Medical artificial intelligence software system including data standardization system, association system, and management system

Abstract

A medical artificial intelligence software data standardization system according to an embodiment of the present invention comprises: a medical data communication unit provided to receive medical data, including at least one of an electronic medical record (EMR) or a medical image, from external medical artificial intelligence software; a standardization unit provided to standardize the medical data; a memory provided to store the standardized medical data and store existing standardized data obtained by processing existing medical data; and a control unit electrically connected to the medical data communication unit, the standardization unit, and the memory, wherein the control unit can standardize the medical data into data of a predetermined format through the standardization unit to generate standardized data.

Description

Medical AI software system including data standardization system, linkage system, and management system

[0001] The present invention relates to a medical artificial intelligence software system capable of performing data standardization, linking with external software, and software management.

[0002]

[0003] Technologies for human convenience are being developed in various fields, and the development of artificial intelligence, in particular, is not lagging behind any other field. AI is being utilized significantly in fields such as medicine, finance, education, and information processing, and has already reached a significant level of development, partially replacing the roles of practitioners in these fields.

[0004] In the medical field, various medical AI software has been developed and widely utilized. These software support medical decision-making based on specific electronic medical record (EMR) data or medical image data.

[0005] However, according to the conventional technology, since each software user, such as a hospital, uses a different software system, there is a problem that a considerable amount of time and effort is required for these software to be interconnected or for data to be transmitted and received.

[0006]

[0007] The problem to be solved by the present invention is to provide a medical artificial intelligence software system that can effectively perform data standardization, linkage with external software, and software management.

[0008]

[0009] According to one embodiment of the present invention, a medical artificial intelligence software data standardization system comprises a medical data communication unit configured to receive medical data including at least one of an electronic medical record (EMR) and a medical image from external medical artificial intelligence software, a standardization unit configured to standardize the medical data, a memory configured to store the medical data in a standardized form and to store existing standardized data in which existing medical data is processed, and a control unit electrically connected to the medical data communication unit, the standardization unit, and the memory, wherein the control unit can standardize the medical data into data in a predetermined format through the standardization unit to generate standardized data.

[0010] In the medical artificial intelligence software data standardization system according to one embodiment of the present invention, the control unit may perform a data preprocessing process of generating preprocessed data by processing at least one of case unification, removal of unnecessary spaces, removal of special characters, removal of meaningless numbers, removal of noise, image quality improvement, and size adjustment for the medical data, and may perform a process of extracting keywords from the preprocessed data.

[0011] In the medical artificial intelligence software data standardization system according to one embodiment of the present invention, the standardization unit includes a natural language processing model, and the control unit performs learning on the preprocessed data using the natural language processing model to generate learning data, and matches the medical data with existing medical data stored in the memory based on the keyword and the learning data using a predetermined matching algorithm to generate first standardized data.

[0012] In the medical artificial intelligence software data standardization system according to one embodiment of the present invention, the standardization unit further includes a data analysis model including a predetermined classification algorithm, and the control unit can extract medical indicators, which are numerical features, from the preprocessed data and extract predefined classification indicators from the preprocessed data using the classification algorithm.

[0013] In the medical artificial intelligence software data standardization system according to one embodiment of the present invention, the control unit can identify the similarity between the existing standardization data stored in the memory and the medical indicator and the classification indicator, and generate second standardization data based on the identified similarity.

[0014] In the medical artificial intelligence software data standardization system according to one embodiment of the present invention, the control unit can generate the standardized data by integrating the first standardized data and the second standardized data based on pre-adjusted weights.

[0015] The medical artificial intelligence software data standardization system according to one embodiment of the present invention further includes a user interface, and the control unit provides the standardized data to the user through the user interface, and when a data modification input from the user is obtained through the user interface, updates the standardized data based on the modification input, and stores the updated standardized data in the memory.

[0016] In the medical artificial intelligence software data standardization system according to one embodiment of the present invention, the standardization data is stored in the memory, and the control unit distinguishes non-matching data that is not included in the standardization data among the medical data or does not match the existing standardization data among the updated standardization data, stores the non-matching data in the memory, and provides the non-matching data to the user using the user interface.

