System for providing artificial intelligence-based analysis result for biometric data of patient and method thereof

Abstract

The present disclosure provides a system for acquiring biometric data and providing an artificial intelligence-based analysis result. The system according to an embodiment of the present disclosure includes: an artificial intelligence (AI) system that acquires, by using an AI-based model, AI analysis result data by analyzing electrocardiogram data of a patient, and processes the AI analysis result data on the basis of a protocol of an electronic medical record unit, and transmits the processed AI analysis result data to the electronic medical record unit; and the electronic medical record unit that manages a medical record of the patient including the electrocardiogram data and the AI analysis result data, wherein the electronic medical record unit integrates the processed AI analysis result data received from the AI system with examination result data of other biometric data of the patient and provides the integrated result data.

Description

System and method for providing artificial intelligence-based analysis results of a patient's biometric data

[0001] The present disclosure relates to deep learning technology in the medical field, and more particularly, to a system and method for analyzing a patient's biometric data and providing analysis results based on artificial intelligence.

[0002] Electrocardiogram (ECG) data is a recording of electrocardiogram signals generated by microcurrents within the heart. Accurate analysis of ECG data is essential in the medical field, as it can lead to early detection of cardiovascular disease or structural abnormalities in patients. Currently, most ECG analysis relies on the human eye. Specifically, physicians observe ECG data, analyze its waveform, or identify characteristic information to diagnose patients. However, this method has inherent limitations due to its reliance on the human eye. It is difficult to recognize all the subtle features or patterns inherent in ECG data, which can lead to overlooking important diagnostic information. Furthermore, the fact that interpretation results can vary depending on the physician's skill level also contributes to poor accuracy.

[0003] Meanwhile, recent advancements in information and communication technology have led to the application of artificial intelligence (AI) in various fields. In particular, various attempts are being made to integrate AI into the medical field, which previously relied on specialized and limited resources such as doctors and researchers to diagnose patients' illnesses, to improve efficiency. Recently, AI has been actively utilized to analyze electrocardiogram (ECG) data to predict potential heart disease in patients.

[0004] However, there is a problem with smooth system integration between companies providing services based on AI technology and hospitals using them. Electronic Medical Record (EMR) systems used within hospitals follow different data formats and management systems for each hospital, making it difficult to integrate existing medical test data (e.g., blood tests, imaging tests, etc.) with new AI-based analysis results. Figure 1 is an exemplary diagram that provides a method for providing AI analysis results for existing electrocardiogram (ECG) data within an electronic medical record environment. Referring to Figure 1, in a conventional EMR system, a UI (User Interface) containing multiple medical test result items listed for a patient is displayed on the display, and medical staff can select a specific item to view and receive the corresponding test result information on the same UI. However, ECG analysis results based on AI technology provided by companies outside the hospital may have a different data format from the test results of the existing EMR system, and therefore are not directly integrated into the UI (600) provided by the hospital EMR, but are provided through a separate pop-up window (610). This delivery method can make it difficult for medical professionals to intuitively compare existing medical examination data with AI-based analysis results, and can cause workflow disruptions.

[0005] The present disclosure is conceived in response to the aforementioned background technology, and aims to provide a system and method for providing artificial intelligence-based analysis results for a patient's biometric data.

[0006] However, the problems to be solved in this disclosure are not limited to the problems mentioned above, and other problems not mentioned can be clearly understood based on the description below.

[0007] A system for providing an artificial intelligence-based analysis result according to an embodiment of the present disclosure for realizing the above-described task includes an artificial intelligence system that analyzes electrocardiogram data of a patient using an artificial intelligence-based model to obtain AI (Artificial Intelligence) analysis result data, processes the AI ​​analysis result data based on a protocol of an electronic medical record unit, and transmits the AI ​​analysis result data to the electronic medical record unit, and an electronic medical record unit that manages the medical record of the patient including the electrocardiogram data and the AI ​​analysis result data, and the electronic medical record unit integrates the processed AI analysis result data received from the artificial intelligence system with the test result data of other biometric data of the patient and provides the result.

[0008] Alternatively, the artificial intelligence system may input the electrocardiogram data into the artificial intelligence-based model to obtain a score corresponding to the likelihood of a disease, compare the score with a reference value to determine the disease grade of the determined patient, and obtain the score and the disease grade of the patient as the AI ​​analysis result data.

[0009] Alternatively, the artificial intelligence system may identify the structural format of a test result UI (User Interface) in which test result data of the patient's other biometric data is displayed, and process the AI ​​analysis result data based on the protocol of the electronic medical record unit so that the AI ​​analysis result data corresponds to the structural format of the test result UI.

[0010] Alternatively, the other biometric data is blood data, and the artificial intelligence system can integrate the patient's blood test result data and the processed AI analysis result data and transmit them to the electronic medical record unit so that the AI ​​analysis result is included on the patient's blood test result UI.

[0011] Alternatively, the patient's blood test result UI includes a list composed of blood test results according to a plurality of blood test items, and when the electronic medical record unit receives the processed AI analysis result data from the artificial intelligence system, it adds the AI ​​analysis result field to the list, and when receiving a request for viewing the patient's blood test result, it can display the blood test result UI including the AI ​​analysis result.

[0012] Alternatively, the AI ​​analysis result may be in XML format, and the protocol of the electronic medical record unit may include the HL7 protocol and the FHIR protocol.

[0013] Alternatively, the system further includes a receiver unit for acquiring and storing biometric data, and the artificial intelligence system, when a storage event of electrocardiogram data in the receiver unit is detected, acquires the stored electrocardiogram data from the receiver unit, and analyzes the stored electrocardiogram data using the artificial intelligence-based model to obtain an analysis result.

