Method of providing user interface for prediction of diseases and device of providing user interface for prediction of diseases using the same

Abstract

The present invention provides a method for providing a user interface for disease prediction implemented by a processor, comprising the steps of: receiving at least one clinical data among heart sound data, electrocardiogram data, and chest X-ray data for an individual; classifying the type of clinical data based on a classification algorithm configured to classify the type based on the clinical data, and performing a pre-processing predetermined according to the type of clinical data on the classified clinical data; predicting the risk of developing a cardiovascular disease for an individual based on the pre-processed clinical data using at least one of a plurality of models each configured to predict a cardiovascular disease based on at least one of the heart sound data, electrocardiogram data, chest X-ray data, and clinical data; and providing the risk of developing a cardiovascular disease for an individual.

Description

Technology Field

[0001] The present invention relates to a method for providing a user interface for disease prediction and an apparatus for providing a user interface for disease prediction using the same. More specifically, the invention relates to a method for providing a user interface and an apparatus for providing a user interface for disease prediction using the same, configured to determine whether additional examination is necessary by considering various types of clinical data regarding an individual suspected of having a cardiovascular disease. Background Technology

[0002] Cardiovascular disease refers to conditions affecting the heart and major arteries, and has been reported as the leading cause of disease worldwide. These cardiovascular diseases are caused by disorders of the heart and blood vessels and may specifically include coronary heart disease (heart attack), coronary artery disease, elevated blood pressure (hypertension), peripheral artery disease, rheumatic heart disease, congenital heart disease, and heart failure. More specifically, coronary artery disease usually occurs when the coronary arteries supplying blood to the heart become blocked or narrowed due to atherosclerosis. Additionally, myocardial infarction can develop when a coronary artery is completely blocked by atherosclerosis, leading to the death of heart muscle tissue, while angina pectoris is a condition caused by the narrowing of coronary arteries that can induce chest tightness or chest pain.

[0003] Meanwhile, identifying individuals at risk of developing cardiovascular disease can be an important strategy for more effectively preventing or treating the disease. Traditionally, cardiovascular disease could only be diagnosed using physical methods after it had progressed to a certain extent, which limited early diagnosis or prediction. While diagnosis is typically performed by imaging the heart and coronary arteries with X-rays and ultrasound using contrast equipment, diagnostic results can vary depending on the proficiency of the medical staff. Furthermore, although several risk factors measurable from blood samples to predict cardiovascular disease—such as those related to LDL and HDL cholesterol levels—are currently used clinically, many patients with atherosclerosis do not possess these risk factors. Moreover, cardiovascular disease can occur in many individuals who do not exhibit these risk factors.

[0004] Therefore, there is an ongoing demand for methods to detect the risk of cardiovascular disease development in patients.

[0005] The background description of the invention is provided to facilitate a better understanding of the present invention. The matters described in the background description should not be construed as an acknowledgment that they exist as prior art. The problem to be solved

[0006] Meanwhile, the inventors of the present invention noted that changes in clinical data would precede the onset of cardiovascular disease clinically as part of the human body's physiological response.

[0007] More specifically, the inventors of the present invention noted that the characteristics of clinical data obtainable from an individual may be associated with the risk of developing cardiovascular disease.

[0008] Consequently, the inventors of the present invention recognized that features extracted from electrocardiogram data, heart sound data, and chest X-ray data can be used to predict the risk of cardiovascular disease.

[0009] As a result, the inventors of the present invention have developed a user interface providing system that provides an assessment of the risk of developing cardiovascular disease based on the characteristics of electrocardiogram data, heart sound data, and chest X-ray data obtained from an individual.

[0010] Meanwhile, the inventors of the present invention were able to apply a prediction model trained to predict the risk of developing cardiovascular disease based on electrocardiogram data, heart sound data, and chest X-ray data to a user interface providing system.

[0011] In particular, the inventors of the present invention recognized that the clinical data obtainable from each individual may differ. Accordingly, the inventors of the present invention intended to build a learning model for each case by considering cases where all three types of clinical data are obtainable, cases where two of the clinical data are obtainable, or cases where only a single type of clinical data is obtainable.

[0012] Furthermore, the inventors of the present invention intended to further apply a classification model configured to classify the type of data obtained from an entity, thereby enabling appropriate preprocessing for each data.

