetic resonance imaging in older adults
JP4157573B2 (ja) 問診票データ作成装置
WO2022145542A1 (fr) Procédé et dispositif d'aide au diagnostic d'une maladie cardiovasculaire
Kaul et al. In search of the vulnerable plaque: is there any light at the end of the catheter?
US20180130555A1 (en) Systems and methods for intelligent admissions
JP2010017519A (ja) 神経精神病の診断を生成するためのコンピュータ実装方法、コンピュータ・プログラム製品、装置及びシステム
WO2019132165A1 (fr) Procédé et programme de fourniture de rétroaction sur un résultat chirurgical
Louis et al. Problems and controversies in tremor classification
Wang et al. Value of medical history in ophthalmology: A study of diagnostic accuracy
Okuno et al. Valvular and nonvalvular atrial fibrillation in patients undergoing transcatheter aortic valve replacement
Egbe et al. Potential benefits of ambulatory blood pressure monitoring in coarctation of aorta
WO2011136491A2 (fr) Système basé sur un système de terminologie pour le support d'une définition d'objet de données
WO2022010168A1 (fr) Procédé de prédiction du risque de démence chez un patient atteint de la maladie de parkinson et dispositif de prédiction de risque de démence
Tahri Sqalli et al. Eye tracking technology in medical practice: a perspective on its diverse applications
Glass et al. Comparison of two fall-risk assessment tools used in a long-term care facility
Hougham et al. Empirical research on informed consent with the cognitively impaired
WO2020022825A1 (fr) Procédé et dispositif électronique de détection d'état de santé assistée par intelligence artificielle (ai) dans un réseau de l'internet des objets
WO2023080697A1 (fr) Procédé de division de signal cardiaque et dispositif de division de signal cardiaque utilisant ledit procédé de division de signal cardiaque
WO2019132166A1 (fr) Procédé et programme d'affichage d'image d'assistant chirurgical
WO2022145541A1 (fr) Procédé et dispositif d'aide au diagnostic de maladie rénale
Cao et al. Relations between white coat effect of blood pressure and arterial stiffness
Narsavage et al. Factors associated with referral of elderly individuals with cardiac and pulmonary disorders for home care services following hospital discharge

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20968088

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20968088

Country of ref document: EP

Kind code of ref document: A1