Nursing care work assistance system, nursing care work assistance device, nursing care work assistance method, and program

Abstract

This nursing care work assistance system for assisting in nursing care work comprises: a data aggregation unit that aggregates data related to nursing care; and a processing unit that performs data processing by using the data. The data aggregation unit stores metadata and a subnet of the data stored in an external storage of the system. The data aggregation unit accesses the data in the external storage by using the subnet and stores the data in a concealed state. The nursing care work assistance system further comprises a processing model aggregation unit that aggregates processing models that are used by the processing unit for the data processing. The processing model aggregation unit stores the processing models in a concealed state.

Description

Nursing Care Business Support System, Nursing Care Business Support Device, Nursing Care Business Support Method, and Program

[0001] The present invention relates to a nursing care business support system, a nursing care business support device, a nursing care business support method, and a program.

[0002] Aging is progressing in countries around the world. For example, in Japan and Germany, the proportion of people aged 65 or older in the total population (aging rate) exceeds 21%, and they are called super-aged societies. In the future, aging is expected to progress rapidly not only in developed countries but also in developing countries. Due to the progress of aging, the need for nursing care is increasing even more. On the other hand, the decline in the quality of nursing care due to the shortage of nursing care personnel has become a problem.

[0003] In order to address such problems, the digital transformation (DX) of nursing care is required. By the DX of nursing care, stakeholders such as the nursing care service user himself / herself, nursing care providers, medical institutions, and local governments can share and utilize information regarding the nursing care service user. As a result, it is expected that even with limited nursing care personnel, efficient and high-quality nursing care can be provided. However, the DX of nursing care has not been fully implemented yet.

[0004] The Japanese government is currently promoting the DX of nursing care. The Japanese government (Ministry of Health, Labour and Welfare) provides an information system named the Scientific Nursing Care Information System (LIFE: Long-term care Information system For Evidence) for the purpose of improving the quality of nursing care (Non-Patent Document 1). LIFE collects data on care plans, the implementation status of care, and the progress of the condition of nursing care service users after care from nursing care facilities and workplaces across the country in a unified format, analyzes the collected data, and feeds back the results to the nursing care services of each nursing care provider. In this way, LIFE supports evidence-based nursing care (scientific nursing care) in order to improve the quality of nursing care.

[0005] Ministry of Health, Labour and Welfare, "About the Scientific Care Information System (LIFE)", [online], [Accessed October 8, 2024], Internet < URL: https: / / www.mhlw.go.jp / stf / shingi2 / 0000198094_00037.html >

[0006] However, LIFE had the following challenges: (1) Data entry was time-consuming, and concrete feedback that could contribute to quality improvement could not be obtained. (2) Because data was not kept confidential within the system, businesses were concerned about insufficient security and tended to avoid entering detailed data, only providing general feedback on areas for improvement.

[0007] This invention has been made in view of these circumstances, and aims to provide a caregiving support system, a caregiving support device, a caregiving support method, and a program that can obtain concrete feedback from the system that can contribute to quality improvement without requiring time and effort for data input.

[0008] According to a first aspect of the present invention, a caregiving support system is a system that supports caregiving operations, comprising a data aggregation unit that aggregates data related to caregiving, and a processing unit that performs data processing using the data, wherein the data aggregation unit stores metadata and subnets of the data stored in external storage of the system.

[0009] According to a second aspect of the present invention, the caregiving support device is a device that supports caregiving operations, and stores metadata and subnets of caregiving data stored in external storage of the device, and provides the data to a processing unit that performs data processing.

[0010] According to a third aspect of the present invention, a method for supporting nursing care operations includes the steps of: a data aggregation unit that aggregates data related to nursing care storing metadata and subnets of the data stored in external storage of the data aggregation unit; and a processing unit that performs data processing using the data.

[0011] According to a fourth aspect of the present invention, the program causes a computer to perform the following steps: a data aggregation unit that aggregates data related to nursing care stores metadata and subnets of the data stored in external storage of the data aggregation unit; and a processing unit that performs data processing using the data.