[0013] As a result, the inventors of the present invention were able to provide a user interface that not only predicts and provides the risk of developing cardiovascular disease based on clinical data when a user loads it, but also distinguishes and outputs the type of any clinical data when it is selected.

[0014] As a result, the user interface providing system based on prediction and classification models was able to provide accurate and reliable prediction information regarding objects.

[0015] The inventors of the present invention anticipated that, through this user interface providing system, medical staff or patients could more quickly determine whether additional examinations are necessary as the individual's cardiovascular disease risk is provided. Furthermore, the inventors anticipated that such a system could advance the timing of treatment, thereby improving treatment outcomes, increasing patient survival rates, and reducing treatment costs.

[0016] Accordingly, the problem that the present invention aims to solve is to provide a method and apparatus for providing a user interface for disease prediction, configured to classify clinical data regarding an object received through a classification model, perform predetermined preprocessing on the classified clinical data, and predict and provide the risk of developing cardiovascular disease using a prediction model.

[0017] The problems of the present invention are not limited to those mentioned above, and other unmentioned problems will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0018] To solve the problem described above, a method for providing a user interface for disease prediction according to an embodiment of the present invention is provided. The information provision method according to an embodiment of the present invention comprises the steps of: receiving at least one clinical data among heart sound data, electrocardiogram data, and chest X-ray data for an individual, performed by a processor; classifying at least one type of clinical data using a classification model based on a classification algorithm configured to classify the types of clinical data; performing a pre-processing predetermined according to the type of clinical data on the classified clinical data; predicting the risk of developing cardiovascular disease for an individual based on the pre-processed clinical data using at least one of a plurality of models each configured to predict cardiovascular disease based on at least one combination of heart sound data, electrocardiogram data, chest X-ray data, and clinical data; and providing the risk of developing cardiovascular disease for an individual.

[0019] To solve the problem described above, a device for providing a user interface for disease prediction according to another embodiment of the present invention is provided. An information providing device according to one embodiment of the present invention includes a receiving unit configured to receive at least one clinical data among heart sound data, electrocardiogram data, and chest X-ray data for an individual, a processor, and an output unit. At this time, the processor classifies at least one clinical data using a classification model based on a classification algorithm configured to classify the types of clinical data, performs pre-processing predetermined according to the type of clinical data on the classified clinical data, and predicts the risk of developing cardiovascular disease for the individual based on the pre-processed clinical data using at least one of a plurality of models each configured to predict cardiovascular disease based on at least one combination of heart sound data, electrocardiogram data, chest X-ray data, and clinical data. Furthermore, the output unit is configured to provide the risk of developing cardiovascular disease for the individual.

[0020] Specific details of other embodiments are included in the detailed description and drawings. Effects of the invention

[0021] The present invention can contribute to the accurate diagnosis of cardiovascular disease by providing a plurality of predictive models configured to predict cardiovascular disease based on various clinical data.

[0022] In particular, the present invention can provide a user interface that predicts the risk of developing cardiovascular disease with higher accuracy by classifying each type of data and applying a preprocessing procedure for feature extraction and ensemble.

[0023] More specifically, the present invention provides a user interface providing system configured to apply a plurality of prediction models learned according to the type of clinical data and configured to probabilistically provide the risk of developing cardiovascular disease, thereby providing clinical information that can contribute to the determination of the necessity of additional clinical examinations.

[0024] Accordingly, the present invention provides the risk of cardiovascular disease development of an individual, thereby enabling medical staff or patients to determine more quickly whether additional examinations are necessary.

[0025] Furthermore, the present invention can advance the timing of treatment to improve treatment outcomes, increase patient survival rates, and reduce treatment costs.

[0026] The effects according to the present invention are not limited to those exemplified above, and a wider variety of effects are included within the present invention. Brief explanation of the drawing

[0027] FIG. 1a is a schematic diagram illustrating a system for providing a user interface for disease prediction using biosignal data according to one embodiment of the present invention. FIG. 1b is a schematic diagram illustrating a device for providing a user interface for disease prediction according to one embodiment of the present invention. FIGS. 2a and 2b are schematic flowcharts illustrating a method for providing a user interface for disease prediction according to an embodiment of the present invention. FIGS. 3a to 3c illustrate, in an exemplary manner, the output section of a device for providing a user interface for disease prediction according to one embodiment of the present invention. Specific details for implementing the invention

[0028] The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the scope of the claims.

