JP2021099867A - Care plan formation support system - Google Patents

Abstract

To provide a care plan formation support system that can support formation of a care plan by outputting a probability of improving the care necessity level of a care receiver according to a situation of individual care receivers for each care plan.SOLUTION: A care plan formation support system 100 comprises a care plan output unit 110 functioning as a processing unit, a determination unit 120, a probability calculation unit 130, a processing change unit 140, a storage unit 150 functioning as a storage medium, a display unit 160, and an input unit 170. The determination unit 120 determines whether a care necessity level has been improved using a care plan output from the care plan output unit 110. The probability calculation unit 130 newly calculates a probability for improving the care necessity level of a care receiver using a result of the determination by the determination unit 120. The processing change unit 140 changes a care plan output support model by using as teacher data a combination with the result of the determination by the determination unit 120.SELECTED DRAWING: Figure 1

Description

The present invention relates to a care plan creation support system.

A care plan for a person requiring long-term care or a person requiring long-term care (hereinafter collectively referred to as "the person being cared for") who receives care or support from another person in daily life including bathing, excretion, and meals is called a care plan. The care plan includes a specific content of the long-term care service and a service plan schedule that defines the date and time when the long-term care service is provided.

The care plan is usually created by the care manager, taking into account the long-term care benefits determined by the care recipient according to the degree of care required. When a person who wants to be certified as requiring long-term care applies for certification of long-term care, the long-term care insurer who received the application has an accredited investigator of an accredited research institution investigate the physical condition of the applicant. The above survey is conducted by selecting a preset item from two or more options that suits the applicant's situation, or by inputting an item in text. The results of this survey are input to a computer or the like and used for the primary determination of the degree of care required by a predetermined calculation method. The long-term care insurer has the long-term care certification examination committee, which is the insurer of the long-term care insurer, certify the degree of care required of the applicant based on the above-mentioned primary judgment, other special notes of the care recipient, and the written opinion of the attending physician. The degree of long-term care required thus certified sets the upper limit of long-term care benefits for the long-term care recipient.

After the upper limit of the long-term care benefit is set, the care manager prepares a care plan suitable for the long-term care recipient in consideration of the above long-term care benefit. At this time, the care manager or the person entrusted with the care manager newly evaluates the condition of the care recipient at the time of creating the care plan (hereinafter, also simply referred to as “assessment”). After that, the care manager creates the above care plan based on the current situation of the evaluated care recipient (hereinafter, also simply referred to as “assessment result”).

Patent Documents 1 to 6 include assessments for the purpose of streamlining care plan creation, suppressing deterioration of the quality of care plans due to the ability of care managers, and homogenizing the care plans provided. A method for assisting in the creation of a care plan from the results and a system for implementing the above method have been proposed.

By the way, the number of long-term care recipients is increasing due to the declining birthrate and aging population that have been progressing in recent years. In addition, since there are very few cases in which the degree of long-term care is improved by long-term care, the number of long-term care recipients with a high degree of long-term care is increasing as the number of long-term care recipients increases. Part of the funds required for long-term care benefits are covered by long-term care benefit costs, which are a type of social security costs, but the long-term care benefit costs per long-term care recipient increase as the degree of long-term care required increases. Therefore, due to the increase in the number of long-term care recipients and the accompanying increase in the number of long-term care recipients who require a high degree of long-term care, the long-term care benefit costs are also increasing year by year. For example, the population of so-called elderly people aged 65 and over in Japan is about 34 million in 2015 (1 in 3.7), while it is about 36.5 million in 2025 (population 3.3). It is expected to increase to 1 person), and the long-term care benefit cost is expected to increase from 10.1 trillion yen in 2015 to 21 trillion yen in 2025.

In addition, the increase in the number of long-term care recipients and the increase in the number of long-term care recipients who require a high degree of long-term care have a great impact on society. For example, problems such as so-called long-term care turnover, in which employees are forced to leave their jobs to care for their close relatives, have become apparent, and these problems reduce the quality of life of the care recipient and those related to the care recipient. It can also cause social loss by reducing the labor force.

Japanese Unexamined Patent Publication No. 11-149499 Japanese Unexamined Patent Publication No. 2000-3404 Japanese Unexamined Patent Publication No. 2000-3391 Japanese Unexamined Patent Publication No. 2001-101279 Japanese Unexamined Patent Publication No. 2006-146410 JP-A-2010-205263

As mentioned above, the financial and social burdens are increasing due to the increase in the number of long-term care recipients, and it is necessary to suppress these burdens. In particular, as the degree of long-term care increases, the amount of long-term care benefits and the amount of social burden increase, so it is desirable to curb the increase in the number of long-term care recipients. However, an effective method for reducing the number of long-term care recipients who require long-term care has not yet been established.

For example, if the care plan is devised to improve the degree of care required by the long-term care recipient, it is considered that the increase in the number of long-term care recipients with a high degree of long-term care can be suppressed. However, there are only a few examples of care plans that have been able to improve the level of long-term care, and even if these examples are applied to other long-term care recipients in different situations, the level of long-term care is expected to improve as well. It is not always possible.