[0014] Alternatively, when a storage event of electrocardiogram data in the receiver unit is detected, the artificial intelligence system can compare the stored electrocardiogram data with electrocardiogram data for the patient previously stored in the receiver unit to determine whether the stored electrocardiogram data is new electrocardiogram data.

[0015] Alternatively, when the artificial intelligence system receives a request for AI analysis results from the electronic medical record unit, it can obtain the electrocardiogram data from the receiver unit and analyze the electrocardiogram data using the artificial intelligence-based model to obtain the AI ​​analysis results.

[0016] The artificial intelligence system of the present disclosure converts and processes AI analysis result data into the same format as existing test result data managed by the EMR system, thereby enabling medical staff to intuitively compare and utilize existing test data and AI analysis results without a separate UI or additional data conversion work or system construction.

[0017] Figure 1 is an exemplary diagram providing a method for providing AI analysis results for existing electrocardiogram data within an electronic medical record environment.

[0018] FIG. 2 is a diagram illustrating an electronic medical record environment according to one embodiment of the present disclosure.

[0019] FIG. 3 is a flowchart illustrating a method for providing an artificial intelligence-based analysis result of an artificial intelligence system according to one embodiment of the present disclosure.

[0020] FIG. 4 is an exemplary diagram providing an artificial intelligence-based analysis result of an artificial intelligence system according to one embodiment of the present disclosure.

[0021] Below, embodiments of the present disclosure are described in detail with reference to the attached drawings so that those skilled in the art can easily implement the present disclosure. The embodiments presented in this disclosure are provided to enable those skilled in the art to utilize or implement the contents of the present disclosure. Accordingly, various modifications to the embodiments of the present disclosure will be apparent to those skilled in the art. That is, the present disclosure may be implemented in various different forms and is not limited to the embodiments described below.

[0022] Throughout the specification of this disclosure, identical or similar drawing numbers refer to identical or similar components. Furthermore, for the purpose of clearly describing the disclosure, drawing numbers for parts in the drawings that are not relevant to the description of the disclosure may be omitted.

[0023] The term "or" as used herein is intended to mean an inclusive "or" rather than an exclusive "or." That is, unless otherwise specified herein or clear from context, "X employs A or B" should be understood to mean either of its natural inclusive permutations. For example, unless otherwise specified herein or clear from context, "X employs A or B" can be interpreted to mean either X employs A, X employs B, or X employs both A and B.

[0024] The term "and / or" as used herein should be understood to refer to and include all possible combinations of one or more of the related concepts listed.

[0025] The terms "comprises" and / or "comprising" as used herein should be understood to mean the presence of certain features and / or components. However, it should be understood that the terms "comprises" and / or "comprising" do not exclude the presence or addition of one or more other features, other components, and / or combinations thereof.

[0026] Unless otherwise specified in this disclosure or unless the context makes it clear that the singular form is being referred to, the singular should generally be construed to include “one or more.”

[0027] The term "Nth (N is a natural number)" used in this disclosure can be understood as an expression used to mutually distinguish components of this disclosure based on a predetermined standard such as a functional perspective, a structural perspective, or convenience of explanation. For example, components performing different functional roles in this disclosure can be distinguished as a first component or a second component. However, components that are substantially the same within the technical spirit of this disclosure but must be distinguished for convenience of explanation may also be distinguished as a first component or a second component.

[0028] The term "acquisition" as used in this disclosure may be understood to mean not only receiving data through a wired or wireless communication network with an external device or system, but also generating data in an on-device form.

[0029] Meanwhile, the term "module" or "unit" used in the present disclosure can be understood as a term referring to an independent functional unit that processes computing resources, such as a computer-related entity, firmware, software or a part thereof, hardware or a part thereof, or a combination of software and hardware. At this time, the "module" or "unit" may be a unit composed of a single element, or a unit expressed as a combination or set of multiple elements. For example, as a narrow concept, a "module" or "unit" may refer to a hardware element of a computing device or a set thereof, an application program that performs a specific function of software, a processing process implemented through software execution, or a set of instructions for program execution, etc. In addition, as a broad concept, a "module" or "unit" may refer to the computing device itself that constitutes the system, or an application running on the computing device, etc. However, since the above-described concept is only an example, the concept of “module” or “part” may be defined in various ways within a range understandable to those skilled in the art based on the contents of the present disclosure.

[0030] The term "model" as used herein may be understood as a system implemented using mathematical concepts and language to solve a specific problem, a set of software units to solve a specific problem, or an abstract model of a processing process to solve a specific problem. For example, a neural network "model" may refer to the entire system implemented as a neural network that has problem-solving capabilities through learning. In this case, the neural network can have problem-solving capabilities by optimizing the parameters connecting nodes or neurons through learning. A neural network "model" may include a single neural network or a set of neural networks that are a combination of multiple neural networks.

[0031] The term "data" used in this disclosure may include "image," "signal," and the like. The term "image" used in this disclosure may refer to multidimensional data composed of discrete image elements. In other words, "image" may be understood as a term referring to a digital representation of an object visible to the human eye. For example, "image" may refer to multidimensional data composed of elements corresponding to pixels in a two-dimensional image. "Image" may refer to multidimensional data composed of elements corresponding to voxels in a three-dimensional image.

[0032] The explanation of the above terms is intended to aid understanding of the present disclosure. Therefore, unless explicitly stated as limiting the contents of the present disclosure, it should be noted that the above terms are not intended to limit the technical ideas of the present disclosure.

