Support device, support method, and program

Abstract

To provide a support device enabling whether a patient problem is appropriately set, to be efficiently confirmed.SOLUTION: A support device 1 comprises a prediction unit 2, a calculation unit 4, and an output unit 6. The prediction unit 2 predicts a patient problem to be addressed in order to achieve the goal of the patient. The calculation unit 4 calculates the degree of coincidence between a predicted patient problem which is a predicted patient problem and an actual patient problem which is the actually addressed patient problem. The output unit 6 performs control in such a manner that an alert is output when the calculated degree of coincidence is less than a predetermined first threshold.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an assistance device, an assistance method, and a program. [Background technology]

[0002] In facilities (such as rehabilitation hospitals) where activities (hereinafter simply referred to as "activities") aimed at improving abilities, including physical functions, such as rehabilitation, are carried out, often by therapists such as physical therapists under the direction of a doctor. In this regard, Patent Document 1 discloses a medical information display device that supports updating of a rehabilitation plan to reflect a patient's daily condition. The device disclosed in Patent Document 1 acquires rehabilitation information that indicates the patient's condition when undergoing rehabilitation, and daily information that indicates the patient's condition when not undergoing the rehabilitation. The device disclosed in Patent Document 1 also outputs a rehabilitation screen that displays the rehabilitation information and the daily information in chronological synchronization. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2020-184219 Summary of the Invention [Problem to be solved by the invention]

[0004] When providing rehabilitation or other treatment (hereinafter sometimes simply referred to as "treatment") to a patient, therapists often determine the content of rehabilitation after setting patient tasks to be resolved in order to achieve the patient's activity goals, for example under the direction of a doctor. If the patient tasks are not set appropriately, the necessary treatment to achieve the goals may not be provided, and the goals may not be achieved. Therefore, it is necessary to confirm whether the patient tasks are set appropriately.

[0005] The purpose of the present disclosure has been made to solve such problems, and is to provide an assistance device, assistance method, and program that can efficiently check whether patient tasks are set appropriately. [Means for solving the problem]

[0006] The support device according to the present disclosure includes a prediction means for predicting patient issues that should be addressed in order to achieve the patient's goals, a calculation means for calculating the degree of agreement between actual patient issues, which are patient issues that have actually been addressed for the patient, and predicted patient issues, which are predicted patient issues, and an output means for controlling so that an alert is output when the degree of agreement is less than a predetermined first threshold.

[0007] In addition, the support device disclosed herein predicts patient issues that should be addressed in order to achieve the patient's goals, calculates the degree of agreement between actual patient issues, which are patient issues that have actually been addressed for the patient, and predicted patient issues, which are predicted patient issues, and performs control so that an alert is output when the degree of agreement is less than a predetermined first threshold.

[0008] In addition, the program disclosed herein causes a computer to execute the following steps: predicting patient issues that should be addressed in order to achieve the patient's goals; calculating the degree of agreement between actual patient issues, which are patient issues that have actually been addressed for the patient, and predicted patient issues, which are predicted patient issues; and performing control so that an alert is output when the degree of agreement is less than a predetermined first threshold. [Effects of the Invention]

[0009] According to the present disclosure, it is possible to provide an assistance device, an assistance method, and a program that can efficiently check whether a patient task has been set appropriately. [Brief explanation of the drawings]

[0010] [Figure 1]1 is a diagram illustrating an overview of a support device according to an embodiment of the present disclosure. [Figure 2] 1 is a flowchart illustrating an assistance method executed by the assistance device according to an embodiment of the present disclosure. [Figure 3] 1 is a diagram illustrating a support system according to a first embodiment. [Figure 4] FIG. 1 is a diagram illustrating a configuration of a support device according to a first embodiment. [Figure 5] FIG. 2 is a diagram for explaining the timing of processing performed by the support device according to the first embodiment. [Figure 6] FIG. 2 is a diagram illustrating predicted patient tasks and actual patient tasks according to the first embodiment. [Figure 7] FIG. 2 is a diagram illustrating patient information according to the first embodiment. [Figure 8] 4 is a flowchart showing a support method executed by the support device according to the first embodiment. [Figure 9] 10 is a flowchart illustrating a first example of a method for calculating a prediction agreement executed by a prediction agreement calculation unit according to the first embodiment. [Figure 10] 10 is a flowchart illustrating a second example of the method for calculating a prediction agreement executed by the prediction agreement calculation unit according to the first embodiment. [Figure 11] 4 is a flowchart showing a method of determining the type of an alert executed by an alert type determining unit according to the first embodiment; [Figure 12] 4 is a diagram illustrating an example of an alert output under the control of an alert output unit according to the first embodiment; FIG. [Figure 13] 10 is a flowchart showing a support method executed by the support device according to the second embodiment. [Figure 14] FIG. 10 is a diagram illustrating an example of an alert output under the control of an alert output unit according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] (Summary of Embodiments of the Present Disclosure) Prior to describing the embodiments of the present disclosure, an overview of the embodiments of the present disclosure will be described. Fig. 1 is a diagram illustrating an overview of a support device 1 according to the embodiments of the present disclosure. The support device 1 is, for example, a computer such as a server or a personal computer.