In addition, in order to implement the above-mentioned ingenuity of the care plan, it is considered important to improve the efficiency of creating the care plan and to homogenize the created care plan. On the other hand, Patent Documents 1 to 6 describe that the purpose of improving the efficiency of care plan creation and homogenizing the created care plan is achieved by having a computer assist in the creation of a care plan. (Hereinafter, a system that supports the creation of a care plan is also simply referred to as a "care plan creation support system." A care plan creation support system is a system that can output useful information for a care manager to create a care plan. Often, it may have only the function of outputting only the long-term care service that should be included in the care plan, or it may have the function of simultaneously outputting the service plan schedule that defines the date and time for providing the above-mentioned long-term care service.) .. However, these documents do not consider the creation of a care plan that can improve the level of long-term care required.

The present invention has been made in view of the above problems, and a care plan is created by outputting a recommended care plan that can be expected to improve the degree of long-term care according to the situation of each individual care recipient. The purpose is to provide a care plan creation support system that can support.

One embodiment of the present invention for solving the above problems includes an input receiving unit that receives input of the result of an assessment performed on the long-term care person, a storage unit that stores a plurality of candidate care plans, and the long-term care person. It has a probability output unit that outputs the probability that the situation of the long-term care recipient will be improved by implementing the candidate care plan, and the probability output unit is a learning-processed probabilistic inference model of the assessment. The present invention relates to a care plan creation support system that inputs a result and the candidate care plan and outputs the probability that the care-requiring degree of the long-term care recipient improves.

INDUSTRIAL APPLICABILITY According to the present invention, a care plan creation support system capable of supporting the creation of a care plan by outputting a recommended care plan that can be expected to improve the degree of long-term care according to the situation of each individual care recipient is provided. Will be done.

FIG. 1 is a block diagram showing an example of a configuration of a care plan creation support system according to an embodiment of the present invention. FIG. 2 is a flowchart showing an example of processing by the care plan creation support system. FIG. 3 is a flowchart showing a specific process included in the output of the care plan (step S100). FIG. 4 is a schematic view showing the configuration of the intermediate unit in the first embodiment. FIG. 5 is a flowchart showing a specific process included in the learning process (step S200). FIG. 6 is a flowchart showing a configuration of a care plan creation support system according to another embodiment of the present invention. FIG. 7 is a block diagram showing a configuration of a care plan creation support system for still another embodiment of the present invention. FIG. 8 is a block diagram showing a configuration of a care plan creation support system for still another embodiment of the present invention.

[First Embodiment]
FIG. 1 is a block diagram showing an example of the configuration of the care plan creation support system 100 according to the embodiment of the present invention. The care plan creation support system 100 according to the present embodiment displays a care plan output unit 110, a determination unit 120, a probability calculation unit 130, a processing change unit 140, and a storage unit 150 that functions as a storage medium. A unit 160 and an input unit 170 are provided.

Although not shown, the care plan creation support system 100 includes, for example, a CPU (Central Processing Unit) as a processor, a storage medium such as a ROM (Read Only Memory) storing a control program, a RAM (Random Access Memory), and the like. It is equipped with a working memory and a communication circuit. In this case, the functions of the above-mentioned processing units are realized by the CPU executing a control program (corresponding to the "care plan creation support program" of the present invention). At least a part of the program for executing the processing by the care plan output unit, the determination unit, the probability calculation unit, and the processing change unit is stored in the server, but as in the fourth embodiment described later. , At least a part of the above program may be stored in the cloud server.

When the care plan output unit 110 provides care to the care recipient using a preset care plan (hereinafter, also simply referred to as “providing a care plan”), the degree of care required of the care recipient is determined. It is a processing unit that acquires the probability of improvement for each preset care plan and outputs the care plan with a higher acquired probability as a recommended care plan. For example, the care plan output unit 110 inputs the input situation of the long-term care recipient to a probabilistic inference model (hereinafter, simply “care plan output support model”) to provide preset care. Obtain the probability that the care-requiring degree of the care recipient will improve when the plan is provided. The stochastic inference model is an inference value for each predetermined attribute (type of care plan in the present embodiment) (in the present embodiment, the degree of care required is improved) by performing a predetermined operation on the input value. It is an inference model that outputs (probability of doing). After that, the care plan output unit 110 refers to the probability of improvement in the degree of long-term care output for each of the care plans, and selects, for example, a care plan having a higher probability of improvement in the degree of long-term care. It is a processing unit that outputs a care plan recommended for the long-term care recipient.

The determination unit 120 is a processing unit that determines whether or not the current care-requiring degree of the care recipient has improved from the past care-requiring degree of the care recipient. For example, the determination unit 120 determines whether or not the degree of need for long-term care of the long-term care recipient has improved after the long-term care is provided to the long-term care recipient using the care plan output from the care plan output unit 110. The determination unit 120 may determine whether or not the degree of care required of the care recipient has improved after the care recipient has been cared for using a care plan separately created by the care manager.

The probability calculation unit 130 uses the result of the determination by the determination unit 120 to provide the care recipient when care is provided according to the preset care plan (particularly, the care plan output by the care plan output acquisition 110). It is a processing unit that newly calculates the probability that the degree of long-term care required will improve. For example, the probability calculation unit 130 includes the total number of records stored in the storage unit 150 that have the same care plan used for the care recipient and the care plan, and the degree of care required included in the above records. The number of records determined to have improved is calculated, the ratio of the number of records determined to have improved the degree of need for long-term care to the total number of records is calculated, and the calculated ratio is calculated as the above-mentioned long-term care recipient. It is the probability that the degree of long-term care required by the person will improve.