[0012] According to the present invention, it is possible to obtain specific feedback from the system that can contribute to quality improvement without requiring extra effort to input data.

[0013] This is a diagram illustrating a care support system according to the first embodiment of the present invention. This is a flowchart showing a care work support method according to the first embodiment of the present invention. This is a diagram illustrating an example of creating a report to be submitted to the national or local government. This is a diagram illustrating an example of extracting know-how from individual care data. This is a diagram illustrating a care support system according to the second embodiment of the present invention. This is a diagram illustrating a data lake of the prior art. This is a diagram illustrating a virtual data lake according to the present invention. This is a diagram illustrating a data sandbox according to the present invention. This is a diagram illustrating a data sandbox according to the present invention.

[0014] The embodiments of the present invention will be described below with reference to the drawings, but the following embodiments are not intended to limit the invention as defined in the claims.

[0015] <First Embodiment> Hereinafter, a caregiving support system, caregiving support device, caregiving support method, and program according to the first embodiment of the present invention will be described with reference to Figures 1 to 4.

[0016] Figure 1 is a diagram illustrating a care support system according to the first embodiment of the present invention. The care support system 100 is a system operated by a care support platform operator. Hereinafter, the care support system 100 may be referred to as the "care support platform".

[0017] The care support platform 100 receives care-related data 121a, 121b, and 121c from care providers 120a, 120b, and 120c. The care support platform 100 processes the data in response to requests from data users 130a and 130b. The data processing uses the care-related data 121a, 121b, and 121c, and the processing models (programs) 141a, 141b, and 141c that are necessary for the requests from data users 130a and 130b. The care support platform 100 allows data users 130a and 130b to refer to the data Da and Db resulting from the data processing.

[0018] The care support system 100 includes a data sandbox 110. The data sandbox 110 includes a data aggregation unit 111, a processing model aggregation unit 112, a confidentiality processing unit 113, and a processing result storage unit 114.

[0019] The data sandbox 110 is a mechanism for processing data related to nursing care 121a, 121b, 121c and processing models (programs) 141a, 141b, 141c while keeping them confidential from third parties. Details of the data sandbox will be described later.

[0020] The data aggregation unit 111 provides the caregiving data 121a, 121b, and 121c for data processing. The data aggregation unit 111 includes a search unit (not shown) for data users to search for caregiving data. The data aggregation unit 111 does not store copies of the caregiving data 121a, 121b, and 121c themselves, but rather the metadata 122a, 122b, and 122c of each of these data, as well as subnets 123a, 123b, and 123c. In other words, the data aggregation unit 111 is a virtual data lake. Details of the virtual data lake, metadata, and subnets will be described later.

[0021] The processing model aggregation unit 112 provides processing models (programs) 141a, 141b, and 141c for data processing. The processing model aggregation unit 112 includes a search unit (not shown) for data users to search for processing models. The processing model aggregation unit 112 stores the processing models (programs) 141a, 141b, and 141c that perform data processing. The processing models (programs) include a generative AI capable of natural language processing.

[0022] The data aggregation unit 111 and the processing model aggregation unit 112 are equipped with an encryption unit (not shown) that encrypts the data and processing models related to caregiving, respectively. Here, encryption refers to, for example, encryption using a common key.

[0023] The data concealment processing unit 113 performs data processing using the caregiving-related data and processing model obtained by the data user through a search. The data concealment processing unit 113 performs data processing using the concealed data and processing model in an concealed state. The data concealment processing unit 113 has an input unit (not shown) that accepts input from the data user. For example, prompts for performing data processing using generation AI are input to the input unit.

[0024] The processing result storage unit 114 stores the data resulting from the data processing. The result data is stored in the processing result storage unit 114 in an anonymized state. Data users 130a and 130b access the data Da and Db stored in the processing result storage unit 114 using information necessary to access the anonymized data (e.g., a shared key).

[0025] Care providers 120a, 120b, and 120c are businesses that operate care facilities.