[0012] The support device 1 performs processing related to setting tasks (patient tasks) to be solved in order to achieve the goal of activities (capacity improvement activities) aimed at improving the capabilities of a patient, such as rehabilitation. Note that in the following explanation, an example will be described in which the capacity improvement activity is treatment such as rehabilitation, but the capacity improvement activity is not limited to rehabilitation (treatment).

[0013] The support device 1 has a prediction unit 2, a calculation unit 4, and an output unit 6. The prediction unit 2 functions as a prediction means. The calculation unit 4 functions as a calculation means. The output unit 6 functions as an output means.

[0014] 2 is a flowchart showing a support method executed by the support device 1 according to an embodiment of the present disclosure. The prediction unit 2 predicts a patient task that should be addressed in order to achieve the patient's goal (step S2). The calculation unit 4 calculates a degree of agreement between a predicted patient task, which is a predicted patient task, and an actual patient task, which is a patient task that has actually been addressed for the patient (step S4). The output unit 6 performs control so that an alert is output when the calculated degree of agreement is less than a predetermined first threshold (step S6).

[0015] Here, the output alert corresponds to a warning that the patient task may not be set appropriately and therefore the goal may not be achieved. The output unit 6 may also perform control to display the alert on a display device (user interface) provided in the support device 1. Alternatively, the output unit 6 may perform control to display the alert on a device other than the support device 1, such as a user terminal. The output unit 6 may also perform control to output the alert by voice or the like.

[0016] For example, in rehabilitation hospitals, in order to optimize the length of hospital stays, a discharge date and a target patient condition at the time of discharge (hereinafter simply referred to as "target") are set, and rehabilitation is carried out toward that target. However, there are cases where the target cannot be achieved as planned, resulting in "discharge delays," in which patients are unable to be discharged at the set time. Therefore, it is desirable to reduce discharge delays.

[0017] One of the causes of delayed discharge is the inappropriate setting of patient issues that need to be resolved in order to achieve goals. Generally, in rehabilitation hospitals, rehabilitation is provided to patients by therapists such as physical therapists under the direction of a doctor. Therapists create rehabilitation plans, for example, according to the following flow. That is, therapists implement rehabilitation, for example, according to the following flow, under the guidance of a doctor. In the rehabilitation flow described below, goals are set, and then rehabilitation to achieve those goals is implemented. That is, therapists confirm the rehabilitation goals (rehabilitation objectives) established at the time of the patient's admission, and then implement rehabilitation to achieve those goals. (1) Setting goals based on the patient's condition and needs (2) Identifying patient issues that need to be resolved to achieve the goal (3) Deciding on rehabilitation to solve the problem (4) Rehabilitation

[0018] For example, if a patient's goal is set as "to be able to go up and down stairs," the factors that hinder going up and down stairs are set as the patient's challenge. In this example, if the patient is unable to go up and down stairs due to a decline in lower limb function, "lower limb function" is set as the patient's challenge. Rehabilitation to improve lower limb function is then set. Then, by implementing this rehabilitation, the therapist will address the patient's challenge of "lower limb function."

[0019] Here, since the patient's condition changes daily, the above steps (2) to (4) are repeatedly executed daily depending on the patient's condition. Since rehabilitation is carried out in the above order, if the "patient task setting" is not done appropriately, it will be difficult to achieve the goal. This is because if the "patient task setting" is not done appropriately, there is a risk that the patient will not be provided with the rehabilitation necessary to achieve the goal. Therefore, it is important that the patient task setting is done appropriately.