The processing change unit 140 includes a care plan created from the past situation of the care recipient and the past situation of the care recipient and provided to the care recipient in the past, and a determination result by the determination unit 120. It is a processing unit that changes the care plan output support model using the combination with and as teacher data. Specifically, the processing change unit 140 is a combination of the same care recipient's situation and the care plan in that the probability of being acquired for a certain care plan when the care recipient's situation is input to the care plan output unit 110. The care plan output support model is changed so as to approach the probability calculated by the probability calculation unit 130.

Details of the processing by these processing units will be described later.

One of the features of the care plan creation support system 100 is that the care plan output support model is changed by learning processing. According to the learning process, it is possible to mechanically extract the features contained in the past data and change the care plan output support model so that similar results are output for inputs with similar features. Is. Therefore, for the situation of a certain long-term care person, it is determined whether or not the degree of care required has improved for each care plan provided, and these data (the situation of the long-term care person, the care plan, and each of the above care plans). By performing the above learning process using the above judgment result), the care plan output support model is changed so that a care plan with a high probability of improving the degree of long-term care can be output even in situations with similar characteristics. it can. As a result, the care plan creation support system 100 has a higher probability of improvement after the need for long-term care, depending on the characteristics of the situation of the long-term care recipient, even for individual long-term care recipients having various situations. Can be output as a recommended care plan. In addition, by preparing a large number of such data and repeating the above learning process, the accuracy of outputting a care plan with a higher probability that the degree of care required will improve for any situation of the long-term care recipient will increase. ..

The storage unit 150 is a storage medium capable of writing and reading.

The storage unit 150 can be a storage medium such as a database. The storage unit 150 contains the care recipient information, the date or time, the status of the care recipient, the care plan, the degree of care required of the care recipient, and the need for long-term care after the care is performed using the care plan. Stores a record associated with the degree or the change in the degree of long-term care required as a result of providing the above care plan. A unique record number may be assigned to the above record.

The above date or date is the date or date when the care recipient was assessed.

The above-mentioned care recipient information is information for uniquely identifying the care recipient who receives care, and may include an ID number unique to the care recipient, the name of the care recipient, gender, date of birth, and the like. ..

The situation of the care recipient can be the result of the assessment obtained by the assessment of the care recipient by the care manager or the person entrusted by the care manager. In addition, as the situation of the long-term care recipient, the contents described in the attending physician's opinion such as the general condition survey on the service usage status, the survey result on special notes, and the name of the disease may be used.

The care plan is a care plan created by the care manager from the situation of the long-term care recipient and provided to the long-term care recipient.

The degree of need for long-term care of the person to be cared for is the degree of need for long-term care of the person to be cared for certified by the long-term care certification examination committee when the above care plan is created.

The degree of need for long-term care after long-term care is provided using the above care plan is determined by the long-term care certification examination committee after providing the care plan to the same long-term care recipient. Degree.

The change in the degree of need for nursing care is data that can be aggregated, which is determined by the determination by the determination unit 120. For example, "1" is assigned to a record containing a care plan determined to have improved the degree of long-term care, and "0" is assigned to a record containing a care plan determined to have not improved the degree of long-term care. Granted.

At the time of initial setting, the storage unit 150 prepares and stores in advance a number of the records that can be calculated by the probability calculation unit 130.

The display unit 160 is a display device such as a smartphone, a PC, or a TV carried by the care manager, and displays a recommended care plan according to a display control signal input from the care plan output unit 110. The display unit 160 can be used by the care manager who sees the displayed care plan to create a care plan to be provided to the care recipient.

The input unit 170 includes various operation keys such as a numeric keypad and a start key, an audio receiver, a video receiver such as a camera, etc., receives various input operations by a user, generates input data, and generates input data. The data is output to the care plan output unit 110 and the determination unit 120. The input unit 170 is a variety of signal receiving devices configured to receive input data electrically output from another medium such as a receiving connector and output it to the care plan output unit 110 and the determination unit 120. You may. The input data output from the input unit 170 is output to the care plan output unit 110 and used for the output of the care plan, and is also output to the determination unit 120 to be output to the care plan output by the probability calculation unit 130 and the processing change unit 140. Used to change the support model.

FIG. 2 is a flowchart showing an example of processing by the care plan creation support system 100. The care plan creation support system 100 receives data on the situation of the care recipient and outputs a recommended care plan for the care recipient based on a predetermined stochastic inference model (step S100). Then, the care plan creation support system 100 performs a learning process for changing the stochastic inference model so that a care plan having a higher probability of improving the care-requiring degree of the care recipient can be output (step S200).

Although FIG. 2 shows a mode in which the learning process (step S200) is performed after the output of the care plan (step S100), the output of the care plan (step S100) is performed after the learning process (step S200). Alternatively, the output of the care plan (step S100) and the learning process (step S200) may be performed at the same time in parallel.

Hereinafter, the input data used for the processing by the care plan creation support system 100 will be described, and then more specific processing included in the output of the care plan (step S100) and the learning processing (step S200) will be described.

(Input data)
The input data is information on the assessment performed on the care recipient in order to create a new care plan, the care recipient information, the date or time of the assessment performed on the care recipient, and the care recipient. And the degree of care required of the care recipient on the date or date of the above assessment. It is preferable that the care recipient information included in the input data and the care recipient information stored in the storage unit 150 consist only of the same items, but the determination unit 120 can determine the same care recipient. As long as there is, only some items may be duplicated.