[0017] A medical artificial intelligence software linkage system according to one embodiment of the present invention includes a medical data communication unit configured to receive medical data from external medical artificial intelligence software, a standardization unit configured to standardize the medical data, the medical data communication unit, and a control unit electrically connected to the standardization unit, wherein the control unit can link operation with the external artificial intelligence software using standardized data generated by standardizing the medical data into data in the same format through the standardization unit.

[0018] In the medical artificial intelligence software linkage system according to one embodiment of the present invention, the control unit may be provided to enable reversible data format conversion for the standardized data and the medical data obtained from the external artificial intelligence software and having a different data format from the standardized data, using the standardization unit.

[0019] In the medical artificial intelligence software linkage system according to one embodiment of the present invention, the control unit can share the standardized data in real time with external medical artificial intelligence software that can be connected via communication.

[0020] The medical artificial intelligence software linkage system according to one embodiment of the present invention further includes a cloud server for processing the medical data and the standardized data, and the control unit can implement distributed computing by processing the medical data and the standardized data through the cloud server and the external medical artificial intelligence software.

[0021] The medical artificial intelligence software linkage system according to one embodiment of the present invention further includes a user interface, and the medical data includes real-time hospital complexity data such as the number of outpatients, the number of inpatients, the number of waiting patients, and the expected waiting time, and the control unit provides the real-time hospital complexity data to the user through the user interface and can transmit the real-time hospital complexity data to external medical artificial intelligence software that is communicatively connectable.

[0022] According to one embodiment of the present invention, a medical artificial intelligence software management system includes a medical data communication unit configured to receive medical data from external medical artificial intelligence software, a standardization unit configured to standardize the medical data, a software classification unit configured to classify the external medical artificial intelligence software to generate classification information, a user interface configured to provide the classification information to a user, and a control unit electrically connected to the medical data communication unit, the standardization unit, the software classification unit, and the user interface, wherein the control unit can provide the classification information generated using standardized data generated by standardizing the medical data into data in the same format through the standardization unit to the user through the user interface.

[0023] In the medical artificial intelligence software management system according to one embodiment of the present invention, the external medical artificial intelligence software includes a plurality of softwares, and the control unit can generate the classification information for the plurality of softwares using the software classification unit.

[0024] In the medical artificial intelligence software management system according to one embodiment of the present invention, the classification information includes transaction information for the plurality of softwares, and the control unit can provide the transaction information to the user through the user interface.

[0025] In the medical artificial intelligence software management system according to one embodiment of the present invention, the transaction information may include at least one of design information, function information, price information, contract information, installation information, and support option information for the plurality of software.

[0026] In the medical artificial intelligence software management system according to one embodiment of the present invention, the software classification unit includes a data analysis model or a machine learning model, and the control unit analyzes or learns the classification information using the data analysis model or the machine learning model to generate a recommendation option, and can provide the recommendation option to the user through the user interface.

[0027]

[0028] According to embodiments of the present invention, operations related to external software can be standardized and utilized efficiently.

[0029] According to embodiments of the present invention, compatibility and interoperability between different software can be improved.

[0030] According to embodiments of the present invention, the work efficiency of software users can be improved.

[0031]

[0032] FIG. 1 is a conceptual diagram illustrating the concept of a medical artificial intelligence software system according to one embodiment of the present invention.

[0033] FIG. 2 is a configuration diagram showing systems that can be implemented by a medical artificial intelligence software system according to one embodiment of the present invention.

[0034] Figure 3 is a configuration diagram showing a medical artificial intelligence software data standardization system according to one embodiment of the present invention.

[0035] Figure 4 is a flowchart showing the operation flow of a medical artificial intelligence software data standardization system according to one embodiment of the present invention.

[0036] FIG. 5 is a flowchart specifically illustrating part of the operation flow of a medical artificial intelligence software data standardization system according to one embodiment of the present invention.

[0037] Figure 6 is a configuration diagram showing a medical artificial intelligence software linkage system according to one embodiment of the present invention.

[0038] Figure 7 is a conceptual diagram illustrating the concept of a medical artificial intelligence software linkage system according to one embodiment of the present invention.