[0020] Setting patient tasks is generally based on experience, making it difficult for inexperienced therapists. Furthermore, therapists are busy. Therefore, even if one therapist sets a patient task to achieve a patient's rehabilitation goals within the scope of a doctor's instructions, it is difficult for another therapist to check daily whether the patient task is optimal for achieving the patient's rehabilitation goals. Furthermore, for example, if another therapist (e.g., a team leader) needs to check whether a patient task set by an inexperienced therapist is optimal for achieving the patient's rehabilitation goals, they may check the patient task set in the electronic medical record. In this case, the other therapist must operate the electronic medical record, which is time-consuming. Therefore, there is a need for an automatic system that can efficiently check whether patient tasks are set appropriately.

[0021] In contrast, the support device 1 according to the present disclosure is configured as described above, and can output an alert when the degree of agreement between the predicted patient task and the actual patient task is low, i.e., when there is a large difference between the predicted patient task and the actual patient task. Here, the actual patient task is, for example, a patient task that is addressed in accordance with the patient task set by a therapist under the direction of a doctor. Furthermore, the predicted patient task is likely to be appropriate for the patient undergoing rehabilitation. Therefore, when there is a large difference between the predicted patient task and the actual patient task, it is possible that the set patient task is inappropriate.

[0022] Therefore, the support device 1 according to the present disclosure can automatically output an alert when there is a possibility that the set patient task is inappropriate. This allows the support device 1 according to the present disclosure to efficiently check whether the patient task is set appropriately. In other words, the support device 1 according to the present disclosure can support the evaluation of the set patient task. Note that even when using the support method executed by the support device 1 and the program that executes the support method, it is possible to efficiently check whether the patient task is set appropriately.

[0023] As described above, according to the present embodiment, the therapist can easily confirm that the set patient task may not be appropriate. In such a case, the therapist can take measures to set an appropriate patient task. For example, it can be confirmed that the set patient task may not be optimal for achieving the patient's rehabilitation goals, even if it is within the scope of the doctor's instructions. The therapist can then take measures to set an appropriate patient task within the scope of the doctor's instructions. As a result, the goals can be achieved as planned, and delays in discharge are reduced.

[0024] (Embodiment 1) Hereinafter, embodiments will be described with reference to the drawings. For clarity of explanation, the following description and drawings have been omitted and simplified as appropriate. In addition, the same elements in each drawing are designated by the same reference numerals, and duplicate explanations have been omitted as necessary.

[0025] 3 is a diagram showing a support system 50 according to the first embodiment. The support system 50 includes one or more user terminals 60 and a support device 100. The support device 100 corresponds to the support device 1 shown in FIG. 1. The user terminal 60 and the support device 100 are connected to each other via a wired or wireless network 52 so as to be able to communicate with each other.

[0026] The assistance device 100 is a computer such as a server or a personal computer. The assistance device 100 provides assistance in setting a patient task to be solved in order to achieve a goal in rehabilitation (activity for improving abilities). Specifically, the assistance device 100 performs control so that an alert is output when there is a possibility that the patient task has not been set appropriately.

[0027] The user terminal 60 is, for example, a computer. The user terminal 60 is, for example, a personal computer (PC) of a user such as a therapist, or a mobile terminal such as a tablet terminal or a smartphone. The user may use the user terminal 60 to input patient information, which is information about the patient. In this case, the user terminal 60 accepts the patient information via an input device. The user terminal 60 then transmits the patient information to the assistance device 100. The assistance device 100 stores the patient information. The patient information will be described later. The user may also use the user terminal 60 to set determined goals, patient tasks, and rehabilitation. In this case, the user terminal 60 transmits information indicating the set goals, patient tasks, and rehabilitation to the assistance device 100. The user may also use the user terminal 60 to input daily information about the patient's condition, the patient tasks being addressed (actual patient tasks), and the rehabilitation performed. In this case, the user terminal 60 transmits information indicating the patient's condition, the actual patient tasks, and the rehabilitation performed to the assistance device 100.

[0028] Furthermore, the user terminal 60 may receive a command to output an alert under the control of the assistance device 100. In this case, the user terminal 60 causes an output device (user interface) such as a display device provided in the user terminal 60 to output (display) an alert corresponding to the command received from the assistance device 100.

[0029] The support device 100 may output an alert to a user interface provided in the support device 100. The alert may be displayed as an image on a display. Alternatively, the alert may be output by a lamp. Alternatively, the alert may be output by sound. Alternatively, the alert may be output by vibration of the user terminal 60 or the like.