[0029] Although terms such as "first," "second," etc., are used to describe various components, it goes without saying that these components are not limited by these terms. These terms are used merely to distinguish one component from another. Therefore, it goes without saying that the first component mentioned below may also be the second component within the technical scope of the present invention.

[0030] Throughout the specification, the same reference numerals refer to the same components.

[0031] The features of each of the various embodiments of the present invention may be combined or combined with one another, either partially or wholly, and as will be fully understood by those skilled in the art, various technical interlocking and operation are possible, and each embodiment may be implemented independently of one another or together in an interlocking relationship.

[0032] Hereinafter, with reference to FIGS. 1a and 1b, an apparatus for providing a user interface for disease prediction according to various embodiments of the present invention will be described in detail.

[0033] FIG. 1a is a schematic diagram illustrating a system for providing a user interface for disease prediction using biosignal data according to an embodiment of the present invention. FIG. 1b is a schematic diagram illustrating a device for providing a user interface for disease prediction according to an embodiment of the present invention.

[0034] First, referring to FIG. 1a, a user interface providing system (1000) for disease prediction is configured to provide clinical information, such as the risk of developing cardiovascular disease, based on clinical data (300) for an individual. At this time, the clinical data (300) applied to the user interface providing system (1000) for disease prediction may be at least one of heart sound data (310), electrocardiogram data (320), and chest X-ray data (330). A device (100) for providing a user interface according to an embodiment of the present invention is configured to output the type of acquired clinical data and to output the risk of developing cardiovascular disease predicted by these data.

[0035] More specifically, referring together with FIG. 1b, a device (100) for providing a user interface for disease prediction according to one embodiment of the present invention may include a receiving unit (110), an input unit (120), an output unit (130), a storage unit (140), and a processor (150).

[0036] At this time, the receiver (110) may be configured to receive at least one clinical data among heart sound data (310), electrocardiogram data (320), and chest X-ray data (330). At this time, the receiver may be further configured to be connected to communicate with an electrocardiogram or heart sound measurement device, and further to a medical imaging device, so as to receive the corresponding data directly from each device.

[0037] The input unit (120) is not limited to a keyboard, mouse, touch screen panel, etc. The input unit (120) may be configured to set up a device (100) for providing a user interface for disease prediction and to direct its operation. More specifically, the input unit (120) may be further configured to receive a selection of clinical data from a user.

[0038] The output unit (130) can display clinical data received by the receiver unit (110). Furthermore, the display unit (130) can display information regarding cardiovascular disease generated by the processor (150).

[0039] The storage unit (140) may be configured to store various clinical data received by the receiving unit (110) and user selections entered through the input unit (120). Furthermore, the storage unit (140) may be further configured to store information related to cardiovascular disease classified or determined by the processor (150) to be described later. However, not limited thereto, the storage unit (140) may be configured to store all data generated during the process in which the processor (150) classifies data based on new clinical data and predicts the risk of developing cardiovascular disease.

[0040] The processor (150) may be based on a classification model based on a classification algorithm configured to classify types of clinical data, and a plurality of prediction models each trained to predict the risk of developing cardiovascular disease according to a combination of data.

[0041] More specifically, the processor (150) may be configured to classify at least one clinical data using a classification model based on a classification algorithm configured to classify types of clinical data, and to perform pre-processing on the classified clinical data according to the type of clinical data. Furthermore, the processor (150) is configured to predict the risk of developing cardiovascular disease for an individual based on the pre-processed clinical data using at least one of a plurality of models each configured to predict cardiovascular disease based on at least one of heart sound data, electrocardiogram data, chest X-ray data, and clinical data.

[0042] Meanwhile, the classification model within the processor (150) can be used as a classification server based on a classification algorithm. Accordingly, within this specification, the classification model can be used interchangeably with the classification server.

[0043] Accordingly, the device (100) for providing a user interface for disease prediction according to one embodiment of the present invention can determine information related to cardiovascular disease with high accuracy.