[0039] Figure 8 is a configuration diagram showing a medical artificial intelligence software management system according to one embodiment of the present invention.

[0040] Figure 9 is a conceptual diagram illustrating the concept of a medical artificial intelligence software management system according to one embodiment of the present invention.

[0041]

[0042] <Explanation of symbols for major parts of the drawing>

[0043] 1: Medical AI Software System

[0044] 2: External software

[0045] 2-1: First external software

[0046] 2-2: Second external software

[0047] 2-3: Second external software

[0048] 2-n: Second external software

[0049] 10: Medical AI Software Data Standardization System

[0050] 100: Medical Data Communication Department

[0051] 110: Standardization Department

[0052] 111: Natural Language Processing Model

[0053] 112: Data Analysis Model

[0054] 120: Memory

[0055] 130: User Interface

[0056] 140: Control Unit

[0057] 20: Medical AI Software Linkage System

[0058] 200: Medical Data Communication Department

[0059] 210: Standardization Department

[0060] 220: Cloud Server

[0061] 230: User Interface

[0062] 240: Control Unit

[0063] 30: Medical AI Software Management System

[0064] 300: Medical Data Communication Department

[0065] 310: Standardization Department

[0066] 320: Software Classification Department

[0067] 330: User Interface

[0068] 340: Control Unit

[0069]

[0070] Hereinafter, with reference to the attached drawings, preferred embodiments of the present invention will be described in detail so that those skilled in the art can easily implement the invention. However, the present invention may be implemented in various different forms and is not limited or restricted by the following examples.

[0071] In order to clearly explain the present invention, a detailed description of a part that is irrelevant to the description or a related known technology that may unnecessarily obscure the gist of the present invention has been omitted, and when adding reference signs to components of each drawing in this specification, the same or similar reference signs are attached to the same or similar components throughout the specification.

[0072] In addition, terms and words used in this specification and claims should not be interpreted as limited to their usual or dictionary meanings, but should be interpreted as meanings and concepts that conform to the technical idea of ​​the present invention based on the principle that the inventor can appropriately define the concept of the term to explain his or her own invention in the best way.

[0073] FIG. 1 is a conceptual diagram showing the concept of a medical artificial intelligence software system according to an embodiment of the present invention, and FIG. 2 is a configuration diagram showing systems that can be implemented by a medical artificial intelligence software system according to an embodiment of the present invention.

[0074] Referring to FIG. 1, the medical artificial intelligence software system (1) can be electrically / communicatively connected to external software (2). The medical artificial intelligence software system (1) can be connected to the external software (2) to perform at least one of data standardization, linkage with the external software, and software management.

[0075] The external software (2) may include multiple software pieces. For example, the external software (2) may include n pieces of software pieces, from the first external software piece (2-1), the second external software piece (2-2), the third external software piece (2-3), to the n-th external software piece (2-n). The above is an example to aid understanding, and it is not necessary to have three or more pieces of software pieces. In other words, the multiple pieces of software pieces may be two pieces of software pieces, and there may be no special restrictions.

[0076] Referring to FIG. 2, the medical AI software system (1) may include a plurality of systems that enable data standardization, linkage with external software, and software management. For example, the medical AI software system (1) may include a medical AI software data standardization system (10) (hereinafter referred to as the "data standardization system"), a medical AI software linkage system (20) (hereinafter referred to as the "linkage system"), and a medical AI software management system (30) (hereinafter referred to as the "management system").

[0077] The above-described data standardization system (10), linkage system (20), and management system (30) do not necessarily have to be implemented in hardware, but can be implemented in software. In other words, since the medical artificial intelligence software system (1) can implement the operations and functions of the data standardization system (10), linkage system (20), and management system (30) through software operation, it is not necessary for each system to be implemented in different configurations when operating. In other words, the data standardization system (10), linkage system (20), and management system (30) can be implemented in overlapping configurations.

[0078] Below, the data standardization system (10), linkage system (20), and management system (30) are described in detail.