[0033] FIG. 2 is a diagram illustrating an electronic medical record environment (1000) according to one embodiment of the present disclosure. However, FIG. 2 is merely an example, and the electronic medical record environment (1000) may include other components for implementing an environment for recording, managing, and providing a patient's electronic medical information. Furthermore, only some of the disclosed components may be included in the electronic medical record environment (1000).

[0034] Referring to FIG. 2, an electronic medical record environment (1000) (Electronic Medical Record System, EMR) is an environment (1000) (or system) for electronically recording and managing medical information related to patient treatment in medical facilities such as hospitals and public health centers, and may also be referred to as a medical system. The electronic medical record environment (1000) may include an artificial intelligence system (100), a receiver unit (200), a middleware unit (300), and an electronic medical record unit (400).

[0035] FIG. 3 is a diagram illustrating a method for generating an artificial intelligence-based reading result sheet in an electronic medical record environment (1000) according to one embodiment of the present disclosure.

[0036] An artificial intelligence system (100) according to one embodiment of the present disclosure acquires biometric data, analyzes the acquired biometric data (10) based on artificial intelligence (AI), and transmits the analysis result based on artificial intelligence (hereinafter, AI analysis result) to an electronic medical record unit (400).

[0037] In particular, the artificial intelligence system (100) converts or processes the format of the data of the artificial intelligence-based analysis results of biometric data acquired from the receiver unit (200) to match the format of the data of the analysis results of other biometric data managed by the electronic medical record unit (400), and transmits it to the electronic medical record unit (400), thereby enabling analysis results of different biometric data to be identified simultaneously in the electronic medical record environment (1000) (or in an electronic device connected to the electronic medical record unit (400).

[0038] That is, the artificial intelligence system (100) not only analyzes biometric data using artificial intelligence, but also performs the function of processing and converting the analysis results so that they can be integrated or made compatible with the analysis results for other biometric data managed by the existing electronic medical record unit (400).

[0039] A receiver unit (200) according to one embodiment of the present disclosure acquires and stores a patient's biometric data (10). The receiver unit (200) may be referred to as a receiver device and may be implemented as at least one electronic device (e.g., a computer, a server, etc.).

[0040] The biometric data (10) acquired by the receiver unit (200) may include biometric signal information of the patient (e.g., waveform of the biometric signal, characteristic information identified from the waveform of the biometric signal), patient information, measurement environment (1000) (setting information of the measurement device, measurement sampling rate information, etc.), and metadata (measurement time, measurement location, data format information, etc.). For example, if the biometric data (10) is electrocardiogram data acquired from an electrocardiogram measurement device, the electrocardiogram data may include electrocardiogram waveform information, electrode location and lead information, and patient information.

[0041] The receiver unit (200) is connected to an external bio-signal measuring device that measures a patient's bio-signal, and can obtain bio-data (10) obtained by the bio-signal measuring device from the bio-signal measuring device. The receiver unit (200) can be connected to at least one bio-signal measuring device by wire or wirelessly. For example, the receiver unit (200) can obtain bio-data (10) of the patient from the bio-signal measuring device that performs network communication with the receiver unit (200) through a communication interface of the receiver unit (200). In addition, the receiver unit (200) can store bio-data (10) obtained from the bio-signal measuring device. For example, the receiver unit (200) can store bio-data (10) in the memory of the receiver unit (200) whenever bio-data (10) is obtained.

[0042] The receiver unit (200) can be connected to various bio-signal measuring devices, which can include an electrocardiogram measuring device, an electroencephalogram measuring device, an brain wave signal measuring device, a blood pressure measuring device, an oxygen saturation measuring device, a blood sugar measuring device, a body temperature measuring device, etc. To this end, the receiver unit (200) can include a plurality of receiver units (200) each connected to a different bio-signal measuring device.

[0043] In addition, the receiver unit (200) may include a bio-signal measuring device so that the receiver unit (200) may directly acquire bio-data. For example, the receiver unit (200) may include at least one electrode, thereby directly acquiring electrocardiogram data for a patient. Hereinafter, for the convenience of explanation of the present disclosure, the bio-signal measuring device will be assumed to be an electrocardiogram measuring device, and thus the bio-data acquired by the receiver unit (200) will be described as electrocardiogram data.

[0044] The receiver unit (200) can convert the acquired electrocardiogram data (10) into image data corresponding to the electrocardiogram data (10). Specifically, the receiver unit (200) can extract electrocardiogram signal information and features of the electrocardiogram data (e.g., patient identification information, measurement time, etc.) from the acquired electrocardiogram data (10), and convert the extracted information into image data including a waveform of the electrocardiogram signal and text representing the features of the electrocardiogram signal. For example, the electrocardiogram data (10) can be in an XML file format. The image data can be an electrocardiogram signal interpretation result sheet.

[0045] Meanwhile, the receiver unit (200) can transmit image data to a middleware unit (300) that connects the receiver unit (200) and the electronic medical record unit (400). Alternatively, the receiver unit (200) can directly transmit the image data to the electronic medical record unit (400).