The above-mentioned situation of the long-term care recipient is the one that best suits the situation of the long-term care recipient from two or three or more options of "Yes (Y)" or "No (N)" for multiple survey items. Can be the result of selection. The types of the plurality of survey items and options are not particularly limited and can be arbitrarily determined. For example, the survey items and options defined for use in certifying the degree of care required can be used. The above-mentioned surveys for certifying the degree of care required include general condition surveys and basic surveys, but either or both of these may be used, and basic survey items and options (74 items as of July 2015). Only may be used. At this time, when the certification criteria for the degree of care required are changed, it is preferable that the above-mentioned survey items and options can be changed accordingly. Further, as for the situation of the long-term care recipient, the result of the above-mentioned general condition survey or the content of the attending physician's opinion may be input as a text.

For example, as of July 2015, the survey items used to determine the degree of long-term care required are "paralysis of the upper left limb (selection from" yes "and" no ")" and "turning over (" can be done without grasping "," Items related to the physical function and moving movement of the care recipient, such as "I can do it if I grab something" and "I can't do it"), "Movement (" Not assisted "," Need to watch over "," I Items related to the living functions of the long-term care recipient, such as "needs club assistance" and "needs full assistance"), "communication of intentions (" the surveyed person can convey intentions to others "," sometimes For cognitive abilities of care recipients such as "can communicate", "almost unable to communicate" and "cannot") and "say the date of birth and age (select from" can "and" cannot ")" Items related to mental and behavioral disorders of the care recipient, such as "become damaging when things are stolen (selection from" not "," sometimes "and" yes ")", "money Management (selection from "not unlocked", "partially unlocked" and "totally unlocked") and "daily decision-making (" can (even in special cases) "," special Items related to adaptation to social life, such as "can be done except in rare cases", "daily difficulty" and "cannot be selected"), and "dialysis (selection from" yes "and" no ")" It consists of 74 items such as items related to medical care received in the past 14 days. These items may be used as they are, some may be selected and used, or other items may be registered and used arbitrarily.

From the viewpoint of saving the trouble of inputting, the number of items of the care recipient's situation can be 5 or more and 20 or less, and further 5 or more and 15 or less. The situation of the care recipient at this time preferably consists of survey items used at the time of evaluation (monitoring) of activities of daily living (ADL) and means of daily living (IADL) performed on the care recipient. As the survey items used in the evaluation of ADL, the evaluation items used in the evaluation method such as the Barthel index, the performance status defined by the Eastern Cooperative Oncology Group (ECOG), and the functional independence measure (FIM) should be used. Can be done.

(Step S100: Output of care plan)
In the output of the care plan (step S100), the care plan output unit 110 inputs the input status of the care recipient into the care plan output support model, and acquires the probability that the degree of care required will improve for each care plan. To do. After that, the care plan output unit 110 outputs a care plan recommended for the long-term care recipient based on the probability that the degree of long-term care required for each of the long-term care plans will improve, and displays the care plan on the display unit 160.

The above care plan includes at least long-term care services. The long-term care service is the content of the specific long-term care service provided to the person being cared for. The above-mentioned long-term care service may be selected from those that can be provided by the long-term care provider. In addition to the specific contents of the long-term care service, the care plan may include a service plan schedule that defines the date and time when the long-term care service is provided.

FIG. 3 is a flowchart showing a specific process included in the output of the care plan (step S100).

First, the care plan output unit 110 inputs the situation of the care recipient into the care plan output support model, and acquires the probability that the degree of care required will improve for each care plan (step S310). After that, the care plan output unit 110 outputs one or a plurality of care plans having a high probability of improving the degree of need for care as a recommended care plan (step S320). For example, the care plan output unit 110 outputs a care plan having a high probability of output (for example, a care plan having the highest probability or a care plan corresponding to the output unit having the 10th probability) as a recommended care plan. To do.

After that, the care plan output unit 110 outputs the output care plan as a display control signal to the display unit 160, and displays the care plan on the display unit 160 (step S330). At this time, when the care plan is changed as in the third embodiment described later, the process returns to step S310, and the care plan output unit 110 causes the display unit 160 to display the modified care plan.

Finally, the care plan output unit 110 newly records the care recipient information, date or date and time, the care recipient's status and the degree of care required, and the output or changed care plan contained in the input data in the storage unit 150. Is registered as (step S340).

The above-mentioned care plan output support model is a probabilistic inference model that outputs a constant inference value for each preset type of care plan in response to the input of the care recipient's situation, and is performed by the processing change unit 140. It is configured to be changeable. The inferred value is the probability that the degree of need for long-term care of the long-term care recipient will improve when each care plan is provided to the long-term care recipient having the input situation. Care plan output support models include, for example, neural networks, Boltzmann machines and decision trees.

Hereinafter, the configuration of the care plan output support model when the care plan output support model includes a neural network will be illustrated.