[0079] Figure 3 is a schematic diagram illustrating a medical AI software data standardization system according to one embodiment of the present invention. The description of the aforementioned embodiment can be applied identically or similarly to the present embodiment.

[0080] The data standardization system (10) may include a medical data communication unit (100).

[0081] The medical data communication unit (100) may be configured to receive medical data including at least one of an electronic medical record (EMR) and a medical image from external medical artificial intelligence software (e.g., external software (2)).

[0082] The data standardization system (10) may include a standardization department (110).

[0083] The standardization department (110) can be provided to standardize medical data.

[0084] The standardization unit (110) may include artificial intelligence models for performing standardization. For example, the standardization unit (110) may include a natural language processing model (111).

[0085] Additionally, the standardization unit (110) may include a data analysis model (112) including a predetermined classification algorithm.

[0086] The data standardization system (10) may include a memory (120).

[0087] The memory (120) can store medical data in a standardized form. In addition, the memory (120) can store existing standardized data in which existing medical data has been standardized.

[0088] The data standardization system (10) may include a user interface (130).

[0089]

[0090] *The user interface (130) can provide the user with information about standardized data processed by the data standardization system (10) so that the user can check it.

[0091] The data standardization system (10) may include a control unit (140).

[0092] The control unit (140) can be electrically connected to the medical data communication unit (100), standardization unit (110), memory (120), and user interface (130).

[0093] The control unit (140) can generate standardized data. For example, the control unit (140) can generate standardized data by standardizing medical data into data in a predetermined format through the standardization unit (110).

[0094] It may be understood that the operations of the data standardization system (10) described below are controlled by the control unit (140).

[0095] Figure 4 is a flowchart illustrating the operational flow of a medical AI software data standardization system according to one embodiment of the present invention. The description of the aforementioned embodiment can be applied identically or similarly to the present embodiment.

[0096] The data standardization system (10) can create standardized data through certain processes.

[0097] The data standardization system (10) can perform data preprocessing according to process S110. For example, the control unit (140) can perform a data preprocessing process to generate preprocessed data by processing medical data by at least one of unifying uppercase and lowercase letters, removing unnecessary spaces, removing special characters, removing meaningless numbers, removing noise, improving image quality, and adjusting size. Through the data preprocessing process for medical data, the data can be organized into a form suitable for analysis / transformation.

[0098] The data standardization system (10) can perform keyword extraction according to process S120. For example, the control unit (140) can perform a process of extracting keywords from preprocessed data.

[0099] For keyword extraction, a database (or memory (120)) containing widely used medical terms, test names, disease names, etc. can be constructed and utilized. Keyword extraction can be performed using text mining techniques, particularly term frequency-inverse document frequency (TF-IDF) analysis and topic modeling.

[0100] After keyword extraction, naming rule learning can be performed. For example, machine learning, particularly supervised learning methods, can be used to learn naming rules that appear in the names of stored medical data. This process uses pre-classified medical data as training data, allowing the system to learn to recognize and understand naming patterns in medical data.

[0101] The data standardization system (10) can generate standardized data through the standardization unit according to process S130. For example, the control unit (140) can generate standardized data using the natural language processing model (111) and / or data analysis model (112) of the standardization unit (110).

[0102] The operation of the data standardization system (10) according to process S130 is described in detail later in FIG. 5.

[0103] Figure 5 is a flowchart specifically illustrating a portion of the operational flow of a medical AI software data standardization system according to one embodiment of the present invention. The description of the aforementioned embodiment can be applied identically or similarly to the present embodiment.

[0104] The control unit (140) can generate standardized data using the natural language processing model (111) and / or the data analysis model (112) of the standardization unit (110). Referring to FIG. 5, the processing flow using the natural language processing model (111) and the data analysis model (112) is described, but it is not necessarily limited thereto, and standardized data can be generated using only one of the natural language processing model (111) and the data analysis model (112).

[0105] Below, the data standardization system (10) is described as a processing flow using the natural language processing model (111) and data analysis model (112) of the standardization department (110).

[0106] The control unit (140) can perform learning on preprocessed data using a natural language processing model (111) to generate learning data (D1).

