Clinical judgement support device, clinical judgement support system, clinical judgement support method, and program

Abstract

To discriminate identification of situation classification and a similar example related to nursing care through machine learning and present details of care for solving the problem of a patient, thereby supporting clinical judgement.SOLUTION: A clinical judgement support device of the present invention comprises: a learning unit 11f that performs machine learning using attribute data of a patient and situation classification data related to nursing care as learning data to build a mathematical model for situation classification discrimination; a storage unit 13 that stores data related to a similar example and details of care and the mathematical model; a classification unit 11c that receives input of the attribute data of the patient and discriminates the situation classification related to clinical judgement on the patient with the mathematical model; an example identification unit 11d that identifies a similar example corresponding to the situation classification; and a care detail identification unit 11e that identifies the details of care corresponding to the similar example.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a technology for supporting clinical decisions that combines evidence-based data related to home nursing care, etc., derived by AI (Artificial Intelligence) technology with the experiential knowledge of visiting nurses, etc. [Background technology]

[0002] In the medical field, digital technology has been advancing rapidly, such as online medical consultations, AI-assisted diagnosis support for doctors, and daily data collection using wearable devices. In the field of nursing, research is being conducted on risk detection of patients using AI technology from biometric data and nursing records of patients at medical institutions overseas, nursing records using voice recognition technology, etc. In addition, various technologies have been proposed to support the creation and management of assessments of patients.

[0003] For example, Patent Document 1 discloses a technology that divides the nursing process into stages of assessment, nursing diagnosis, planning, execution, and evaluation, manages data related to each stage, displays this data, and also accepts and manages the input of data related to each stage. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 11-282933 Summary of the Invention [Problem to be solved by the invention]

[0005] However, home nursing, in particular, requires responding to a greater variety of situations than care in a hospital, and there are many situations where nurses, especially those with little experience, have difficulty in judging the situation and making decisions. Patent Document 1 can manage each piece of data in the nursing process, but it cannot present the care content to solve the patient's problems to the nurse, etc., and does not sufficiently alleviate the sense of difficulty felt by the visiting nurse.

[0006] The present invention has been made in consideration of these problems, and its purpose is to provide technology that supports clinical judgment when visiting nurses and others have difficulty making clinical judgments by identifying situation classifications related to nursing care and distinguishing similar cases through machine learning, and presenting care content to resolve the problems of the patient. [Means for solving the problem]

[0007] In order to solve the above problems, a clinical decision support device according to one embodiment of the present invention is a clinical decision support device capable of freely communicating with a terminal device, and includes a learning unit that constructs a mathematical model for determining a situation classification by performing machine learning using attribute data of a recuperate and situation classification data related to nursing care as learning data, a receiving unit that receives the attribute data of the recuperate from the terminal device, a memory unit that stores at least data related to similar cases and care contents and the mathematical model, a classification unit that accepts input of the attribute data of the recuperate and determines a situation classification related to the clinical judgment of the recuperate using the mathematical model, a case identification unit that identifies the similar case corresponding to the situation classification, a care content identification unit that identifies the care content corresponding to the similar case, and a transmission unit that transmits data related to the similar case and the care content to the terminal device.

[0008] A clinical decision support system according to another aspect of the present invention is a clinical decision support system including a terminal device and a clinical decision support device, wherein the terminal device is equipped with a first transmission unit that transmits attribute data of a recuperating patient to the clinical decision support device and requests the clinical decision support device to present similar cases and care contents, a first reception unit that receives data relating to the similar cases and care contents from the clinical decision support device, and a display unit that performs display based on the data relating to the similar cases and care contents, and the clinical decision support device performs machine learning using the attribute data of the recuperating patient and situation classification data relating to nursing care as learning data. the system comprises a learning unit that constructs a mathematical model for determining a situation classification by performing the above-mentioned processing; a second receiving unit that receives attribute data of the recuperate from the terminal device; a memory unit that stores at least data related to similar cases and care contents and the mathematical model; a classification unit that accepts input of the attribute data of the recuperate and determines a situation classification related to the clinical judgment of the recuperate using the mathematical model; a case identification unit that identifies the similar case corresponding to the situation classification; a care content identification unit that identifies the care content corresponding to the similar case; and a second transmission unit that transmits data related to the similar case and the care content to the terminal device.