The neural network (hereinafter, also simply referred to as “plan output network”) is an input layer composed of a plurality of input units, one or a plurality of intermediate layers each composed of a plurality of intermediate units, and an output composed of a plurality of output units. Has a layer. The plan output network is a multi-layered neural network having N layers having a plurality of intermediate layers from the input layer side to the output layer side, and the intermediate unit of each intermediate layer is in the middle of the previous layer. The information received from one or more intermediate units (or input units) included in the layer is output to one or more intermediate units (or output units) included in the intermediate layer of the next layer. The number of layers (N) of the plan output network may be 3 or more, and can be arbitrarily determined according to the number of assessment items and the like. For example, the number of layers (N) can be 3 or more and 25 or less, preferably 5 or more and 20 or less, and more preferably 8 or more and 16 or less.

The input layer (N = 1) has the same number of input units as the survey items included in the situation of the care recipient. The input unit corresponds to any of the plurality of survey items, and outputs the input value corresponding to each option of the survey item to one or a plurality of intermediate units as it is.

Each of the intermediate layers (N = k (1 <k <N)) has an arbitrary number of the intermediate units. The intermediate unit receives output from one or more intermediate units (or one or more input units of the input layer) of the intermediate layer of the previous layer, and one or more of the intermediate layers of the next layer. Output to the intermediate unit (or the output unit of the output layer). The intermediate layer may be a fully connected layer in which the intermediate units included in the intermediate layer are connected to all of the intermediate units (or input units) in the previous layer, or the intermediate units included in the intermediate layer are intermediate units in the previous layer. It may be a convolutional layer that joins only part of (or the input unit).

At this time, as shown in FIG. 4, the intermediate unit Z of the kth layer receives the outputs w1, w2, w3 and w4 from the intermediate units of the k-1 layer (four in the figure). After that, as shown in the equation (1), the intermediate unit Z sets a value u1 obtained by further adding a predetermined bias b to the sum of the values obtained by multiplying each output by the predetermined weights y1, y2, y3 and y4, respectively. calculate. The weight and bias can be changed by the processing change unit 140.
u1 = w1 x y1 + w2 x y2 + w3 x y3 + w4 x y4 + b ... (1)

The intermediate unit further outputs the value z1 obtained by applying the predetermined activation function f to the above u1 to the intermediate unit (or output unit) of the k + 1th layer. The type of activation function f is not particularly limited, and a sigmoid function, a logistic function, a rectified linear unit (ReLU), a softmax function, and the like can be used. Of these, the activation function f possessed by the intermediate unit is preferably ReLu from the viewpoint of improving the output accuracy and speeding up the processing.

The output layer (N = N) has the same number of output units as the number of preset care plans that can be output. Each output unit receives the output from one or more N-1 intermediate units, and the value obtained by performing the same processing as the intermediate unit is input to the input layer of the care recipient. It is output as an inferred value for the situation (probability that the degree of care required improves). At this time, the activation function f possessed by the output unit has a total of the inferred values output from all the output units of 1, and the inferred values output can be used as they are by performing a probability distribution. Functions are preferred.

At this time, each of the output units is made to correspond one-to-one with each of the care plans that can be output by the preset care plan output unit 110. Then, the inferred value output by each of the above output units is regarded as the probability that the degree of care required of the care recipient having the above situation will improve when each care plan is provided. Therefore, the care plan output unit 110 can select and output the care plan corresponding to the output unit having the larger inferred value as the care plan recommended for the care recipient having the above situation.

The weights and biases of each unit in the intermediate layer and the output layer are based on the combination of the past situation of the care recipient and the care plan selected by the care manager for the care recipient as teacher data at the time of initial setting. , It is preferable to set in advance by the learning process by the process change unit 140.

(Step S200: Learning process)
In the learning process (step S200), the determination unit 120, the probability calculation unit 130, and the process change unit 140 change the above-mentioned care plan output support model by the learning process. Specifically, the determination unit 120 determines whether or not the degree of care required included in the input data has improved from the past degree of care required of the same care recipient registered in the storage unit 150. After that, the probability calculation unit 130 determines that any combination of the care recipient's situation and the care plan included in the storage unit 150 is determined to be a care recipient having the care recipient's situation constituting the combination. On the other hand, the probability that the degree of long-term care will improve when the above care plan is provided is calculated. After that, the processing change unit 140 changes the care plan output support model by using the situation and the care plan of the care recipient constituting the combination and the probability that the degree of care required for the combination is improved as teacher data. For example, when the processing change unit 140 inputs the situation of the care recipient into the neural network, the inference value output from each output unit provides a care plan corresponding to the output unit. Change the weight and bias values of each intermediate unit and output unit so that they approach the values calculated as the probability of improvement in care level.

The above-mentioned past degree of long-term care, the situation of the past long-term care recipient and the past long-term care plan, which will be described later, are the date or date and time of the input data among the records of the same long-term care recipient registered in the storage unit 150. It is the degree of long-term care required, the situation of the long-term care recipient, and the care plan included in the record at the time when the person goes back to the past for a predetermined period from. Specifically, the date or date and time registered in the assessment date or date and time included in the record is included in the record such that the date or time is retroactive by a predetermined period from the assessment date or date and time included in the input data. The items are the above-mentioned past degree of care required, past care plan, and past care recipient status. The predetermined period can be a period from the creation of the care plan to the re-creation of the care plan, for example, 6 months.

FIG. 5 is a flowchart showing a specific process included in the learning process (step S200). The determination unit 120 collates the care recipient information included in the input data with the care recipient information registered in the storage unit 150, and the same care recipient record as the input data is registered in the storage unit 150. It is determined whether or not the information is provided (step S510).