[0026] Care providers need to receive public subsidies, such as insurance, to operate their care facilities. In order to receive these public subsidies, care providers must prove to the national and local governments (Figure 1, 130a) that they are properly managing their care businesses. Therefore, care providers must prepare and submit reports on the care businesses they operate to the national and local governments.

[0027] Care providers employ care staff such as caregivers, care assistants, care managers, care administration staff, and nurses. These care staff plan the care to be provided to the person receiving care and create a care plan. They then provide care based on the care plan. They also record the care provided in a care log.

[0028] Care providers 120a, 120b, and 120c each manage care-related data 121a, 121b, and 121c, respectively. Care-related data 121a, 121b, and 121c include, for example, care plans and care log records.

[0029] The care support system 100 virtually aggregates care-related data 121a, 121b, and 121c held by multiple care providers 120a, 120b, and 120c into a data aggregation unit 111. The care support system 100 automatically creates reports using the confidential data 121a, 121b, and 121c, along with a processing model (program) that includes a generative AI capable of natural language processing. This eliminates the need for care staff to create reports themselves.

[0030] The care support system 100 performs statistical processing of care-related data 121a, 121b, and 121c, as well as insurance point processing and insurance amount calculations based on the results of the statistical processing, simultaneously with the creation of the report. This eliminates the need for national and local governments (130a) that receive the report data to perform insurance point processing and insurance amount calculations themselves.

[0031] The care support system 100 extracts know-how in typical care situations using confidential data 121a, 121b, and 121c, and a processing model (program) that includes a generative AI capable of natural language processing. This allows other care providers (130b in Figure 1) to utilize the accumulated care know-how.

[0032] The data aggregation unit (virtual data lake) 111 of the care support system 100 processes data using confidential data 121a, 121b, and 121c, and a processing model (program) that includes a generative AI capable of natural language processing. This has the following effects: (1) It is no longer necessary to copy the data 121a, 121b, and 121c themselves to the care support system 100, and it is not necessary to input the data into the care support platform. (2) Because the data is confidential within the system, it is possible to alleviate the concerns of care providers that the security is insufficient, and thus more detailed care data can be aggregated and used from a larger number of care providers.

[0033] The following describes the virtual data lake and data sandbox used in this invention.

[0034] [Virtual Data Lake] A virtual data lake is a data infrastructure technology that enables the utilization of ubiquitous data without consolidating it into a single location. In other words, a virtual data lake virtually integrates unevenly distributed data. The virtual data lake will be explained below using Figures 6A and 6B.

[0035] As shown in Figure 6A, when utilizing data scattered across multiple data provider locations (two locations in Figures 6A and 6B: 1020a and 1020b), copies of the data stored at each data provider location (1021a and 1021b) are consolidated into a single location (hereinafter referred to as the "data utilization location") (1010) to create one large data repository. This data repository is called a "data warehouse" if the data is structured, or a "data lake" if the data is unstructured or not. In the following, we will use the term "data lake" consistently to deal with the general case.

[0036] Multiple data provider locations (1020a, 1020b) are connected to a data utilization location (1010) via a network (NW). Therefore, copies of each data can be created at the data utilization location from each data provider location via the network. Once a data lake (1011) is created at the data utilization location, data users can access and utilize the data from the data utilization location rather than from the individual data provider locations.

[0037] When creating a data lake, it is necessary to copy all of the data provider's data, even if the data users actually use is only a small portion of the data aggregated in the data lake. Therefore, creating a data lake can be time-consuming, especially if the data provider's data is large in scale or continuously generated over time.

[0038] Another problem with data lakes is the risk of losing governance over the data. Specifically, copies of data stored by data providers at their own locations are stored in a data lake that the data providers do not control. Consequently, there is a risk that copies of data stored in the data lake may be further copied by data users, and that data copied by data users may be copied and used by third parties, again beyond the control of the data providers.

[0039] The data infrastructure technology that solves these problems associated with data lakes is the virtual data lake. As shown in Figure 6B, the virtual data lake (1012) aggregates not the data (1021a, 1021b) from data providers themselves, but the metadata (1022a, 1022b) of that data. Here, metadata refers to all "data that describes data," such as the location, creator, and format of a given data. Metadata is generally smaller in size than the corresponding data.