[0009] Another aspect of the present invention is a medical decision support method using a clinical decision support device that can freely communicate with a terminal device, in which a learning unit constructs a mathematical model for determining situation classification by performing machine learning using attribute data of a recuperate and situation classification data related to nursing care as learning data, a receiving unit receives the attribute data of the recuperate from the terminal device, a memory unit stores at least data related to similar cases and care contents and the mathematical model, a classification unit accepts input of the attribute data of the recuperate and determines a situation classification related to the clinical judgment of the recuperate using the mathematical model, a case identification unit identifies the similar case corresponding to the situation classification, a care content identification unit identifies the care content corresponding to the similar case, and a transmission unit transmits data related to the similar case and the care content to the terminal device.

[0010] A program according to another aspect of the present invention causes a clinical decision support device, which has a memory unit for storing data relating to similar cases and care contents and a mathematical model, and is capable of communicating with a terminal device, to function as a learning unit that constructs a mathematical model for determining a situation classification by performing machine learning using attribute data of a recuperate and situation classification data related to nursing care as learning data, a receiving unit that receives the attribute data of the recuperate from the terminal device, a classification unit that accepts input of the attribute data of the recuperate and determines a situation classification related to the clinical judgment of the recuperate using the mathematical model, a case identification unit that identifies the similar case corresponding to the situation classification, a care content identification unit that identifies the care content corresponding to the similar case, and a transmitting unit that transmits data relating to the similar case and the care content to the terminal device. Effect of the Invention

[0011] According to the present invention, when visiting nurses or other such people have difficulty making clinical judgments, it is possible to provide technology that supports clinical judgments by identifying situation classifications related to nursing care and distinguishing similar cases through machine learning, and presenting care content to resolve the issues faced by patients. [Brief description of the drawings]

[0012] [Figure 1] 1 is a configuration diagram of a clinical decision support system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a configuration diagram of a clinical decision support device of the system. [Diagram 3] FIG. 2 is a configuration diagram of a terminal device of the system. [Figure 4] 13 is a flowchart showing the flow of processing related to recuperating person information registration. [Diagram 5] 13 is a flowchart showing a process flow relating to presenting similar cases to nurses. [Figure 6] FIG. 13 is a diagram showing a display example of a My Page screen. [Figure 7] FIG. 13 is a diagram showing a display example of a recuperating patient list screen. [Figure 8] FIG. 13 is a diagram showing an example of a display of a recuperating patient information registration screen. [Figure 9] FIG. 13 is a diagram showing an example of a display of a recuperating patient information registration screen. [Figure 10] FIG. 13 is a diagram showing an example of a situation classification / similar case screen display. [Figure 11] FIG. 13 is a diagram showing a display example of a new difficulty registration screen. [Figure 12] FIG. 13 is a diagram showing a display example of a care content list screen. [Figure 13] FIG. 13 is a diagram showing a display example of a care content explanation screen. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0014] In the following embodiment, similar cases and care contents are presented based on the assessment and care contents of similar cases by experienced visiting nurses, thereby quickly discovering and resolving individual issues and difficulties of patients and their families, and improving the quality of care by visiting nurses. In addition, in order to improve the predictive performance of machine learning, data on similar cases and care is continuously collected and accumulated as big data, and is used for learning. This will be described in detail below.

[0015] FIG. 1 shows the configuration of a clinical decision support system according to an embodiment of the present invention.

[0016] As shown in the figure, the clinical decision support system according to the embodiment of the present invention is configured by connecting a clinical decision support device 1 consisting of a server device, etc., a nurse's terminal device 2, and a patient's family member's terminal device 3 to a communication network 4 such as the Internet via wireless or wired communication. Note that various devices such as notebook personal computers, desktop personal computers, smartphones, tablet terminals, etc. can be used as the terminal devices 2 and 3.