As a result of the determination, when the same record of the care recipient is registered (step S510: YES), the determination unit 120 describes the above after the care is performed using the past care plan for the care recipient. The change in the degree of care required of the care recipient (whether or not the degree of care required has improved) is evaluated (step S520). Specifically, the determination unit 120 includes the care-requiring degree (current care-requiring degree) of the care recipient included in the input data and the past record of the same care recipient registered in the storage unit 150. It is determined whether or not the current degree of long-term care is improved (lowered) than the above-mentioned past degree of long-term care by comparing with the degree of long-term care required. If the record of the same care recipient is not registered (step S510: NO), the input data cannot be used for the learning process, so the process in FIG. 5 ends.

As of July 2016, the degree of long-term care required in Japan consists of seven levels: support required 1, support required 2, long-term care required 1, long-term care 2, care required 3, long-term care required 4, and long-term care required 5. It is judged that the degree of need for long-term care is low, and the latter is higher in need of care. Therefore, when the degree of long-term care shifts from the latter classification to the former classification, for example, the past long-term care required level of a certain person to be cared for is "care required 3", but the long-term care required level included in the input data is "needed" When it is "Nursing care 2", it is determined that the degree of need for long-term care of the care recipient has improved.

This determination result is obtained in the record including the past care recipient status and the past care plan used in the above evaluation among the records of the same care recipient included in the storage unit 150. Is registered as the value of (step S530).

After that, the probability calculation unit 130 counts the number of records in which the value of the item “change in care required” is newly registered in the storage unit 150 (step S540). As a result, when it is determined that a predetermined number or more of data are newly registered (step S540: YES), the probability calculation unit 130 obtains the probability that the care-requiring degree of the care recipient is improved (step S550). For example, the probability calculation unit 130 includes the total number of records stored in the storage medium 150 that have the same care plan used for the care recipient and the care plan, and the degree of care required included in the above records. The number of records determined to have improved is obtained, and the ratio of the number of records determined to have improved the degree of long-term care to the total number of records is calculated. When it is not determined that a predetermined number or more of data are newly registered (step S540: NO), the process in FIG. 5 ends.

After that, the process change unit 140 performs a learning process to change the care plan output support model (step S560).

For example, the processing change unit 140 makes the inferred value output from the output unit corresponding to the care plan when the situation of the care recipient is input to the plan output network described above be the obtained ratio. , Change the above plan output network.

Specifically, the processing change unit 140 obtains a difference (error) between the inferred value output by the plan output network when the situation of the long-term care recipient is input and the calculated ratio. Next, the processing change unit 140 has weights and biases (y1, y2, y3 in the above-described example of FIG. 4) possessed by either the intermediate unit and the output unit of the plan output network so that the error is reduced. , Y4 and b) change one or more values.

The method of changing the evaluation network and the plan output network is not particularly limited, and a known method such as a gradient descent method using an error backpropagation method can be used. Further, the type of the error function used to express the above error in the gradient descent method is not particularly limited, and known equations such as the square error and the cross entropy equation can be used.

With such a configuration, the care plan output by the care plan output unit 110 by the processing including the execution of the care plan output support model including the plan output network has a high probability of improving the care-requiring degree of the care recipient. It becomes a thing. In particular, by inputting a large amount of input data and performing learning processing many times, the above combination having a high probability of improving the degree of long-term care for a long-term care recipient having an arbitrary long-term care status is configured. A care plan can be selected and provided to the care recipient.

In addition, by having a computer assist in the creation of a care plan, the objectives of improving the efficiency of creating a care plan and homogenizing the created care plan can be achieved.

In addition, in the conventional care plan creation method, one of the factors hindering the improvement of the quality of the created care plan is whether or not the care recipient who is provided with the care plan included in the care plan is lively. It can be mentioned that the quality of the selection of the care plan was evaluated by the judgment criteria such as, which are easily changed by the subjectivity of the observer. On the other hand, if the care plan is evaluated by an evaluation standard that can objectively determine the size of whether or not the degree of care required is improved, the subjectivity of the observer that affects the evaluation of the care plan can be eliminated as much as possible. ..

In addition, when the care plan output support model is configured to include a multi-layer neural network as in the present embodiment, when a specific care plan is provided among the survey items included in the input care recipient's situation. The care plan output support model can be changed so that the probability that the above-mentioned specific care plan is output is higher when the one having the characteristic that the degree of care required is likely to be improved is input. .. Therefore, even when the situation of a similar care recipient having the above-mentioned characteristics is input, the above-mentioned specific care plan is output with high probability, so that the care plan output unit 110 is not included in the teacher data. It is possible to output a care plan with a high probability that the degree of care required will improve even for the situation of the care recipient (or the number of data is small).

[Second Embodiment]
FIG. 6 is a flowchart showing a specific process included in the learning process (step S200) in another embodiment of the present invention. This embodiment is different from the first embodiment in that a care plan evaluation model, which is a probabilistic inference model different from the care plan output support model, is used to change the care plan output support model by learning processing. Since the other processes and the system configuration are the same as those in the first embodiment, duplicate description will be omitted.