[0046] A middleware unit (300) according to one embodiment of the present disclosure connects a receiver unit (200) and an electronic medical record unit (400). The middleware unit (300) performs a function of routing and transmitting data between the receiver unit (200) and the electronic medical record unit (400). For example, the middleware unit (300) may determine whether a prescription code issued by a medical professional corresponds to data (e.g., an electrocardiogram signal interpretation result sheet) transmitted from the receiver unit (200), and transmit data corresponding to the prescription code to the electronic medical record unit (400). The prescription code may be transmitted from the electronic medical record unit (400). Accordingly, the middleware unit (300) may select an electronic device that has transmitted a prescription code among various electronic devices included in the electronic medical record unit (400), and transmit the data. The middleware unit (300) may be implemented as an electronic device (e.g., a computer, a server, etc.) or as software (e.g., a program, an application, etc.). The prescription code may be a code composed of letters, numbers, symbols, etc. that includes patient identification information (e.g., patient ID, etc.), treatment department information, treatment type information, surgery information, etc.

[0047] In addition, the middleware unit (300) can relay connections to various electronic devices (e.g., bio-signal measuring devices, etc.) and systems (e.g., PACS (Picture Archiving and Communication System), Laboratory Information System (LIS), etc.) within a medical institution connected to the electronic medical record unit (400), and can convert or integrate the format of transmitted medical data. In addition, the middleware unit (300) can perform encryption on medical data or restrict access by external electronic devices (or external systems).

[0048] An electronic medical record unit (400) according to one embodiment of the present disclosure records and manages a patient's medical information. Specifically, the electronic medical record unit (400) stores the patient's medical records, prescription details, test results, etc. entered by medical staff. Data related thereto can be received from the middleware unit (300). In addition, the electronic medical record unit (400) transmits prescription codes, medical information, etc. entered by medical staff to the middleware unit (300).

[0049] The electronic medical record unit (400) may be implemented as a system including an electronic device (e.g., a computer, etc.) including a processor, a storage unit (e.g., a database, etc.), a display device, etc. The electronic medical record unit (400) may display information on the analysis results of various biometric data about the patient on a display. Here, the display may be a component of an electronic device included in the electronic medical record unit equipped with a display used by medical staff to manage the patient. Specifically, the electronic medical record unit (400) may display medical data (e.g., image data corresponding to electrocardiogram data (10) acquired by the receiver unit (200)) transmitted from the middleware unit (300) to the receiver unit (200) on the display. In particular, the electronic medical record unit (400) may display AI analysis result data transmitted from the artificial intelligence system (100) on the display. Meanwhile, the electronic medical record unit (400) may be referred to as an electronic medical record system.

[0050] Below, the operation of an artificial intelligence system (100) according to one embodiment of the present disclosure will be described in detail.

[0051] FIG. 3 is a flowchart illustrating a method for generating an artificial intelligence-based reading result sheet of an artificial intelligence system (100) according to one embodiment of the present disclosure.

[0052] According to one embodiment of the present disclosure, an artificial intelligence system (100) analyzes a patient's electrocardiogram data using an artificial intelligence-based model to obtain AI analysis result data, processes the AI ​​analysis result data based on the protocol of the electronic medical record unit, and transmits the AI ​​analysis result data to the electronic medical record unit (400). The artificial intelligence system (100) may be a hardware device or a part of a hardware device that performs comprehensive processing and calculation of data, or may be a software-based computing environment connected to a communication interface. For example, the artificial intelligence system (100) may be a server that performs intensive data processing functions and shares resources, or may be a client that shares resources through interaction with a server. In addition, the artificial intelligence system (100) may be a cloud system in which a plurality of servers and clients interact to comprehensively process data. Since the above description is only one example related to the type of the artificial intelligence system (100), the type of the artificial intelligence system (100) may be configured in various ways within a range that can be understood by those skilled in the art based on the contents of the present disclosure. For example, the artificial intelligence system (100) can be implemented in various electronic devices such as server devices, desktops, laptops, and smartphones.

[0053] Meanwhile, according to one embodiment of the present disclosure, the artificial intelligence system (100) may include an artificial intelligence server (110) that acquires AI analysis results and a processing server (120) that processes AI analysis result data. However, the present invention is not limited thereto, and may include other configurations for implementing an electronic medical record environment (1000). This is an example of a functionally separate configuration, and the artificial intelligence system (100) may be implemented in a form in which it operates integrally in a single device or in a distributed manner across multiple devices. In addition, the present invention may include other configurations for implementing an electronic medical record environment (1000).

[0054] Referring to FIG. 3, an artificial intelligence server (110) according to an embodiment of the present disclosure can analyze electrocardiogram data (10) using an artificial intelligence model (30) to obtain an AI analysis result (hereinafter, AI analysis result) (S310).

[0055] Specifically, when the receiver unit (200) obtains and stores electrocardiogram data (10) from a biosignal measuring device, the artificial intelligence server (110) can obtain the stored electrocardiogram data (10) from the receiver unit (200). Then, the artificial intelligence server (110) can input the electrocardiogram data (10) into an artificial intelligence model (30) to obtain an analysis result of the electrocardiogram data (10). For example, the artificial intelligence server (110) can extract data related to a biosignal (specifically, data related to a biosignal waveform or data related to a feature identified in a biosignal waveform) from electrocardiogram data (10) in XML format, and input the extracted information into the artificial intelligence model (30) to obtain an analysis result.

[0056] As a more specific example, when the biosignal is an electrocardiogram, the artificial intelligence server (110) can extract data related to the electrocardiogram waveform from the electrocardiogram data and / or data related to features identified in the electrocardiogram waveform, such as P waves (and Q waves, QRS intervals, etc.), and input the extracted and acquired data into the artificial intelligence model (30), thereby obtaining analysis results regarding the electrocardiogram waveform.

[0057] When the artificial intelligence server (110) obtains analysis results regarding electrocardiogram data (10) through the artificial intelligence model (30), it can modify the electrocardiogram data (10) to include the obtained analysis results. That is, if the electrocardiogram data (10) obtained by the receiver unit (200) from the biosignal measuring device is original data, the artificial intelligence unit can modify the original data by including the analysis results in the original data.