[0017] In this configuration, the clinical decision support device 1 constructs a mathematical model for situation classification discrimination by performing machine learning using the attribute data of the recuperate and the situation classification data related to nursing care as learning data. When the attribute data of the recuperate is received from the nurse's terminal device 2, the input of the attribute data of the recuperate is accepted, and the situation classification related to the recuperate's clinical judgment is discriminated by the mathematical model. Then, when a request for presentation of similar cases and care contents is received from the nurse's terminal device 3, the clinical decision support device 1 identifies similar cases corresponding to the situation classification, identifies care contents corresponding to the similar cases, transmits data related to the similar cases and care contents to the nurse's terminal device 3, and prompts the display and confirmation of these. In addition, when the clinical decision support device 1 receives feedback of data from the nurse's terminal device 3 that transmitted the data related to the similar cases and care contents, the clinical decision support device 1 further performs machine learning using the data related to the feedback as learning data, thereby updating the mathematical model for situation classification discrimination. The data related to this feedback can be text data, etc., but is not limited thereto.

[0018] FIG. 2 shows the configuration of a clinical decision support device in the clinical decision support system, and will be described.

[0019] As shown in the figure, the clinical decision support device 1 has a control unit 11, a communication I / F 12, and a storage unit 13. The control unit 11, the communication I / F 12, and the storage unit 13 are communicatively connected via a bus line. In addition to these components, the device may have an operation input unit such as a keyboard or a mouse, and a display unit such as a liquid crystal display for performing various displays. The communication I / F 12 is realized by, for example, a NIC, and is connected to a communication network 4 such as the Internet by wire or wirelessly, and communicates with terminal devices 2, 3, etc.

[0020] The storage unit 13 is realized by, for example, a semiconductor memory element such as a RAM or a flash memory, a HDD, or an optical disk device, and stores in advance the programs to be executed by the control unit 11. The storage unit 13 also has a recuperating patient information storage unit 14, a nurse information storage unit 15, a similar case storage unit 16, a care information storage unit 17, and a learning storage unit 18.

[0021] The recuperator information storage unit 14 stores attribute data of the recuperator, a status classification assigned at the time of registration, etc., linked to the recuperator ID. More specifically, the attribute data of the recuperator includes, for example, the recuperator's date of birth, sex, name of main injury or illness, family members living together, symptoms that are difficult to control, the situation, the recuperator's and family's degree of independence in daily life with dementia, the degree of independence in medical management, the recuperator's and family's level of understanding of the illness, the recuperator's and family's intentions, the uncertainty of intentions, the family's ability to care, and data related to a trusting relationship with a nurse. However, it goes without saying that the data is not limited to these.

[0022] Here, the classification of situations is divided into eight categories: 1) high-life-risk illnesses and conditions, 2) difficulty in building relationships between the patient and their family and nurses, 3) lack of caregiving ability on the part of the caregiver and family, 4) unclear and misaligned wishes of the patient and their family regarding treatment and care, 5) difficulty in managing illnesses and medical treatment independently, 6) difficulty in intervening in family relationships, 7) difficulty in nurses responding to medical care and 8) difficulty in interacting with other professions. However, it goes without saying that the classification is not limited to these.

[0023] The nurse information storage unit 15 stores attribute data of nurses, patient IDs of patients for whom they are providing care, etc., linked to the nurse ID. More specifically, the attribute data of nurses includes data such as the nurse's name, date of birth, sex, years of experience as a visiting nurse, experience in other nursing professions, years of experience as a other nurse, etc.

[0024] The similar case storage unit 16 stores past case information (similar cases) related to clinical judgments in home nursing care and the like, corresponding situation classifications, etc., in association with case IDs. The similar cases to be presented to the nurse's terminal device 3 can be read out by referring to the similar case storage unit 16 using the situation classification assigned to the patient under their care as a key.