[0044] Accordingly, the user can receive clinical information that can contribute to the determination of the need for additional clinical tests through the user interface providing system (1000) for disease prediction.

[0045] Hereinafter, with reference to FIGS. 2a and 2b, a method for providing a user interface for disease prediction according to various embodiments of the present invention

[0046] FIGS. 2a and 2b are schematic flowcharts illustrating a method for providing a user interface for disease prediction according to an embodiment of the present invention.

[0047] Referring to FIG. 2a, first, clinical data is received according to a method for providing a user interface for disease prediction according to an embodiment of the present invention (S210). Next, the types of clinical data are classified by a classification model (S220), and pre-processing is performed according to the types of data (S230). Next, the risk of developing cardiovascular disease of an individual is predicted using a prediction model pre-trained according to the clinical data (S240), and a final result is provided (S250).

[0048] More specifically, in the step (S210) where clinical data is received, at least one clinical data among heart sound data, electrocardiogram data, and chest X-ray data may be received. In this case, the heart sound data may be at least one of AA (aortic arch), PA (pulmonary artery), TA (transverse arch), and MA. Furthermore, the electrocardiogram data may have a PDF file format. However, the types and formats of these clinical data are not limited to those described above.

[0049] According to a feature of the present invention, in the step (S210) where clinical data is received, clinical data that has undergone preprocessing, such as size adjustment or dimension change, may be received.

[0050] Next, in the step (S220) where the type of clinical data is classified, the type of acquired clinical data can be classified.

[0051] For example, referring together with FIG. 2b, in the step (S220) where the type of clinical data is classified, clinical data (300) for a received object may be input into a classification model (220) based on an algorithm configured to classify the type of clinical data. Then, electrocardiogram data (310), heart sound data (320), and chest X-ray data (330) classified by the classification model (220) may be obtained.

[0052] However, the classification of clinical data types may be performed by a variety of methods. For example, in the step (S220) where the types of clinical data are classified, if a user uploads clinical data to a classification server via HTTP POST, the types of clinical data may be distinguished and output by a classification algorithm.

[0053] Next, in the step (S230) where a predetermined preprocessing is performed, features for each data are extracted and then standardized, and if there are multiple classified data, preprocessing can be performed in which they are ensembled and integrated.

[0054] For example, referring again to FIG. 2b, in the step (S230) where a predetermined preprocessing is performed, preprocessing may be performed for each of the electrocardiogram data (310), heart sound data (320), and chest X-ray data (330). More specifically, in the step (S230) where a predetermined preprocessing is performed, for the electrocardiogram data (310), signal information may be extracted and standardization may be performed. Furthermore, for the heart sound data (320), feature information may be extracted and standardization may be performed. Also, for the chest X-ray data, image information may be extracted and standardization may be performed. Furthermore, in the step (S230) where a predetermined preprocessing is performed, the standardized electrocardiogram data (312), the preprocessed heart sound data (322), and the preprocessed chest X-ray data (332) may be ensembled and integrated into a single data. At this time, the ensemble data (232) in which these data are integrated may be in the form of a numeric array, but is not limited thereto.

[0055] Next, in the step (S240) where the risk of developing cardiovascular disease is predicted, the probability of developing cardiovascular disease can be predicted by a pre-trained prediction model based on the type or combination of clinical data.

[0056] For example, referring again to FIG. 2b, in the step (S240) where the risk of developing cardiovascular disease is predicted, since the ensemble data (232) is derived from electrocardiogram data (310), heart sound data (320), and chest X-ray data (330), it may be selectively input into a disease prediction model (242) based on electrocardiogram, heart sound, and chest X-ray data among a plurality of disease prediction models (240). At this time, the plurality of disease prediction models (240) may include a model configured to predict the risk of developing disease using single data from a disease prediction model based on electrocardiogram data, a disease prediction model based on heart sound data, and a disease prediction model based on chest X-ray data. Furthermore, a plurality of disease prediction models (240) may include a disease prediction model based on electrocardiogram and heart sound data, a disease prediction model based on electrocardiogram and chest X-ray data, a disease prediction model based on heart sound and chest X-ray data, and a prediction model based on a combination of two or more data of a disease prediction model based on electrocardiogram, heart sound and chest X-ray data (242). As a result of the step (S240) in which the risk of developing a cardiovascular disease is predicted, an output value (252) may be output from the disease prediction model (242) based on electrocardiogram, heart sound and chest X-ray data. At this time, the output value (252) is the probability of developing a cardiovascular disease such as myocardial infarction, and may be a value between 0 and 1. However, the output form is not limited thereto.