[0058] Meanwhile, the artificial intelligence model (30) may be a neural network model trained in advance to analyze biosignals and output analysis results. For example, the artificial intelligence model (30) may be a neural network model, and the neural network model may be implemented as a multi-layer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network, etc. Meanwhile, the artificial intelligence model (30) may output the likelihood of a disease predicted from the biosignal as an analysis result. Accordingly, the artificial intelligence server (110) may modify the electrocardiogram data (10) by including information on the likelihood of a disease.

[0059] According to one embodiment of the present disclosure, an artificial intelligence server (110) may input electrocardiogram data (10) into an artificial intelligence model (30) to obtain a disease score related to a bio-signal corresponding to the electrocardiogram data (10), and modify the electrocardiogram data (10) to include information regarding the disease score. In other words, the neural network model may output a disease probability predicted from the bio-signal in the form of a score.

[0060] For example, if the disease is left ventricular systolic dysfunction, the artificial intelligence server (110) may obtain learning data including input data composed of multiple electrocardiogram data obtained from different patients and label data in which labels indicating whether each patient has left ventricular systolic dysfunction are assigned, and may train a neural network model in advance using the obtained learning data. The learning data may include feature information (e.g., P wave, QRS complex, T wave, etc. included in the electrocardiogram signal) extracted from the electrocardiogram signal (more specifically, the electrocardiogram signal corresponding to the electrocardiogram data).

[0061] The artificial intelligence server (110) can train a neural network model to determine whether a patient has left ventricular systolic dysfunction based on such feature information. The artificial intelligence server (110) can input training data into the neural network model and calculate a loss function based on the difference between the output value of the neural network model and the label data during the training process. The loss function can be defined as a cross entropy loss or an objective function that optimizes the balance between precision and recall. Based on the calculated loss function, the artificial intelligence server (110) can adjust the weights of the model (30) through backpropagation. By repeating this process, the artificial intelligence server (110) can improve the classification performance of the neural network model regarding left ventricular systolic dysfunction, and ultimately, can obtain a trained neural network model to determine whether a patient has left ventricular systolic dysfunction based on electrocardiogram data.

[0062] When the artificial intelligence server (110) obtains a disease score from the artificial intelligence model (30), it can modify the electrocardiogram data (10) to include the obtained disease score. In particular, the artificial intelligence server (110) inputs the electrocardiogram data into the artificial intelligence-based model (30) to obtain a score corresponding to the possibility of a disease, compares the score with a reference value to determine the disease grade of the determined patient, and may obtain the score and the patient's disease grade as AI analysis result data. For example, if the reference value related to left ventricular systolic dysfunction is 10, and the score obtained by the artificial intelligence server (110) by inputting it into the artificial intelligence model (30) is 20, the artificial intelligence server (110) can confirm that the score is higher than the reference value and determine the patient's status with respect to left ventricular systolic dysfunction to be high risk. At this time, the artificial intelligence server (110) can obtain grade information representing the patient's status together with the score as an analysis result, and modify the electrocardiogram data to include the analysis result.

[0063] In this regard, the artificial intelligence server (110) can extract data related to bio-signals from the electrocardiogram data (10), input the extracted data into the artificial intelligence model (30) to obtain a disease score related to the electrocardiogram data (10), and insert data related to the disease score into the electrocardiogram data (10) to modify the electrocardiogram data (10). For example, the artificial intelligence server (110) can extract data related to bio-signals from the electrocardiogram data (10) in XML format. Specifically, the artificial intelligence server (110) can identify a tag related to the bio-signal and extract data related to the bio-signal as a sub-element of the identified tag. In addition, the artificial intelligence server (110) can input the extracted data into the artificial intelligence model (30) to obtain a disease score that quantifies the possibility of a disease related to the electrocardiogram data (10). In addition, the artificial intelligence server (110) can modify the electrocardiogram data (10) to include disease score information and information on the patient's condition grade regarding the disease by inserting new tags regarding the disease score and information on the patient's condition grade regarding the disease score. In addition, the artificial intelligence server (110) can also include patient identification information, measurement time of the electrocardiogram data, etc. in the electrocardiogram data (10). Alternatively, the artificial intelligence server (110) can transmit only data regarding the AI ​​analysis result (i.e., AI analysis result data) to the processing server (120).

[0064] According to one embodiment of the present disclosure, the artificial intelligence server (110) may include a plurality of neural network models according to the type of biosignal corresponding to the electrocardiogram data (10). At this time, the artificial intelligence server (110) may identify the type of biosignal of the acquired electrocardiogram data (10), select a neural network model corresponding to the type of the identified biosignal from among the plurality of neural network models, and obtain an analysis result. For example, the artificial intelligence server (110) may select only the electrocardiogram data (10) related to the electrocardiogram signal from among the plurality of electrocardiogram data (10) stored in the receiver unit (200), and analyze the selected electrocardiogram data (10) using the artificial intelligence model to obtain an analysis result related to the electrocardiogram signal.

[0065] In addition, according to one embodiment of the present disclosure, the artificial intelligence server (110) may include multiple neural network models according to disease type for the same biosignal. At this time, the artificial intelligence server (110) may identify a set disease type or a disease type requested from the electronic medical record unit (400), select a neural network model corresponding to the type of the identified disease signal from among the multiple neural network models, and obtain an analysis result. Meanwhile, the artificial intelligence server (110) may obtain multiple disease scores corresponding to multiple diseases related to the electrocardiogram data (10) using the multiple neural network models, and may modify the electrocardiogram data (10) to include the multiple disease scores. At this time, the artificial intelligence server (110) may select a neural network model corresponding to the type of biosignal and the disease type according to a request and prescription code from a medical professional transmitted through the electronic medical record unit (400).