[0025] The care information storage unit 17 stores care contents and their details, case IDs of corresponding similar cases, and their priorities, linked to the care IDs. Here, the following care contents are set in advance. That is, first, minimize the impact on life caused by responding to the illness, second, prevent sudden changes and troubles, third, respond to sudden changes, fourth, support the decision-making of the home care recipient and family, fifth, build a trusting relationship with the home care recipient and family, sixth, support the family who is responsible for caregiving, seventh, determine the possibility of continuing home care, eighth, establish a visiting nursing system, ninth, cooperate with the attending physician, tenth, cooperate with multiple professions, etc. For each of these care contents, the corresponding similar case ID is stored, and further, a score is given to the correspondence with the reclassification of these care contents, and the score value is stored together as the priority.

[0026] In addition, the learning memory unit 18 stores various data for machine learning as well as the mathematical model calculated by machine learning.

[0027] The control unit 11 is realized by a CPU, an MPU, etc. The control unit 11 is composed of an integrated circuit such as an ASIC, an FPGA, etc. The control unit 11 reads out and executes a program stored in the storage unit 13, thereby functioning as a communication unit 11a, a management unit 11b, a classification unit 11c, a case identification unit 11d, a care content identification unit 11e, a learning unit 11f, a main control unit 11g, etc.

[0028] In such a configuration, the communication unit 11a functions as a receiving unit that receives attribute data of the patient from the nurse's terminal device 2 via the communication I / F 12, and a transmitting unit that transmits data related to similar cases and care contents to the nurse's terminal device 2.

[0029] When management unit 11b receives recuperate attribute data etc. from nurse's terminal device 2, it assigns a recuperate ID, links the recuperate attribute data etc. to the recuperate ID, and stores and manages the recuperate attribute data etc. in recuperate information storage unit 14. Furthermore, when management unit 11b receives nurse attribute data etc. from nurse's terminal device 3, it assigns a nurse ID, links the nurse attribute data etc. to the nurse ID, and stores and manages the nurse attribute data etc. in nurse information storage unit 15.

[0030] The classification unit 11c receives input of attribute data of the recuperate, and determines a status classification related to the clinical judgment of the recuperate by a mathematical model stored in the learning storage unit 18 of the storage unit 13. This status classification is stored together with the recuperate information storage unit 14.

[0031] The case identifying unit 11d refers to the similar case storage unit 16 to identify a similar case that corresponds to the situation classification of the recuperate.

[0032] The care content identification unit 11e refers to the care information storage unit 17 and identifies the care content corresponding to the similar case identified by the case identification unit 11d. As described above, the care information storage unit 17 stores the care content, the details of the care content, the case ID of the corresponding similar case, and its priority, in association with the care ID. More specifically, the care information storage unit 17 stores the corresponding similar case ID for each care content, and further scores the correspondence with the reclassification of these care contents, and stores the score value together as the priority. Therefore, the care content identification unit 11e can identify the care content and its details corresponding to the similar case identified by the case identification unit 11d, and further the similar case corresponding to the care content.

[0033] The learning unit 11f performs machine learning using the recuperate's attribute data, the situation classification data related to nursing care, and the similar case data as learning data to build a mathematical model for situation classification discrimination, and stores the model in the learning storage unit 18. Furthermore, the learning unit 11f receives feedback of data from the terminal device 3 of the nurse who transmitted the similar case and care content data, and performs further machine learning using the feedback data as learning data to update the mathematical model for situation classification discrimination. This feedback data is assumed to be, for example, text data, but is of course not limited to this.

[0034] Furthermore, the learning unit 11f may perform machine learning using the situation classification data, similar case data, and data related to care content as learning data, to calculate the depth of correspondence between the situation classification, similar cases, and care content as a priority, and store the result in the similar case memory unit 16 and the care information memory unit 17.

[0035] In addition to the above, the main control unit 11g is responsible for various controls such as authentication when the terminal devices 2 and 3 access the system.

[0036] The configuration of the terminal device of the clinical decision support system will be described with reference to Fig. 3. Here, the configuration of the terminal device 2 of the nurse is illustrated, but the terminal device 3 of the family of the patient undergoing treatment has the same configuration.