The care plan evaluation model is a probabilistic inference model that outputs the probability that the degree of care required will improve by implementing the care plan for the input of the combination of the care recipient's situation and the care plan, and is changed by the learning process. It is configured to be possible. The care plan evaluation model includes, for example, a neural network, a Boltzmann machine and a decision tree.

Hereinafter, the configuration when the care plan evaluation model includes a neural network (hereinafter, also simply referred to as “evaluation network”) will be illustrated. At this time, the care plan evaluation model has an input layer composed of a plurality of input units, one or a plurality of intermediate layers each composed of a plurality of intermediate units, and an output layer composed of one output unit. The evaluation network is a multi-layered neural network having N layers having a plurality of intermediate layers from the input layer side to the output layer side, and the intermediate unit of each intermediate layer is the intermediate layer of the previous layer. The information received from one or more intermediate units (or input units) included in is output to one or more intermediate units (or output units) included in the intermediate layer of the next layer. The number of layers (N) possessed by the evaluation network may be 3 or more, and can be arbitrarily determined according to the situation of the care recipient and the number of combinations with the care plan. For example, the number of layers (N) can be 3 or more and 25 or less, preferably 5 or more and 20 or less, and more preferably 8 or more and 16 or less.

In the evaluation network, the input items to the input layer are a combination of the situation of the care recipient and the care plan, and the inferred value output from the output layer is the probability that the degree of care required is improved with respect to the above combination. It can have the same configuration as the plan output network described above.

At this time, the probability calculation unit 130 is among the records stored in the storage medium 150 after it is determined in step S540 described above that a sufficient amount of data is newly registered in the storage unit 150 (step S540: YES). , The total number of records with the same care plan used for the care recipient and the care plan, and the number of records included in the above records that are judged to have improved the degree of care required, and relative to the total number of the above records. , The ratio of the number of records determined to have improved the degree of long-term care required is calculated (step S550). After that, the processing change unit 140 changes the evaluation network so that the above ratio is output when the above combination is input to the care plan evaluation model (step S570).

Specifically, the processing change unit 140 obtains a difference (error) between the probability that the degree of need for nursing care output by the evaluation network improves when the combination is input and the calculated ratio. Next, the processing change unit 140 changes one or more values of the weight and the bias possessed by either the intermediate unit and the output unit of the evaluation network so that the error becomes small.

After that, the process change unit 140 performs the same process as in the first embodiment, and when the status of the care recipient is input to the plan output network, the inferred value output by each output unit is the status of the care recipient. The plan output network is changed so that the degree of need for long-term care output from the evaluation network is improved when the combination with the care plan corresponding to the output unit is input (step S560-2).

The method of changing the evaluation network and the plan output network is not particularly limited, and a known method such as a gradient descent method using an error backpropagation method can be used. Further, the type of the error function used to express the above error in the gradient descent method is not particularly limited, and known equations such as the square error and the cross entropy equation can be used.

With such a configuration, the probability that the degree of care required is improved, which is more accurate than the “ratio of records with improved degree of care required” in the first embodiment, is the target output when the care plan output support model is changed. It can be used as a value. Therefore, the recommended care plan output from the care plan output support model can be considered to have a higher probability of improving the degree of care required.

With such a configuration, it is possible to predict the prognosis of the care recipient by extracting and using only the output from the evaluation network.

[Third Embodiment]
FIG. 7 is a block diagram showing a configuration of a care plan creation support system for still another embodiment of the present invention. The care plan creation support system 700 according to the present embodiment includes a care plan change unit 710 that changes the care plan displayed on the display unit 160.

The care plan change unit 710 is provided with various operation keys such as a numeric keypad and a start key, an audio receiver, or a video receiver such as a camera, and accepts various input operations by a care manager to change the output care plan. Then, the generated input data is transmitted to the care plan output unit 110. The care plan output unit 110 receives the input data and changes the output care plan. Although FIG. 7 shows an example in which the care plan changing unit 710 is configured separately from the display unit 160, for example, the display unit 160 has an operation portion integrally configured on the display screen, such as a touch panel. Alternatively, it may be configured by being connected to a hard key such as a button or a key, and at the same time function as a care plan change unit 710.

The modified care plan is registered in the storage unit 150 as in the first and second embodiments.

At the same time, the display unit 160 may connect to the server of the long-term care provider and display information such as availability and availability of the long-term care service included in the output care plan. With such a configuration, the care manager can change the care plan from the display unit 160 when it is determined that the displayed nursing care service cannot be reserved.

Further, the care plan change unit 710 may be configured to be able to arrange a long-term care service by connecting to the server of the long-term care provider and inputting from the touch panel or the hard key.

[Fourth Embodiment]
FIG. 8 is a block diagram showing a configuration of a care plan creation support system for still another embodiment of the present invention. The care plan creation support system 800 according to the present embodiment has a cloud server 810 which is a virtualization server that cooperates with various devices via the Internet. The cloud server 810 includes a program for executing processing by the above-mentioned care plan output unit 110, determination unit 120, probability calculation unit 130, and processing change unit 140, data stored in a storage medium such as the storage unit 150, and the like. Some or all of is preserved.

The cloud server 810 is owned, managed and operated by an operating company that operates a long-term care business or a contractor entrusted by the operating company (hereinafter, also simply referred to as "operating company, etc. 830"). The operating company or the like 830 accesses the cloud server 810 from one or more processing units 835 such as a PC that can connect to the cloud server 810, and causes the processing unit 835 to execute the above-described processing.