[0107] The control unit (140) can create first standardized data by matching the acquired medical data with existing medical data stored in the memory (120) based on keywords and learning data (D1) extracted using a predetermined matching algorithm.

[0108] The control unit (140) can extract medical indicators and classification indicators (D2) through the data analysis model (112). For example, the control unit (140) can extract medical indicators, which are numerical features, from preprocessed data using the data analysis model (112), and can extract predefined classification indicators from the preprocessed data using a classification algorithm.

[0109] The data analysis model (112) can recognize, classify, and extract data patterns by utilizing machine learning, particularly supervised learning and unsupervised learning algorithms.

[0110] In the case of supervised learning, the model can be trained using predefined normal / abnormal labels, and in the case of unsupervised learning, similar data can be matched by finding natural clustering within the data.

[0111] Normal / abnormal labels or clusters can be created by extracting a portion of the entire medical data and then combining specific data elements. For example, specific elements can be selected based on the importance of the medical data and the analysis objective, such as hospital size (e.g., large hospital, small to medium-sized hospital), diagnosis (e.g., diabetes, hypertension), patient age, and gender. These can be used as criteria for defining each label or cluster. Furthermore, data groups that satisfy specific conditions can be formed, such as "hypertension-related data for male patients aged 40-50 diagnosed in small to medium-sized hospitals."

[0112] The control unit (140) can identify the similarity between existing standardized data (or existing medical data) stored in the memory (120) and the medical indicator and classification indicator (D2).

[0113] For example, similarity identification methods include Euclidean distance, Manhattan distance, and cosine similarity identification methods, which can calculate the distance between each data point and group similar data together. The control unit (140) can generate second standardized data based on the identified similarity (or grouping).

[0114] In other words, the control unit (140) can identify the similarity between existing standardized data (or existing medical data) and medical indicators and classification indicators (D2), and generate second standardized data for medical indicators and classification indicators (D2) whose similarity is greater than or equal to a threshold similarity.

[0115] The control unit (140) can generate standardized data by integrating the first standardized data and the second standardized data based on pre-adjusted weights.

[0116] The control unit (140) can provide standardized data to the user through the user interface (130).

[0117] The data standardization system (10) can also perform updates on standardized data. For example, the control unit (140) can obtain a user's data modification input via the user interface (130). When the data modification input is obtained via the user interface (130), the control unit (140) can update the standardized data based on the modification input and store the updated standardized data in the memory (120).

[0118] The data standardization system (10) can also provide non-matching data to the user.

[0119] For example, the control unit (140) can distinguish non-matching data that is not included in the generated standardized data among the acquired medical data or does not match the existing standardized data among the updated / newly stored standardized data due to modification and store the data in the memory (120).

[0120] The control unit (140) may provide the user with non-matching data using the user interface (130). In this case, the user may make additional correction inputs based on the non-matching data provided through the user interface (130). The control unit (140) may update the standardized data stored in the memory (120) based on the additional correction inputs and store the updated standardized data in the memory (120).

[0121] As described above, the data standardization system (10) standardizes operations related to external software, enabling efficient utilization. Accordingly, users can efficiently manage and utilize various medical AI software in one place, significantly improving work efficiency by providing an integrated approach to the software.

[0122] Figure 6 is a schematic diagram illustrating a medical AI software linkage system according to one embodiment of the present invention. The description of the aforementioned embodiment can be applied identically or similarly to the present embodiment.

[0123] The linkage system (20) may include a medical data communication unit (200). The medical data communication unit (200) of FIG. 6 may correspond to the medical data communication unit (100) of FIG. 3, and may be configured to receive medical data from external medical artificial intelligence software (e.g., external software (2)).

[0124] The linkage system (20) may include a standardization unit (210). The standardization unit (210) of FIG. 6 may correspond to the standardization unit (110) of FIG. 3 and may be provided to standardize medical data.

[0125] The linkage system (20) may include a cloud server (220). The cloud server (220) may be a server for processing medical data and standardized data.

[0126] The linkage system (20) may include a user interface (230). The user interface (230) of FIG. 6 may correspond to the user interface (130) of FIG. 3, and may provide the user with information about standardized data processed by the linkage system (20) so that the user can confirm it.