[0037] As shown in the figure, the terminal device 2 (3) has a control unit 21, a communication unit I / F 22, a display unit 23, an operation unit 24, and a storage unit 25. The units 21 to 25 are connected to each other so as to be able to communicate with each other via a bus line. The communication I / F 22 is realized by, for example, a NIC or the like, and is connected to a communication network 4 such as the Internet by wire or wirelessly, and communicates with the clinical decision support device 1 and the like.

[0038] The display unit 23 is realized by a liquid crystal display or the like, and performs various displays. The operation unit 24 is realized by a mouse, a keyboard, or the like, and accepts various operational inputs by the user. The display unit 23 and the operation unit 24 may be integrated into a touch panel. The storage unit 25 is realized by, for example, a semiconductor memory element such as a RAM or a flash memory, a HDD, or an optical disk device, and stores the programs executed by the control unit 21.

[0039] The control unit 21 is realized by a CPU, an MPU, etc., and may be configured by an integrated circuit such as an ASIC, an FPGA, etc. The control unit 21 executes a program stored in the storage unit 25, thereby functioning as a communication unit 21a, a display control unit 21b, and a main control unit 21c.

[0040] In such a configuration, the communication unit 21a functions as a transmission unit that requests data in a format such as HTML related to similar cases and care contents from the clinical decision support device 1, and a reception unit that receives data in a format such as HTML related to similar cases and care contents from the clinical decision support device 1. The display control unit 21b controls the display unit 23 to display data in a format such as HTML related to similar cases and care contents sent from the clinical decision support device 1. In addition, the main control unit 21c is responsible for various controls such as sending various requests from the terminal device.

[0041] When registering the nurse's attribute data from the nurse's terminal device 2, a screen like that shown in Fig. 6 is displayed on the display unit 23, and after inputting the nurse's name, date of birth, sex, years of experience as a visiting nurse, experience in other nursing jobs, years of experience as a other nurse, etc. as the nurse's attribute data, the communication unit 21a transmits these attribute data to the clinical decision support device 1 under the control of the main control unit 21c by pressing the register button, and the attribute data is registered on the clinical decision support device 1. As described above, in addition to these attribute data, the nurse information storage unit 15 also stores the IDs of the patients in charge, so that it is possible to present a list of patients in charge for each nurse, as shown in Fig. 7.

[0042] Hereinafter, the processing procedure for registering and classifying patient information by the clinical decision support system will be described with reference to the flowchart of FIG.

[0043] In the nurse's terminal device 2, the main control unit 21c makes a login request to the clinical decision support device 1 (S1). In the clinical decision support device 1, the communication unit 11a receives this login request, and the main control unit 11g performs a predetermined authentication (S2). If the authentication is successful, data in a format such as HTML related to a recuperate information registration screen is transmitted to the recuperate and family terminal device 2 (S3). In the recuperate and family terminal device 2, the communication unit 21a receives this data, and under the control of the display control unit 21b, the recuperate information registration screen is displayed on the display unit 23 (S4).

[0044] The patient information registration screen is configured, for example, as shown in Figures 8 and 9. As shown in the figures, the patient's name, date of birth, sex, main illness, family members living together, and difficult-to-control symptoms and conditions can be entered as basic information, and the patient and family conditions can be selected from the following specifications: dementia daily life independence level, medical management independence level, level of understanding of illness, uncertainty of will and intention, trust with nurses, and family caregiving ability. In addition, tasks and assessments can also be entered as text data, for example.

[0045] On the nurse's terminal device 2, input and selection are made on these screens of the display unit 23, and when the registration (update) button is selected, the main control unit 21c transmits this as the patient's attribute data to the clinical decision support device 1 via the communication unit 21a (S5).

[0046] In the clinical decision support device 1, the communication unit 11a receives attribute data (S6), the classification unit 11c refers to a mathematical model stored in the learning memory unit 18 to perform situation classification based on the attribute data (S7), and the management unit 11b registers (updates) and manages the attribute data and situation classification, etc. in the recuperate information storage unit 14 by linking them to the recuperate ID (S8). This completes one example of processing related to recuperate information registration and classification. The classification unit 11c may perform situation classification each time the recuperate information is updated or at a specified timing, and update the situation classification in the recuperate information storage unit 14 as appropriate.