At this time, the input unit 170 can be a device (for example, a smartphone, a PC, a TV, etc.) that can be connected to the Internet owned by a group 820 such as an assessment business operator that assesses the long-term care recipient. The input data input from the device as the input unit 170 is transmitted from the device to the cloud server 810. At this time, it is preferable that the input unit 170 is configured to automatically delete personal information such as the name of the care recipient, the date of birth, and the status of the care recipient after transmission.

At this time, the display unit 160 can be a device (for example, a smartphone, a PC, a TV, etc.) that can be connected to the Internet owned by a group 840 such as a care manager who creates a care plan. The care plan output by the care plan output unit 110 is transmitted to and displayed on the above-mentioned device as the display unit 160. When the group 840 modifies the care plan from the display unit 160, the modified data is transmitted to the cloud server 810.

The group 820 and the group 840 may be the same business operator or individual.

It should be noted that each of the above embodiments is merely an example of embodiment in carrying out the present invention, and the technical scope of the present invention should not be construed in a limited manner by these. That is, the present invention can be implemented in various forms without departing from its gist or its main features.

For example, in each of the above embodiments, data in which the degree of care required changes by one level and data in which the degree of care required changes by two or more stages are treated in the same manner as data with improved degree of care required, but it is required depending on the past care plan. The degree of improvement may be weighted according to how many times the degree of care has improved and used in the learning process.

Further, in each of the above embodiments, the storage unit 150 is described as a component incorporated in the care plan creation support system, but the storage unit 150 is configured independently of the care plan creation support system to be a care plan creation support system. The data for performing each of the above processes may be configured to be referenceable by connecting from. Also at this time, in the fourth embodiment, the storage unit 150 may be stored in the cloud server.

Further, in each of the above embodiments, the result of the determination by the determination unit 120 is registered in the storage unit 150, and the probability calculation unit 130 newly calculates the probability that the degree of long-term care is improved by using the registered determination result. However, the past care-requiring degree and the care-requiring degree included in the input data are registered in the storage unit 150, and the determination unit 120 compares the above two care-requiring degrees registered in the storage unit 150 and described above. The change in the degree of long-term care may be determined. At this time as well, the change in the degree of need for long-term care as a result of the long-term care provided using the care plan may be separately registered in the storage unit 150.

Further, in the third embodiment, the care plan output unit 110 outputs a plurality of care plans as recommended care plans, and one care plan selected by the care manager is input from the care plan change unit 710. , The selected and input care plan may be registered in the storage unit 150.

Further, in each of the above embodiments, the display unit 160 is described as a smartphone or tablet PC that can be carried by the care manager, but it is recommended that the care manager is displayed on the screen as a desktop PC or the like installed in a business establishment or the like. You may create a care plan while checking the care plan. The output unit 160 may not only display the care plan on the screen but also read it out as voice.

This application is an application claiming priority based on Japanese Application No. 2016-155859 filed on August 8, 2016, and the contents described in the claims, specification and drawings of the application are the present. Incorporated in the application.

According to the care plan creation support system of the present invention, a care plan with a high probability of improving the degree of long-term care can be output according to the situation of each individual care recipient, and the accuracy of the output is improved by learning processing. be able to. Therefore, if the care plan created by using the care plan creation support system of the present invention is provided to the long-term care recipient, the degree of care required of the long-term care recipient can be expected to be improved. If the degree of long-term care required by long-term care recipients improves, it is expected that long-term care benefit costs will decrease and long-term care turnover will decrease, and the increase in long-term care will reduce the financial and social burden.

100, 700, 800 Care plan creation support system 110 Care plan output unit 120 Judgment unit 130 Probability calculation unit 140 Processing change unit 150 Storage unit 160 Display unit 170 Input unit 710 Care plan change unit 810 Cloud server 820 Group 830 Operating company, etc. 835 Processing unit 840 group

Claims (5)

An input reception unit that accepts input of the results of assessments performed on the care recipient,
It has a storage unit that stores a plurality of candidate care plans and a probability output unit that outputs the probability that the situation of the long-term care recipient is improved by implementing the candidate care plan for the long-term care recipient.
The probability output unit inputs the result of the assessment and the candidate care plan into the learning-processed stochastic inference model, and outputs the probability that the care-requiring degree of the care recipient will improve.
Care plan creation support system. The care plan creation according to claim 1, wherein the probability output unit calculates the ratio of the care recipient whose degree of care required is improved by the candidate care plan to the number of implementations of the candidate care plan as the probability of improvement. Support system. A determination unit that determines a change in the degree of need for long-term care of the long-term care recipient after executing the candidate care plan for the long-term care recipient.
It has a processing change unit that changes the stochastic inference model based on the result of the determination by the determination unit.
The care plan creation support system according to claim 1 or 2. Based on the result of the determination by the determination unit, it has a probability calculation unit that newly calculates the probability that the care-requiring degree of the care recipient will be improved by the candidate care plan.
The processing change unit changes the stochastic inference model so that the probability that the care-requiring degree of the care recipient, which is output by the stochastic inference model, is improved is close to the probability calculated by the probability calculation unit.
The care plan creation support system according to claim 3. The care plan creation support system according to any one of claims 1 to 4, wherein the probability output unit outputs the probability that the required care level of the care recipient is improved.

Images