[0057] Finally, in the step (S250) where the final result is provided, the output value according to the optional input to the prediction model can be provided to the user.

[0058] For example, referring again to Fig. 2b, in the step (S250) where the final result is provided, the probability of myocardial infarction, '0.8', is provided through the output unit, and the user can check the risk of cardiovascular disease development for the subject.

[0059] Meanwhile, in the step (S250) where the final result is provided, the output information by the prediction models is converted into an HTTP response and JSON format, and then the JSON response may be parsed by visualization software and output through the output unit.

[0060] By the method for providing a user interface for disease prediction according to various embodiments of the present invention, the user can more easily determine whether to perform additional examinations on an individual.

[0061] Next, with reference to FIGS. 3a to 3c, a user interface provided by a method for providing a user interface for disease prediction according to various embodiments of the present invention will be described exemplarily.

[0062] FIGS. 3a to 3c illustrate, in an exemplary manner, the output section of a device for providing a user interface for disease prediction according to one embodiment of the present invention.

[0063] First, referring to FIG. 3a, a device (100) for providing a user interface for disease prediction according to one embodiment of the present invention may have an input section (120) and an output section (130) that are identical as a touch screen panel. At this time, the output section (130) provides an object information section (132), a chest X-ray data display section (133), an electrocardiogram data display section (134), a heart sound data display section (135), and a cardiovascular disease incidence probability display section (137).

[0064] More specifically, the individual information section (132) may provide information such as the patient's gender, age, date of birth, name, and patient ID. Furthermore, the chest X-ray data display section (133) may display and provide the individual's chest X-ray data among the clinical data uploaded to the classification server by the user. Additionally, the electrocardiogram data display section (134) may display and provide the individual's electrocardiogram data among the clinical data uploaded to the classification server by the user, and the heart sound data display section (135) may display and provide the individual's heart sound data among the clinical data uploaded to the classification server by the user. At this time, the heart sound data displayed in the heart sound data display section (135) may be AA, PA, MA, and TA, and when the user selects the play button, each heart sound data (320) may be output.

[0065] Meanwhile, the user may also select clinical data to be used to predict the risk of developing cardiovascular disease through a selection button in the chest X-ray data display area (133), electrocardiogram data display area (134), and heart sound data display area (135).

[0066] For example, with reference to FIG. 3b, if the user excludes these data through the selection button in the chest X-ray data display area (133) and the selection button in the heart sound data display area (135), the risk of cardiovascular disease development for the individual can be calculated using only the electrocardiogram data provided in the electrocardiogram data display area (134). Furthermore, with reference to FIG. 3c, if excluded through the selection button in the electrocardiogram data display area (134), the risk of cardiovascular disease development for the individual can be calculated using the chest X-ray data provided in the chest X-ray data display area (133) and the heart sound data provided in the heart sound data display area (135).

[0067] Furthermore, when a user drags any clinical data, the clinical data may be moved to a chest X-ray data display area (133), an electrocardiogram data display area (134), or a heart sound data display area (135) depending on the type of data. This allows the user to check what the data is.

[0068] Meanwhile, the cardiovascular disease occurrence probability display (137) can provide the probability of cardiovascular disease occurrence output based on data selected by the user. That is, the user can check the risk of disease occurrence for the patient through the cardiovascular disease occurrence probability display (137).

[0069] Accordingly, users can more easily check the risk of developing cardiovascular disease in the individual and decide whether to conduct additional examinations.

[0070] As shown in the various embodiments above, the present invention can provide clinical information that contributes to the determination of the necessity of additional clinical examinations by providing a user interface providing system based on a plurality of prediction models configured to probabilistically provide the risk of developing cardiovascular disease by learning according to the type of clinical data.

[0071] Accordingly, the present invention provides the risk of cardiovascular disease development of an individual, thereby enabling medical staff or patients to determine more quickly whether additional examinations are necessary.