[0047] As mentioned above, the situation classification is divided into eight categories: 1) high-life-risk illnesses and conditions, 2) difficulty in building relationships between the patient and their family and nurses, 3) lack of caregiving ability of the caregiver and family, 4) unclear and misaligned wishes of the patient and their family regarding treatment and recuperation, 5) difficulty in managing illnesses and medical treatment independently, 6) difficulty in intervening in family relationships, 7) difficulty in nurses responding to medical care and 8) difficulty in relationships with other professions. However, it goes without saying that the classification is not limited to these.

[0048] Next, a process procedure for presenting similar cases and care details to nurses by the clinical decision support system will be described with reference to the flowchart of FIG.

[0049] In the nurse's terminal device 2, the main control unit 21c makes a login request to the clinical decision support device 1 (S11). In the clinical decision support device 1, the communication unit 11a accepts the login request, and the main control unit 11g performs a predetermined authentication (S12), and if the authentication is successful, transmits data in a format such as HTML related to a recuperating patient list screen to the nurse's terminal device 2 (S13). In the nurse's terminal device 2, the communication unit 21a receives this data, and under the control of the display control unit 21b, the recuperating patient list screen is displayed on the display unit 23 (S14).

[0050] Next, when a recuperator is selected from the list of recuperators on the nurse's terminal device 2, and a request for the presentation of the situation classification and similar cases is made by the main control unit 21c via the communication unit 21a (S15), the communication unit 11a of the clinical decision support device 1 accepts the request for the situation classification and the presentation of similar cases (S16). Then, the case identification unit 11d identifies the situation classification of the selected recuperator by referring to the recuperator information storage unit 14, and then identifies similar cases corresponding to the situation classification by referring to the similar case storage unit 16 (S17), and transmits the similar case data, etc., in a format such as HTML to the nurse's terminal device 2 via the communication unit 11a (S18).

[0051] In the nurse's terminal device 2, the communication unit 21a receives the similar case data and the like, and under the control of the display control unit 21b, displays a situation classification / similar case screen on the display unit 23 (S19).

[0052] This situation classification / similar case screen is configured as shown in Fig. 10. As shown in the figure, on the situation classification / similar case screen, similar cases related to each situation classification are presented in descending order of score (priority) so that further presentation of their care contents can be selected. That is, in the example of Fig. 10, for example, the situation classification of a life-threatening medical condition / condition has the highest score and is therefore presented at the top, and the title of its care contents is presented in a selectable manner.

[0053] If the similar case does not match the presented difficulty on the situation classification / similar case screen, the screen transitions to a screen as shown in Fig. 11, and the user is prompted to input the reason why the similar case does not match the difficulty by text data or the like. This text data is transmitted to the clinical decision support device 1 and is used when the learning unit 11f updates the mathematical model.

[0054] Next, in the nurse's terminal device 2, one care content is selected from the situation classification / similar case screen, and the main control unit 21c makes a request for presentation of a care content list via the communication unit 21a (S20). In the clinical decision support device 1, the communication unit 11a accepts the request for presentation of the care content list (S21). Then, the care content specification unit 11e specifies the details of the care content by referring to the care information storage unit 17 (S22), and transmits data related to the care content list in a format such as HTML via the communication unit 11a to the nurse's terminal device 2 (S23). In the nurse's terminal device 2, the communication unit 21a receives the data related to the care content list, and displays the care content list screen on the display unit 23 under the control of the display control unit 21b (S24).

[0055] This care content list screen is configured as shown in FIG. 12. As shown in the figure, details of the care content are presented in order of priority. In this example, "responding to sudden changes" has the highest priority (priority 1) as a care content, and is therefore presented at the top left of the screen. When an item for which more details are desired to be confirmed is selected on the nurse's terminal device 3 from this care content list screen, the screen switches to a care content explanation screen as shown in FIG. 13, making it possible to confirm the explanation of the care content in more detail. This completes a series of processes relating to the presentation of similar cases and care contents, etc.