[0066] Meanwhile, according to one embodiment of the present disclosure, when the artificial intelligence server (110) receives an artificial intelligence result request from the electronic medical record unit (400), it can obtain electrocardiogram data (10) from the receiver unit (200), and analyze the electrocardiogram data (10) using the artificial intelligence model (30) to obtain an analysis result. Specifically, the artificial intelligence server (110) can receive an artificial intelligence result request from the electronic medical record unit (400). In particular, the artificial intelligence server (110) can receive an artificial intelligence result request of the electronic medical record unit (400) from the middleware unit (300). In addition, when the artificial intelligence server (110) receives an artificial intelligence result request from the electronic medical record unit (400), it can obtain electrocardiogram data (10) stored in the receiver unit (200) and obtain an analysis result regarding the obtained electrocardiogram data (10).

[0067] According to one embodiment of the present disclosure, when a storage event of electrocardiogram data in the receiver unit is detected, the artificial intelligence server (110) may obtain the stored electrocardiogram data from the receiver unit and analyze the stored electrocardiogram data using an artificial intelligence-based model (30) to obtain an analysis result. Specifically, when it is confirmed that new biometric data is stored in the receiver unit (200), the artificial intelligence server (110) may obtain new biometric data from the receiver unit (200) and analyze the new biometric data using the artificial intelligence model (30) to obtain an analysis result. In this regard, the artificial intelligence server (110) may monitor the backup folder of the receiver unit (200) and detect whether new biometric data is stored in the backup folder of the receiver unit (200), i.e., a storage event of new electrocardiogram data. In addition, when an event in which new biometric data is stored is detected, the artificial intelligence server (110) may obtain new biometric data from the backup folder of the receiver unit (200). And, by using the artificial intelligence model (30)(30), analysis results regarding new biometric data can be obtained.

[0068] In particular, when an event of storing electrocardiogram data in the receiver unit is detected, the artificial intelligence server (110) can compare the electrocardiogram data for the patient previously stored in the receiver unit with the stored electrocardiogram data to determine whether the stored electrocardiogram data is new electrocardiogram data. Specifically, when electrocardiogram data (10) is stored in the receiver unit (200), the artificial intelligence server (110) can compare the newly stored electrocardiogram data (10) with the existing biometric data previously stored in the receiver unit (200) to determine whether the stored electrocardiogram data (10) is new biometric data. When the artificial intelligence server (110) monitors the backup folder of the receiver unit (200), and an event of storing new biometric data in the receiver unit (200) is detected, the artificial intelligence server (110) can distinguish, based on the patient, whether the newly stored electrocardiogram data (10) is newly measured and acquired electrocardiogram data (10) or whether existing biometric data has been modified and copied and stored. To this end, the artificial intelligence server (110) can determine whether the stored electrocardiogram data (10) is new biometric data by comparing newly stored electrocardiogram data (10) with existing biometric data for the same patient to determine similarity or by comparing metadata to determine the measurement time.

[0069] Meanwhile, the receiver unit (200) may receive a request for analysis of electrocardiogram data (10) from the electronic medical record unit (400). Here, the request for analysis of electrocardiogram data (10) may be a request for transmission of electrocardiogram data (10) that does not include an artificial intelligence result. That is, when the receiver unit (200) is requested to transmit electrocardiogram data (10) that does not include an artificial intelligence result, the receiver unit (200) may add a flag for non-request for AI analysis result to the electrocardiogram data (10) and store it in a backup folder, or store the electrocardiogram data in a backup folder other than the backup folder monitored by the artificial intelligence server (110), and convert the electrocardiogram data into image data and transmit it to the middleware unit (300). In addition, the middleware unit (300) may transmit the image data to the electronic medical record unit (400).

[0070] Referring to FIG. 3, according to one embodiment of the present disclosure, the processing server (120) may obtain AI analysis result data from the artificial intelligence server (110), process the AI ​​analysis result data based on the protocol of the electronic medical record unit (400), and transmit the AI ​​analysis result data to the electronic medical record unit (400) (S320).

[0071] Specifically, the processing server (120) can extract AI analysis result data from the modified biometric data acquired from the artificial intelligence server (110). For example, the processing server (120) can extract AI analysis result data from the modified biometric data in XML format. The processing server (120) can process the extracted AI analysis result data to match the format of other analysis result data managed by the electronic medical record unit (400). In particular, the processing server (120) can process the AI ​​analysis result data based on the method and form of the UI (User Interface) that displays other test results or analysis results of other biometric data for the patient on the display in the electronic medical record unit (400). To this end, the processing server (120) can process the AI ​​analysis result data with reference to the protocol that the electronic medical record unit (400) uses to transmit and receive data. At this time, the protocol of the electronic medical record unit (400) may include the HL7 protocol and the FHIR protocol.

[0072] According to one embodiment of the present disclosure, the processing server (120) may identify an order corresponding to AI analysis result data, i.e., an order requesting AI analysis, based on examination order information received from the electronic medical record unit (400) (or middleware unit (300)). For example, the processing server (120) may identify a specific examination order corresponding to AI analysis result data based on a patient identifier, examination name, order code, etc. The processing server (120) may structure or process the identified specific examination order to correspond to the identified specific examination order and transmit the structured or processed data to the electronic medical record unit (400). The processing server (120) may convert the AI ​​analysis result data into a visualizable examination report format and generate it in the form of a PDF document, an image, a graph, or the like. The generated visualization data may be integrated into and output by the electronic medical record unit (400). In addition, the generated visualization data may be generated based on the same layout and template as other existing examination results.