[0127] The linkage system (20) may include a control unit (240). The control unit (240) of FIG. 6 may correspond to the control unit (140) of FIG. 3. The control unit (240) may be electrically connected to the medical data communication unit (200), the standardization unit (210), the cloud server (220), and the user interface (230).

[0128] The operation of the linked system (20) can be linked with external medical artificial intelligence software (e.g., external software (2)). For example, the control unit (240) can link the operation with external artificial intelligence software (e.g., external software (2)) by using standardized data generated by standardizing medical data into data of the same format through the standardization unit (210).

[0129] Figure 7 is a conceptual diagram illustrating a medical AI software linkage system according to one embodiment of the present invention. The description of the aforementioned embodiment can be applied identically or similarly to the present embodiment.

[0130] The linkage system (20) can perform reversible data format conversion. For example, the control unit (240) can control the standardization unit (210) to perform reversible data format conversion. Specifically, reversible data format conversion can be performed on standardized data standardized through the standardization unit (210) and medical data obtained from external artificial intelligence software (e.g., external software (2)) and having a different data format from the standardized data.

[0131] In other words, the linkage system (20) can reversibly convert standardized data and medical data whose formats are different from those used in the external software (2). In this case, the linkage system (20) can also control the external software (2) by converting the standardized data into medical data in the format of the external software (2) and transmitting it to the external software (2). That is, the control unit (240) can convert the standardized data into external medical data and transmit it to the external software (2), and accordingly, the control unit (240) can also control the external software (2).

[0132] In this case, when control of the control unit (240) is blocked by the external software (2), control by the control unit (240) is not performed, but when control of the control unit (240) is not blocked by the external software (2), control can be performed.

[0133] The linkage system (20) can share standardized data in real time with external artificial intelligence software (e.g., external software (2)). That is, the control unit (240) can share standardized data in real time with external software (2) that can be connected to the communication.

[0134] External software (2) may be software from another hospital, may be software that patients can download, and may not have any special restrictions.

[0135] The linked system (20) can implement distributed computing through a cloud server (220) and external artificial intelligence software (e.g., external software (2)). For example, the control unit (240) can implement distributed computing by processing medical data and standardized data through the cloud server (220) and external artificial intelligence software (e.g., external software (2)). In this case, rapid computing can be performed even when processing massive amounts of data.

[0136] The linkage system (20) can provide real-time hospital complexity data to users and can also share it with external artificial intelligence software (e.g., external software (2)).

[0137] For example, medical data may include real-time hospital complexity data such as the number of outpatients, the number of inpatients, the number of waiting patients, and the expected waiting time. The control unit (240) may provide the real-time hospital complexity data to the user through the user interface (230). In addition, the control unit (240) may transmit the real-time hospital complexity data to an external artificial intelligence software (e.g., external software (2)) that can be communicatively connected. In this case, the linkage system (20) and the external artificial intelligence software (e.g., external software (2)) may share the real-time hospital complexity data with each other. In addition, if the external software (2) includes software that can be downloaded by the patient, the patient may also check the real-time hospital complexity data.

[0138] As described above, the linkage system (20) can improve compatibility and interoperability between different software. Specifically, it can contribute to the provision of more accurate and comprehensive medical services by enabling users to comprehensively analyze and utilize data generated from various software.

[0139] Figure 8 is a schematic diagram illustrating a medical AI software management system according to one embodiment of the present invention. The medical AI software management system may be a system capable of providing an integrated purchasing solution for medical AI software. The description of the aforementioned embodiment may be applied to the present embodiment in the same or similar manner.

[0140] The management system (30) may include a medical data communication unit (300). The medical data communication unit (300) of FIG. 8 may correspond to the medical data communication unit (100) of FIG. 3, and may be configured to receive medical data from external medical artificial intelligence software (e.g., external software (2)).

[0141] The management system (30) may include a standardization unit (310). The standardization unit (310) of FIG. 8 may correspond to the standardization unit (110) of FIG. 3 and may be provided to standardize medical data.

[0142] The management system (30) may include a software classification unit (320).