[0040] As shown in FIG. 6B, the virtual data lake (1012) stores subnets (1023a, 1023b) of the data (1021a, 1021b) of data providers in association with the metadata (1022a, 1022b) of that data. Here, a subnet is identification information (e.g., a URL using http) that specifies the connection destination of the data of a specific data provider in a system consisting of the bases of multiple data providers and data usage bases.

[0041] With these configurations, the virtual data lake (1012) virtually aggregates ubiquitous data and manages it centrally. Thereby, the virtual data lake brings advantages to both data users and data providers.

[0042] The virtual data lake enables data users to search for and acquire data quickly across the bases of multiple data providers. Thereby, data users can efficiently use only the necessary data on demand.

[0043] The virtual data lake enables data providers to always manage their own data at their own bases and provide data to data users in response to requests. Thereby, it becomes easier for data providers to maintain governance over their own data.

[0044] For example, by using the virtual data lake, a mechanism can be realized that makes data users comply with the data usage policies created by data providers. The data usage policy describes the content that users (data providers, data users) have agreed on in advance regarding the use of data. The data usage policy includes, for example, whether data can be made public, whether data can be taken out, and the conditions under which derivative data can be used.

[0045] The present invention has, for example, the following effects by using the virtual data lake.

[0046] When a care provider refers to general know-how information about care extracted from data provided by multiple care providers and wishes to know the individual examples in which the know-how was used, there may be a desire to obtain more detailed information about the specific examples of each corresponding measure. For example, in the case of a care recipient suffering from dementia who develops symptoms of delusion of theft, when the care provider investigates the corresponding measures for the deterioration of the symptoms and the measures for reducing the symptoms, there may be a desire to obtain more detailed information about the specific examples of each corresponding measure. In such cases, according to the present invention, it becomes easier to search, inquire, request disclosure, and refer to the source data associated with the know-how information.

[0047] After a care recipient who had been admitted to a facility operated by a certain care provider transferred to another facility operated by another care provider, in a specific case of individual care, the care provider after the transfer may wish to know what kind of care the care provider before the transfer provided. In preparation for such cases, it is inefficient to transfer all data related to care between care providers in advance because it takes time to copy the data and it is necessary to manage copies of a large amount of data that may not be used at the facility after the transfer. In such cases, according to the present invention, the care provider after the transfer can request disclosure and refer to the data managed by the facility before the transfer as needed, and can save the trouble of managing the data before the transfer at the facility after the transfer.

[0048] There may be cases where a care provider does not want to permanently store a copy of the data of the care recipients it manages on an external storage. For example, when a care provider is concerned about the security of an external storage, or when a care provider is concerned that governance over the data it manages will become ineffective. Even in such cases, according to the present invention, since the virtual data lake does not need to have a copy of the data of each care provider, the concerns of the care provider can be resolved. This promotes the integrated use of the data of multiple care providers.

[0049] [Data Sandbox] A data sandbox is a technology that allows data to be processed while keeping the data and the algorithms used for processing confidential. Here, a sandbox refers to a special execution environment for software, an area with severely restricted external access. Below, we will explain an example of data processing using a data sandbox using Figures 7A, 7B, and 7C.

[0050] As shown in Figure 7A, data processing using the data sandbox is implemented using a platform (1121), an authentication infrastructure (1122), and a remote access server (1123).

[0051] The platform (1121) is a computing environment provided by a platform operator. The computing environment includes a CPU, memory, and data storage (none of which are shown in the diagram). The platform (1121) forms a data sandbox (1110) according to the data usage policy (DP). The platform (1121) includes a policy management unit (1111) that manages the data usage policy (DP).

[0052] In the data sandbox (1110), a container (1112) is created. Within the container (1112), an isolated execution environment (Trustened Execution Environment, hereinafter abbreviated as "TEE") (1113) is generated for each data usage policy (DP). Hardware encryption technology is used in the TEE. That is, the CPU is configured to encrypt and use memory areas. This allows data processing to be performed with data and programs encrypted, preventing memory access by users with OS administrative privileges. Therefore, the confidentiality of data processing can be enhanced.