[0072] Furthermore, the present invention can advance the timing of treatment to improve treatment outcomes, increase patient survival rates, and reduce treatment costs.

[0073] Although embodiments of the present invention have been described in more detail with reference to the attached drawings, the present invention is not necessarily limited to these embodiments and may be modified in various ways within the scope of the technical spirit of the present invention. Accordingly, the embodiments disclosed in the present invention are intended to explain, not limit, the technical spirit of the present invention, and the scope of the technical spirit of the present invention is not limited by these embodiments. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of protection of the present invention shall be interpreted by the claims below, and all technical spirits within an equivalent scope shall be interpreted as being included within the scope of rights of the present invention. Explanation of the symbols

[0074] 100: Device for providing a user interface for disease prediction 110: Receiver 120: Input section 130: Output section 132: Object Information 133: Chest X-ray Data Display Section 134: ECG Data Display Area 135: Heart sound data display section 137: Cardiovascular disease incidence probability indicator 140: Storage section 150: Processor 220: Classification Model 232: Ensemble Data 240: Disease Prediction Model 242: Disease prediction model based on electrocardiogram, heart sounds, and chest X-ray data 252: Output value 300: Clinical Data 310: ECG data 312: Preprocessed ECG data 320: Heart sound data 322: Preprocessed heart sound data 330: Chest X-ray data 332: Preprocessed chest X-ray data 1000: User Interface Provision System for Disease Prediction

Claims

Claim 1 A method for providing a user interface for disease prediction implemented by a processor, comprising: receiving at least one clinical data among heart sound data, electrocardiogram data, and chest X-ray data for an entity; classifying the type of the at least one clinical data using a classification model based on a classification algorithm configured to classify the type of clinical data; performing a preprocessing predetermined according to the type of clinical data, wherein if the classified clinical data is electrocardiogram data, signal information is extracted and standardization is performed; if the classified clinical data is heart sound data, feature information is extracted and standardization is performed; and if the classified clinical data is chest X-ray data, image information is extracted and standardization is performed; integrating the preprocessed electrocardiogram data, the preprocessed heart sound data, and the preprocessed chest X-ray data into ensemble data; and as a model configured to predict the risk of disease onset, a disease prediction model based on electrocardiogram and heart sound data, a disease prediction model based on electrocardiogram and chest X-ray data, and heart sound and chest X-ray data A method for providing a user interface for disease prediction, comprising the steps of: predicting the risk of developing cardiovascular disease for said individual based on clinical data on which the preprocessing has been performed, using a plurality of disease prediction models, including a disease prediction model based on a plurality of disease prediction models, and a prediction model based on a combination of two or more data of a disease prediction model based on electrocardiogram, heart sound, and chest X-ray data; and providing the risk of developing cardiovascular disease for said individual. Claim 2 A receiving unit configured to receive at least one clinical data among heart sound data, electrocardiogram data, and chest X-ray data for an entity; a processor; and an output unit, wherein the processor classifies the type of at least one clinical data using a classification model based on a classification algorithm configured to classify the type of clinical data, and for the classified clinical data, performs preprocessing in which signal information is extracted and standardization is performed if the classified clinical data is electrocardiogram data, performs preprocessing in which feature information is extracted and standardization is performed if the classified clinical data is heart sound data, and performs preprocessing in which image information is extracted and standardization is performed if the classified clinical data is chest X-ray data, and performs preprocessing predetermined according to the type of clinical data, and integrates the preprocessed electrocardiogram data, the preprocessed heart sound data, and the preprocessed chest X-ray data into ensemble data, and a model configured to predict the risk of disease onset, comprising a disease prediction model based on electrocardiogram and heart sound data, a disease prediction model based on electrocardiogram and chest X-ray data, and a disease prediction model based on heart sound and chest X-ray data. A device for providing a user interface for disease prediction, configured to predict the risk of developing cardiovascular disease for an individual based on clinical data on which the preprocessing has been performed, using a plurality of disease prediction models including a prediction model based on a combination of two or more data of disease prediction models based on electrocardiogram, heart sound, and chest X-ray data, and an output unit configured to provide the risk of developing cardiovascular disease for the individual.

Images