[0073] The processing server (120) can determine the structural format of the test result UI (User Interface) in which the test result data of the patient's other biometric data is displayed, and process the AI ​​analysis result data based on the protocol of the electronic medical record unit (400) so that the AI ​​analysis result data corresponds to the test result UI structural format.

[0074] In particular, the processing server (120) can acquire analysis result data (hereinafter, “other analysis result data”) for other biometric data other than biometric data corresponding to the AI ​​analysis result data, and process the AI ​​analysis result data and the other analysis result data by integrating them. Here, the processing by integration may mean processing the AI ​​analysis result data and the other analysis result data so that their structures and formats are consistent.

[0075] Specifically, the processing server (120) may receive analysis result data for other biometric data from a biometric signal measuring device or a separate analysis server that acquires other biometric data, including AI analysis result data received from the artificial intelligence server (110). Here, the analysis result data for other biometric data may be analysis results based on other biometric data other than electrocardiogram data, such as heart rate, blood pressure, body temperature, and blood oxygen saturation. In this case, each analysis result may have a different format and structure.

[0076] The processing server (120) can integrate various analysis result data received from multiple sources and process them into a format suitable for the protocol and user interface structure of the electronic medical record unit (400) within the hospital. For example, each result data can be standardized to follow a medical data standard protocol such as HL7 or FHIR, and reconstructed according to the field structure (e.g., test name, result value, unit, reference range, judgment result, etc.) available in the test result UI of the electronic medical record unit (400).

[0077] Thereafter, the processing server (120) transmits the processed analysis result data to the electronic medical record unit (400), thereby allowing medical staff to check the analysis result in the same manner as the existing test result.

[0078] The electronic medical record unit (400) can integrate the processed AI analysis result data received from the processing server (120) with the test result data of the patient's other biometric data and provide it (S330).

[0079] Specifically, if the test result UI provided by the electronic medical record system has a certain data structure, the processing server (120) can display the AI ​​analysis result data according to the structure. For example, if the electronic medical record system displays the test result of other biometric data on the test result UI by configuring it with fields such as 'Test Name', 'Result Value', 'Unit', 'Reference Range', and 'Interpretation', the processing server (120) can convert the AI ​​analysis result data into the same format and provide it to the electronic medical record unit (400). Accordingly, the AI ​​analysis result data can be converted to have the same data structure as other existing test result data (e.g., blood test result data), and through this, the AI ​​analysis result can be output in conjunction with the test result UI of other biometric data in the electronic medical record system.

[0080] In addition, the processing server (120) can perform unit conversion and code conversion of the AI ​​analysis result to maintain consistency with the data processing method of the electronic medical record system. For example, if the AI ​​analysis result is provided as 'Left ventricular systolic dysfunction risk score: 85 points', it can be converted to 'Test name: Artificial intelligence (AI) left ventricular systolic dysfunction risk', 'Result value: 85', 'Unit: Point', 'Reference range: 0-100', 'Judgment: High' according to the test result format required by the electronic medical record system. In addition, if the electronic medical record system uses the HL7 (Health Level 7) or FHIR (Fast Healthcare Interoperability Resources) protocol for exchanging and storing medical data, the processing server (120) can convert the AI ​​analysis result data into a format suitable for the corresponding protocol and transmit it.

[0081] FIG. 4 is an exemplary diagram providing an artificial intelligence-based analysis result of an artificial intelligence system (100) according to one embodiment of the present disclosure.

[0082] The processing server (120) can adjust the sorting and grouping of data so that AI analysis results can be effectively displayed within the electronic medical record system's test results UI. The processing server (120) can adjust the data arrangement order so that AI analysis results are provided within a list alongside existing blood test items, or create a separate section to allow medical staff to intuitively view the analysis results.

[0083] Referring to FIG. 4, when the other biometric data is blood data, the electronic medical record unit (400) may, upon receiving the processed AI analysis result data (and blood analysis result data) from the processing server (120), integrate the patient's blood test result data and the processed AI analysis result data to include the AI ​​analysis result on the patient's blood test result UI. Specifically, the patient's blood test result UI (700) includes a list composed of blood test results according to a plurality of blood test items, and upon receiving the processed AI analysis result data from the processing server (120), the electronic medical record unit (400) may add an AI analysis result field to the list, and upon receiving a request for viewing the patient's blood test results, display the blood test result UI (710) including the AI ​​analysis result.

[0084] Through this, the artificial intelligence system (100) of the present disclosure converts and processes AI analysis result data into the same format as existing test result data managed in an electronic medical record system, thereby enabling medical staff to intuitively compare and utilize existing test data and AI analysis results without a separate UI or additional data conversion work. Furthermore, by automating data linkage between the electronic medical record system and the artificial intelligence server, complex existing manual processes can be minimized and data consistency can be maintained.