[0053] Furthermore, data processing confidentiality can be enhanced by using secure computation protocols instead of, or in conjunction with, hardware encryption technology. Secure computation is a technology that protects data during the data processing process, rather than during data communication or storage. This allows processing to be performed "without viewing the contents of the data."

[0054] One example of a secure computation protocol is multi-party computation based on secret sharing. Secret sharing is a method of dividing data into multiple encrypted data, distributing and storing the divided encrypted data on multiple servers, and using some or all of these as needed. Multi-party computation is a method in which multiple servers that store the divided encrypted data perform calculations and exchange the encrypted data according to predetermined procedures.

[0055] The authentication infrastructure (1122) is a computing environment. The authentication infrastructure (1122) authenticates users who are eligible to use the data sandbox (1110).

[0056] Individual users who have access to the data sandbox (1110) provide their own data and programs to the platform (1121). In the case of using a virtual data lake, the user providing the data (user A providing data D in Figure 7A) can provide the aforementioned metadata and subnet (collectively represented as D' in Figure 7A) instead of creating a copy of data D on the platform (1121).

[0057] On the platform (1121), the TEE (1113) performs data processing using data and programs provided by the user. The TEE (1113) stores the data resulting from the data processing. In Figure 7A, data processing is performed using data D provided by user A and program P provided by user B, and the resulting data R is obtained.

[0058] The platform provider may provide some of the data or programs themselves. For example, if the platform provider provides program P, then user B is the same as the platform provider. However, this does not immediately mean that the platform provider is allowed to access user A's data D.

[0059] The data sandbox (1110) controls access to the data being processed (D or D'), programs (P), and the resulting data (R) of the data processing in accordance with the data usage policy (DP). This prevents unauthorized execution of the prevention program (P) and unauthorized acquisition of the data being processed (D or D') and the resulting data (R). In Figure 7B, among the users who can use the data sandbox (1110), a user (C) who is permitted to take out the resulting data (R) can access the resulting data (R), but cannot access the data (D or D') or programs (P) within the TEE (1113).

[0060] The data sandbox (1110) restricts access from within the TEE (1113) to the outside in accordance with the data usage policy (DP). This prevents a program (P) from taking data (D, R) outside in violation of the data usage policy (DP).

[0061] The data sandbox (1110) deletes the TEE (1113) that performed the data processing as soon as the data processing is complete. As shown in Figure 7C, for example, the data sandbox (1110) deletes the container (1112) (see Figure 7B) containing the TEE (1113). This reduces the risk of data (D, R) or programs (P) being leaked.

[0062] The remote station server (1123) is a server provided by the CPU vendor over the internet. The remote station server (1123) is a server for users to verify the authenticity of the TEE (1113). The remote station server (1123) retrieves the configuration information of the TEE (1113) and verifies whether the configuration is authentic. This allows the user to confirm that the TEE (1113) was created using functions provided by the CPU vendor and has not been tampered with.

[0063] A data sandbox using TEE and remote attestation allows users to verify that the use of data and algorithms brought into an isolated execution environment does not violate data usage policies, and that data and programs are kept confidential using TEE generated through genuine means.

[0064] As described above, by using a data sandbox, the data and programs provided by an authenticated user can be made inaccessible to anyone other than the user who provided them, while only the results of data processing using that data and those programs can be made accessible to those authorized to view those results.

[0065] By using a data sandbox, the present invention achieves, for example, the following effects:

[0066] In this invention, a generative AI is used to present information useful for future care based on past care-related text data, such as individual care logs. Therefore, in this invention, it is necessary to use a Large Language Model (LLM) to learn and generate text data.

[0067] Here, the text data used for learning includes information that should not be disclosed to third parties, such as the personal information of those receiving care and the trade secrets of care facilities (e.g., information regarding the operation of the care facility). Therefore, if copies of past care-related text data are collected at a data utilization center and LLM is created there, there is a risk that information that should not be disclosed to third parties may be leaked through the data utilization center.