[0056] As described above, according to the embodiment of the present invention, it is possible to realize clinical judgment support for visiting nurses' care that reflects the experiential knowledge of experienced visiting nurses. Here, clinical judgment refers to the process of grasping the nursing problems of the patient and his / her family and determining care to solve the problems. When it is difficult to carry out nursing care or solve the problems of the patient and his / her family, optimal advice derived from similar cases can be given, which makes it possible to optimize clinical judgment, educate nurses, and improve the quality of nursing care.

[0057] That is, by operating the terminal device, the nurse can input the patient's information, and can receive the classification of the condition and display, similar cases, and the care contents corresponding to the similar cases, so that the nurse can identify and refer to the care contents from the similar cases, and can provide feedback in the form of responses to a questionnaire afterwards.The clinical decision support device can accumulate this feedback and use it for learning, thereby further improving the accuracy of judgment.

[0058] In the embodiment of the present invention, a method is used in which known data with high learning efficiency is actively collected and learning is performed with a small amount of data. Information regarding causal relationships in the judgment process of visiting nurses with high levels of specialized knowledge is quantified and classified, and the technology incorporates augmented intelligence that combines AI and the clinical judgment of visiting nurses.

[0059] Although the embodiment of the present invention has been described above, it goes without saying that the present invention is not limited to this embodiment and various improvements and modifications can be made without departing from the spirit and scope of the present invention. [Explanation of symbols]

[0060] 1...clinical decision support device, 2...terminal device (for nurses), 3...terminal device (for family members of patients undergoing treatment), 4...communication network, 11...control unit, 11a...communication unit, 11b...management unit, 11c...classification unit, 11d...case identification unit, 11e...care content identification unit, 11f...learning unit, 11g...main control unit, 12...communication I / F, 13...memory unit, 14...patient information memory unit, 15...nurse information memory unit, 16...similar case memory unit, 17...care information memory unit, 18...learning memory unit, 21...control unit, 21a...communication unit, 21b...display control unit, 21c...main control unit, 22...communication I / F, 23...display unit, 24...operation unit, 25...memory unit.

Claims

1. A clinical decision support device capable of communicating with a terminal device, A learning unit that performs machine learning using attribute data of recuperates and situation classification data related to nursing care as learning data to construct a mathematical model for situation classification discrimination; a receiving unit for receiving attribute data of the recuperate from the terminal device; A storage unit that stores at least data related to similar cases and care contents and the mathematical model; a classification unit that receives an input of attribute data of the recuperate and determines a status classification related to a clinical judgment of the recuperate using the mathematical model; a case identification unit that identifies the similar case corresponding to the situation classification; a care content specification unit that specifies the care content corresponding to the similar case; A clinical decision support device comprising: a transmission unit that transmits data relating to the similar cases and the care contents to the terminal device.

2. The attribute data of the recuperating patient is data related to the recuperating patient's date of birth, sex, name of the main injury or illness, family members living with the recuperating patient, symptoms that are difficult to control, the recuperating patient's and family's degree of independence in daily life due to dementia, degree of independence in medical management, the recuperating patient's and family's level of understanding of the recuperating patient's condition, the recuperating patient's and family's intentions, uncertainty of intentions, family caregiving ability, and trusting relationship with the nurse. The clinical decision support device according to claim 1 .

3. The classification of situations is made up of eight categories: first, illnesses and conditions that pose a high risk to life; second, the difficulty of building relationships between the patient and their family and nurses; third, insufficient caregiving capacity of the caregiver and family; fourth, unclear and misaligned wishes of the patient and their family regarding treatment and care; fifth, difficulty in managing illnesses and medical treatment independently; sixth, difficulty in intervening in family relationships; seventh, difficulty for nurses to respond to medical care and care; and eighth, difficulty in interacting with other professions. The clinical decision support device according to claim 1 .