[0143] The software classification unit (320) may be configured to classify external medical artificial intelligence software (e.g., external software (2)) and generate classification information. For example, the software classification unit (320) may generate classification information of the external software (2) based on medical data or converted standardized data of the external medical artificial intelligence software (e.g., external software (2)).

[0144] The software classification unit (320) may include a data analysis model or a machine learning model, and may use the data analysis model or the machine learning model to generate classification information based on medical data or converted standardized data. Furthermore, the software classification unit (320) may not only generate classification information, but also generate recommendation options for the classification information generated using the data analysis model or the machine learning model.

[0145] The management system (30) may include a user interface (330). The user interface (330) of FIG. 8 corresponds to the user interface (130) of FIG. 3 and may be configured to provide classification information to the user.

[0146] The management system (30) may include a control unit (340). The control unit (340) of FIG. 8 may correspond to the control unit (140) of FIG. 3. The control unit (340) may be electrically connected to the medical data communication unit (300), the standardization unit (310), the software classification unit (320), and the user interface (330).

[0147] The management system (30) can provide classification information to the user. For example, the control unit (340) can use the standardization unit (310) to standardize medical data into data of the same format and provide the user with classification information generated using the standardized data through the user interface (330).

[0148] Figure 9 is a conceptual diagram illustrating a medical AI software management system according to one embodiment of the present invention. The description of the aforementioned embodiment can be applied identically or similarly to the present embodiment.

[0149] As described above, the management system (30) can generate classification information. In this case, there may be cases where multiple software programs are included in the external medical AI software (e.g., external software (2)). In this case, the control unit (340) can generate classification information for the multiple software programs using the software classification unit (320).

[0150] The classification information includes transaction information for external software (2), and may include transaction information for multiple software programs if multiple software programs are available. The control unit (340) may provide the transaction information to the user through the user interface (330). The classification information may include not only transaction information, but also software management information and software security information.

[0151] Transaction information may include at least one of design information, functional information, price information, contract information, installation information, and support option information for external software (2) or multiple external software.

[0152] The management system (30) may also provide users with recommended options for multiple external software programs based on classification information. For example, the control unit (340) may analyze or learn classification information for multiple external software programs using a data analysis model or machine learning model of the software classification unit (320). Specifically, the software classification unit (320) may analyze or learn transaction information such as design information, functional information, price information, contract information, installation information, and support option information based on predefined priorities or weights. Accordingly, recommended options may be generated based on the analysis or learning in the software classification unit (320).

[0153] The control unit (340) can provide recommendation options to the user through the user interface (330).

[0154] Although the present invention has been described above with reference to limited embodiments and drawings, the present invention is not limited thereto, and various embodiments are possible within the scope equivalent to the technical idea of ​​the present invention and the patent claims to be described below by a person having ordinary skill in the art to which the present invention pertains.

[0155]

[0156] According to the present invention, operations related to external software can be standardized and utilized efficiently, compatibility and interoperability between different software are improved, and work efficiency of software users is improved, thereby making it more effective in the field of artificial intelligence software utilized in the medical field.

Claims

1. A medical data communication unit capable of receiving medical data including at least one of an electronic medical record (EMR) and medical images from external medical artificial intelligence software; A standardization department established to standardize the above medical data; The above medical data is standardized and stored, and a memory is provided to store existing standardized data in which existing medical data is processed; and It includes the medical data communication unit, the standardization unit, and the control unit electrically connected to the memory, The above control unit, A medical artificial intelligence software data standardization system that creates standardized data by standardizing the medical data into a predetermined format through the above standardization department.

2. In claim 1, The above control unit, Perform a data preprocessing process to generate preprocessed data by processing at least one of case unification, removal of unnecessary spaces, removal of special characters, removal of meaningless numbers, removal of noise, improvement of image quality, and resizing of the above medical data. A medical artificial intelligence software data standardization system that performs a process of extracting keywords from the above preprocessed data.

3. In claim 2, The above standardization part includes a natural language processing model, The above control unit, Using the above natural language processing model, learning is performed on the above preprocessed data to generate learning data, A medical artificial intelligence software data standardization system that creates first standardized data by matching the medical data with existing medical data stored in the memory based on the keywords and the learning data using a predetermined matching algorithm.