[0068] As described above, this invention utilizes a data sandbox along with a virtual data lake. Therefore, copies of past text data related to caregiving are not aggregated at data utilization locations. In addition, third parties cannot access the caregiving data provided by caregiving providers or the LLM itself. Consequently, the risk of information that should not be known to third parties being leaked can be reduced.

[0069] The above describes the virtual data lake and data sandbox used in the present invention. The data aggregation unit 111 (virtual data lake) of the caregiving support system 100 is a caregiving support device according to the first embodiment of the present invention.

[0070] [Caregiving Support Method] The caregiving support method according to the first embodiment of the present invention will be described below with reference to Figures 2 to 4.

[0071] Figure 2 is a flowchart showing a caregiving support method according to the first embodiment of the present invention. The caregiving support method consists of a first step S101, a second step S102, a third step S103, a fourth step S104, a fifth step S105, a sixth step S106, a seventh step S107, an eighth step S108, and a ninth step S109.

[0072] First, the care staff stores the care plan and care log data (for example, 121a in Figure 1; see Figure 1 again below) in the storage of the care provider (120a) (S101). Next, the care provider issues a subnet (an externally accessible link) (123a) of the stored data to the care staff (S102). Next, the care staff creates metadata (122a) of the stored data and registers the created metadata (122a) along with the subnet (123a) in the data aggregation unit (111) of the care support system (100) (S103).

[0073] Next, the data users (130a, 130b) use the search unit (not shown) provided in the data aggregation unit 111 to search for the data (121a) they wish to use (S104). Next, the data users (130a, 130b) use the search unit (not shown) provided in the processing model aggregation unit 112 to search for the processing model (141a) they wish to use (S105). Next, the data users (130a, 130b) register the information necessary to refer to the confidential data (e.g., a common key) of the search results (121a, 141a) obtained in S104 and S105, and the results of data processing, with the confidentiality processing unit (113) (S106).

[0074] Next, the data concealment processing unit (113) downloads data (121a) from the storage of the care provider (120a) in a concealed state using the subnet (123a) registered in the data aggregation unit (111). Data concealment is performed by a data concealment unit (not shown) provided in the data aggregation unit 111. The data concealment processing unit (113) downloads the processing model (141a) from the processing model aggregation unit 112 in a concealed state. Data concealment of the processing model is performed by a data concealment unit (not shown) provided in the processing model aggregation unit 112 (S107).

[0075] Next, the data concealment processing unit 113 performs data processing using the concealed data and the concealed state processing model, and stores the resulting data in the concealed state in the processing result storage unit 114 (S108). Finally, the data users 130a and 130b access the data Da and Db stored in the processing result storage unit 114 using the information necessary to access the concealed data (S109).

[0076] Below, using Figures 3 and 4, an example of data processing in step S108 of the eighth step is shown. The processing model is a generative AI capable of natural language processing.

[0077] Figure 3 illustrates an example of creating a report to be submitted to the national or local government. The left side of Figure 3 shows a schematic example of prompts used to obtain the data necessary for creating the report from the input document (nursing record) using generation AI. The right side of Figure 3 shows the result of data processing by generation AI (output example).

[0078] As shown in Figure 3, by entering a natural language command along with the output items and input document, such as "You are a caregiving clerk. Please extract words from the entered document to match the following output items. If there are any items that cannot be extracted, please contact me," the data for the items necessary for creating the report will be automatically generated.

[0079] Figure 3 schematically shows a simple example, but in reality, the items required to create a report include a wide range of items such as the business establishment number, insurer number, insured person number, the name of the person receiving care, gender, date of birth, height, weight, body temperature, amount of food eaten, type of food eaten, date of certification, date of admission, level of independence in daily living, toilet use, bathing, walking on level ground, climbing stairs, dressing, and bowel control. According to the present invention, the effort required of care staff to search for data on each of these items and create a report can be eliminated.

[0080] Figure 4 illustrates an example of extracting know-how from individual care data. The left side of Figure 4 shows a schematic example of prompts used to obtain the data necessary for know-how extraction from input documents (home care service plan, nurse's record) using a generation AI. The right side of Figure 4 shows the results (know-how) of data processing by the generation AI.