[0085] According to one embodiment of the present disclosure, the artificial intelligence system (100) may further include a receiver unit (hereinafter, referred to as a second receiver unit). The second receiver unit may obtain modified biometric data (11) from the artificial intelligence server (110) and transmit it to the processing server (120) (or middleware unit (300)). Specifically, the artificial intelligence server (110) may modify the biometric data (10) by including an analysis result obtained through an artificial intelligence-based model in the biometric data (10), and transmit the modified biometric data to the second receiver unit. At this time, the second receiver unit may convert the received modified biometric data (11) into image data. In particular, when the analysis result corresponds to a disease score, the second receiver unit may convert the modified biometric data into image data including information about the disease score and transmit the converted image data to the processing server (120). The processing server (120) can transmit the received image data together with the processed AI analysis data to the electronic medical record unit (400). For example, by linking the processed AI analysis data and the image data, the processing server (120) can display the image data on the display of the electronic medical record unit (400) through a pop-up window when a user selects the AI ​​analysis result displayed on the electronic medical record unit (400).

[0086] According to one embodiment of the present disclosure, the artificial intelligence server (110) can transmit the modified biometric data to the second receiver unit, including the biometric data (10) and the analysis results. That is, the artificial intelligence server (110) can transmit the initially acquired biometric data (i.e., original data) together with the modified biometric data to the second receiver unit. At this time, the second receiver unit can convert both biometric data (10) (i.e., original data and modified biometric data) into image data and transmit it to the middleware unit (300). Through this, medical staff using the electronic medical record unit (400) can compare and review the interpretation result sheet including the artificial intelligence-based analysis result and the original interpretation result sheet. At this time, the artificial intelligence server (110) may obtain image data corresponding to the biometric data (10) (i.e., original data) converted by the receiver unit (200) that initially obtained the biometric data, and transmit the obtained image data to the second receiver unit, and transmit it to the processing server (130) together with the image data corresponding to the modified biometric data (11). At this time, the processing server (120) may connect the processed AI analysis data and the image data corresponding to the original data and the image data including the AI ​​analysis result, and then transmit it to the electronic medical record unit (400).

[0087] Meanwhile, a non-transitory computer readable medium storing a program sequentially performing a method for providing an artificial intelligence-based analysis result of an artificial intelligence system according to the present disclosure may be provided.

[0088] A non-transitory readable medium refers to a medium that permanently stores data and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, the various applications or programs described above may be stored and provided on non-transitory readable media, such as a CD, DVD, hard disk, Blu-ray disc, USB, memory card, or ROM.

[0089] The various embodiments of the present disclosure described above can be combined with additional embodiments and modified within a scope that would be understood by those skilled in the art in light of the detailed description set forth above. It should be understood that the embodiments of the present disclosure are illustrative in all respects and not restrictive. For example, each component described as a single component may be implemented in a distributed manner, and likewise, components described as distributed may be implemented in a combined manner. Accordingly, all changes or modifications derived from the meaning, scope, and equivalent concepts of the claims of the present disclosure should be construed as being included within the scope of the present disclosure.

Claims

1. In a system that provides analysis results based on artificial intelligence, An artificial intelligence system that analyzes a patient's electrocardiogram data using an artificial intelligence-based model to obtain AI (Artificial Intelligence) analysis result data, processes the AI ​​analysis result data based on the protocol of the electronic medical record unit, and transmits the AI ​​analysis result data to the electronic medical record unit; and An electronic medical record unit for managing the patient's medical records including the electrocardiogram data and the AI ​​analysis result data; The above electronic medical record unit, Provides the processed AI analysis result data received from the artificial intelligence system by integrating it with the test result data of the patient's other biometric data. System.

2. In paragraph 1, The above artificial intelligence system, By inputting the above electrocardiogram data into the above artificial intelligence-based model, a score corresponding to the possibility of a disease is obtained, the disease grade of the patient is determined by comparing the score with a reference value, and the score and the disease grade of the patient are obtained as the AI ​​analysis result data. System.

3. In paragraph 2, The above artificial intelligence system, Identifying the structural format of the test result UI (User Interface) in which the test result data of the other biometric data of the patient is displayed, and processing the AI ​​analysis result data based on the protocol of the electronic medical record unit so that the AI ​​analysis result data corresponds to the test result UI structural format. System.

4. In paragraph 3, The above biometric data is, It's blood data, The above artificial intelligence system, The patient's blood test result data and the processed AI analysis result data are integrated and transmitted to the electronic medical record unit so that the AI ​​analysis result is included on the patient's blood test result UI. System.

5. In paragraph 4, The UI of the blood test results of the above patient is, Contains a list of blood test results according to multiple blood test items, The above electronic medical record unit, When receiving the processed AI analysis result data from the artificial intelligence system, adding the AI ​​analysis result field to the list, and when receiving a request to inquire about the patient's blood test result, displaying the blood test result UI including the AI ​​analysis result. System.

6. In paragraph 1, The above AI analysis results are: It is in XML format, The protocol of the above electronic medical record unit is: Including HL7 protocol and FHIR protocol, System.

7. In paragraph 1, Further comprising a receiver unit for acquiring and storing biometric data, The above artificial intelligence system, When a storage event of electrocardiogram data in the receiver unit is detected, the stored electrocardiogram data is obtained from the receiver unit, and the stored electrocardiogram data is analyzed using the artificial intelligence-based model to obtain an analysis result. System.

8. In paragraph 7, The above artificial intelligence system, When a storage event of electrocardiogram data in the receiver unit is detected, the stored electrocardiogram data is compared with the electrocardiogram data for the patient previously stored in the receiver unit to determine whether the stored electrocardiogram data is new electrocardiogram data. System.

9. In paragraph 7, The above artificial intelligence system, When receiving a request for AI analysis results from the electronic medical record unit, the electrocardiogram data is obtained from the receiver unit, and the electrocardiogram data is analyzed using the artificial intelligence-based model to obtain the AI ​​analysis results. System.

Images