4. The learning unit receives feedback of data from the terminal device of the nurse that transmitted the data related to the similar cases and the care contents, and further performs machine learning using the data related to the feedback as learning data, thereby updating a mathematical model for situation classification and discrimination. The clinical decision support device according to claim 1 .

5. A clinical decision support system including a terminal device and a clinical decision support device, The terminal device a first transmission unit that transmits attribute data of the recuperate to the clinical decision support device and requests the clinical decision support device to present similar cases and care contents; a first receiving unit that receives data related to the similar cases and care contents from the clinical decision support device; a display unit that displays information based on the data related to the similar cases and care contents, The clinical decision support device comprises: A learning unit that performs machine learning using attribute data of recuperates and situation classification data related to nursing care as learning data to construct a mathematical model for situation classification discrimination; A second receiving unit that receives attribute data of the recuperate from the terminal device; A storage unit that stores at least data related to similar cases and care contents and the mathematical model; a classification unit that receives an input of attribute data of the recuperate and determines a status classification related to a clinical judgment of the recuperate using the mathematical model; a case identification unit that identifies the similar case corresponding to the situation classification; a care content specification unit that specifies the care content corresponding to the similar case; a second transmission unit that transmits data related to the similar cases and the care contents to the terminal device. Clinical decision support system.

6. The attribute data of the recuperating patient is data related to the recuperating patient's date of birth, sex, name of the main injury or illness, family members living with the recuperating patient, symptoms that are difficult to control, the recuperating patient's and family's degree of independence in daily life due to dementia, degree of independence in medical management, the recuperating patient's and family's level of understanding of the recuperating patient's condition, the recuperating patient's and family's intentions, uncertainty of intentions, family caregiving ability, and trusting relationship with the nurse. The clinical decision support system of claim 5.

7. The classification of situations is made up of eight categories: first, illnesses and conditions that pose a high risk to life; second, the difficulty of building relationships between the patient and their family and nurses; third, insufficient caregiving capacity of the caregiver and family; fourth, unclear and misaligned wishes of the patient and their family regarding treatment and care; fifth, difficulty in managing illnesses and medical treatment independently; sixth, difficulty in intervening in family relationships; seventh, difficulty for nurses to respond to medical care and care; and eighth, difficulty in interacting with other professions. The clinical decision support system of claim 5.

8. The learning unit receives feedback of data from the terminal device of the nurse that transmitted the data related to the similar cases and the care contents, and further performs machine learning using the data related to the feedback as learning data, thereby updating a mathematical model for situation classification and discrimination. The clinical decision support system of claim 5.

9. A method using a clinical decision support device that can communicate with a terminal device, comprising: The learning unit performs machine learning using the attribute data of the patient and the situation classification data related to nursing care as learning data to construct a mathematical model for situation classification discrimination, A receiving unit receives attribute data of the recuperator from the terminal device, a storage unit stores at least data relating to similar cases and care contents and the mathematical model; A classification unit receives an input of attribute data of the recuperate and determines a status classification related to a clinical judgment of the recuperate using the mathematical model; a case identification unit that identifies the similar case corresponding to the situation classification; a care content identification unit identifying the care content corresponding to the similar case; A transmission unit transmits data relating to the similar cases and the care contents to the terminal device. Clinical decision support method.

10. A clinical decision support device that includes a storage unit that stores data related to similar cases and care contents and a mathematical model and is capable of freely communicating with a terminal device, A learning unit that constructs a mathematical model for classifying and discriminating between conditions by performing machine learning using attribute data of patients and condition classification data related to nursing care as learning data; A receiving unit for receiving attribute data of the recuperate from the terminal device; a classification unit that receives an input of attribute data of the recuperate and determines a status classification related to a clinical judgment of the recuperate using the mathematical model; a case identification unit that identifies the similar case corresponding to the situation classification; A care content identification unit that identifies the care content corresponding to the similar case; and A program that functions as a transmission unit that transmits data relating to the similar cases and the care contents to the terminal device.

Images