4. In claim 3, The above standardization unit further includes a data analysis model including a predetermined classification algorithm, The above control unit, Extract medical indicators, which are numerical features, from the above preprocessed data, A medical artificial intelligence software data standardization system that extracts predefined classification indicators from the preprocessed data using the above classification algorithm.

5. In claim 4, The above control unit, Identifying the similarity between the existing standardized data stored in the above memory and the above medical indicator and the above classification indicator, A medical artificial intelligence software data standardization system that generates second standardized data based on the identified similarity.

6. In claim 5, The above control unit, A medical artificial intelligence software data standardization system that generates the standardized data by integrating the first standardized data and the second standardized data based on pre-adjusted weights.

7. In claim 6, Including more user interfaces, The above control unit, Providing the standardized data to the user through the user interface, A medical artificial intelligence software data standardization system that obtains a user's data modification input through the user interface, updates the standardized data based on the modification input, and stores the updated standardized data in the memory.

8. In claim 7, The above standardized data is stored in the above memory, The above control unit, Among the above medical data, non-matching data that is not included in the above standardized data or does not match the existing standardized data among the above updated standardized data is distinguished and stored in the above memory, A medical artificial intelligence software data standardization system that provides the user with the above-mentioned non-matching data using the above-mentioned user interface.

9. Medical data communication unit designed to receive medical data from external medical artificial intelligence software; A standardization department established to standardize the above medical data; and Including the medical data communication unit and the control unit electrically connected to the standardization unit, The above control unit, A medical artificial intelligence software linkage system that links operation with the external artificial intelligence software by using the standardized data generated by standardizing the medical data into data of the same format through the standardization department.

10. In claim 9, The above control unit, A medical artificial intelligence software linkage system that uses the above standardization unit to enable reversible data format conversion for the standardized data and the medical data obtained from the external artificial intelligence software and having a different data format from the standardized data.

11. In claim 10, The above control unit, A medical artificial intelligence software linkage system that shares the above standardized data in real time with external medical artificial intelligence software that can be connected via communication.

12. In claim 9, Further comprising a cloud server for processing the above medical data and the above standardized data, The above control unit, A medical artificial intelligence software linkage system that implements distributed computing by processing the medical data and the standardized data through the cloud server and the external medical artificial intelligence software.

13. In claim 9, Including more user interfaces, The above medical data includes real-time hospital complexity data such as the number of outpatients, number of inpatients, number of waiting patients, and expected waiting time. The above control unit, Provide the above real-time hospital complexity data to the user through the above user interface, A medical artificial intelligence software linkage system that transmits the above real-time hospital complexity data to an external medical artificial intelligence software that can be connected to a communication network.

14. Medical data communication unit designed to receive medical data from external medical artificial intelligence software; A standardization department established to standardize the above medical data; A software classification unit designed to classify the above external medical artificial intelligence software and generate classification information; A user interface designed to provide the above classification information to the user; and It includes a control unit electrically connected to the medical data communication unit, the standardization unit, the software classification unit, and the user interface, The above control unit, A medical artificial intelligence software management system that provides the user with the classification information created using the standardized data created by standardizing the medical data into data of the same format through the standardization department through the user interface.

15. In claim 14, The above external medical artificial intelligence software includes multiple softwares, The above control unit, A medical artificial intelligence software management system that creates classification information for the plurality of softwares using the software classification unit.

16. In claim 15, The above classification information includes transaction information for the plurality of software, The above control unit, A medical artificial intelligence software management system that provides the above transaction information to the user through the user interface.

17. In claim 16, The above transaction information is: A medical artificial intelligence software management system including at least one of design information, function information, price information, contract information, installation information, and support option information for the above plurality of software.

18. In claim 17, The above software classification unit includes a data analysis model or a machine learning model, The above control unit, The above classification information is analyzed or learned using the data analysis model or the machine learning model to generate a recommendation option, A medical artificial intelligence software management system that provides users with recommendation options through the above user interface.

Images