[0081] As shown in Figure 4, by inputting commands in natural language, know-how regarding illness, symptoms, and the progression of symptoms and living conditions can be extracted from data such as care plans. In addition to the generative AI, a processing model (program) that performs statistical analysis of numerical data may also be used. This allows other care providers to utilize the accumulated know-how.

[0082] <Second Embodiment> Hereinafter, a caregiving support system according to the second embodiment of the present invention will be described with reference to Figure 5. Components common to the first embodiment will be numbered the same and their descriptions will be omitted.

[0083] As shown in Figure 5, the caregiving support system 200 according to the second embodiment of the present invention differs from the first embodiment in that the processing model aggregation unit 212 downloads processing models (programs) 241a to 241c from the provider locations 240a to 240c of the processing models (programs).

[0084] The processing model aggregation unit 212, like the processing model aggregation unit 112 of the first embodiment, includes a data concealment unit (not shown). As a result, the download of processing models (programs) and data processing by the data concealment processing unit 113 using the processing models (programs) are performed in a confidential state. Consequently, the risk of processing models (programs) being leaked to third parties is reduced, making it possible to solicit processing models (programs) from a larger number of providers.

[0085] The processing model aggregation unit 212 may download processing models (programs) as needed by data users. This eliminates the need for the processing model aggregation unit 212 to prepare large-capacity storage sufficient to store multiple processing models (programs).

[0086] Although preferred embodiments of the present invention have been described above with reference to the attached drawings, it goes without saying that the present invention is not limited to these examples. The shapes and combinations of the constituent members shown in the above examples are merely examples, and can be modified in various ways based on design requirements, etc., without departing from the spirit of the present invention.

[0087] For example, the apparatus of the present invention can also be realized using a computer and a program, and the program can be recorded on a recording medium or provided via a network.

[0088] For example, the care support systems 100 and 200 may be linked to a public database containing information on the person receiving care. The information on the person receiving care held in the public database may include, for example, information on disabilities and illnesses, information on medication history, information on health checkups, and information on vaccinations.

[0089] 100, 200... Nursing care support platform (nursing care support system) 110... Data sandbox 111... Data aggregation unit (virtual data lake, nursing care support device) 112, 212... Processing model aggregation unit 113... Confidential processing unit (processing unit) 114... Processing result storage unit 120a, 120b, 120c... Nursing care provider (external storage) 121a, 121b, 121c... Data 122a, 122b, 122c... Metadata 123a, 123b, 123c... Subnet 130a, 130b... Data user 141a, 141b, 141c, 241a, 241b, 241c... Processing model (program)

Claims

1. A system for supporting nursing care operations, comprising: a data aggregation unit for aggregating data related to nursing care; and a processing unit for performing data processing using the data, wherein the data aggregation unit stores metadata and subnets of the data stored in external storage of the system.

2. The caregiving support system according to claim 1, wherein the data aggregation unit accesses the data in the external storage using the subnet and stores the data in an anonymized state.

3. The caregiving support system according to claim 2, wherein the processing unit comprises a processing model aggregation unit that aggregates processing models used for the data processing, and the processing model aggregation unit stores the processing models in an confidential state.

4. The caregiving support system according to claim 2, wherein the processing unit performs the data processing while the data is kept confidential.

5. The caregiving support system according to claim 1, wherein the metadata and subnet are deleted after the data processing is completed.

6. A device for supporting nursing care operations, which stores metadata and subnets of nursing care-related data stored in external storage of the device, and provides the data to a processing unit for data processing.

7. A method for supporting nursing care operations, comprising the steps of: a data aggregation unit that aggregates nursing care data, storing metadata and subnets of the data stored in external storage of the data aggregation unit; and a processing unit that performs data processing using the data.

8. A program that causes a computer to perform the following steps: a data aggregation unit that aggregates data related to nursing care stores metadata and subnets of the data stored in external storage of the data aggregation unit; and a processing unit that performs data processing using the data.

Images