[0030] 4 is a diagram showing the configuration of the support device 100 according to the first embodiment. The support device 100 has, as its main hardware components, a control unit 102, a storage unit 104, a communication unit 106, and an interface unit 108 (IF; Interface). The control unit 102, the storage unit 104, the communication unit 106, and the interface unit 108 are interconnected via a data bus or the like. The user terminal 60 shown in FIG. 3 may also have the hardware configuration shown in FIG. 4.

[0031] The control unit 102 is a processor such as a CPU (Central Processing Unit). The control unit 102 functions as an arithmetic device that performs control processing, arithmetic processing, etc. The storage unit 104 is a storage device such as a memory or a hard disk. The storage unit 104 is, for example, a ROM (Read Only Memory) or a RAM (Random Access Memory). The storage unit 104 has a function to store control programs, arithmetic programs, etc. executed by the control unit 102. The storage unit 104 also has a function to temporarily store processing data, etc. The storage unit 104 may include a database.

[0032] The communication unit 106 performs processing necessary for communicating with the user terminal 60 (and other devices) via the network 52. The communication unit 106 may include a communication port, a router, a firewall, etc. The interface unit 108 (IF; Interface) is, for example, a user interface (UI). The interface unit 108 has an input device such as a keyboard, a touch panel, or a mouse, and an output device such as a display or a speaker. The interface unit 108 accepts data input operations by a user (operator) and outputs information to the user.

[0033] The support device 100 according to the first embodiment has, as its components, a patient information storage unit 110, a patient problem prediction unit 120, an actual patient problem acquisition unit 130, a prediction agreement calculation unit 140, an alert output determination unit 150, an alert type determination unit 160, and an alert output unit 170. The alert type determination unit 160 also has a past comparison unit 162 and a target comparison unit 164.

[0034] The patient information storage unit 110 functions as a patient information storage means. The patient problem prediction unit 120 corresponds to the prediction unit 2 shown in Figure 1. The patient problem prediction unit 120 functions as a patient problem prediction means (prediction means). The actual patient problem acquisition unit 130 functions as an actual patient problem acquisition means (acquisition means). The prediction agreement calculation unit 140 corresponds to the calculation unit 4 shown in Figure 1. The prediction agreement calculation unit 140 functions as a prediction agreement calculation means (calculation means, agreement calculation means).

[0035] The alert output determination unit 150 functions as an alert output determination means (output determination means). The alert type determination unit 160 functions as an alert type determination means (type determination means). The alert output unit 170 corresponds to the output unit 6 shown in FIG. 1. The alert output unit 170 functions as an alert output means (output means). Furthermore, the past comparison unit 162 functions as a past comparison means. The target comparison unit 164 functions as a target comparison means.

[0036] Each of the above-described components can be realized, for example, by executing a program under the control of the control unit 102. More specifically, each component can be realized by the control unit 102 executing a program stored in the storage unit 104. Alternatively, each component may be realized by recording the necessary program on an arbitrary non-volatile recording medium and installing it as needed. Each component may not necessarily be realized by software programs, but may be realized by any combination of hardware, firmware, and software. Each component may also be realized using a user-programmable integrated circuit, such as an FPGA (field-programmable gate array) or a microcomputer. In this case, a program consisting of each of the above-described components may be realized using this integrated circuit. The above also applies to other embodiments described later. Specific functions of each component will be described later.

[0037] Using the above-described components, the support device 100 predicts a patient task that should be addressed in order to achieve the patient's goal. The support device 100 also calculates the degree of agreement between the predicted patient task and the actual patient task. Furthermore, the support device 100 determines whether the calculated degree of agreement is less than a predetermined first threshold (alert output determination), and performs control so that an alert is output when the degree of agreement is less than the first threshold.

[0038] FIG. 5 is a diagram illustrating the timing of processing performed by the assistance device 100 according to the first embodiment. FIG. 5 shows the passage of time (number of days elapsed) during rehabilitation. If a patient is hospitalized, the passage of time during rehabilitation may correspond to the number of days of hospitalization. In the example of FIG. 5, a patient task is predicted for each period Ta from the start of rehabilitation. The time point at which the patient task is predicted is defined as time point A. Furthermore, for each time point A (each period Ta), an evaluation may be made as to whether or not the goal can be achieved as planned. The period Ta is a predetermined period. The period Ta is, for example, 10 days, but is not limited to this. Furthermore, an alert output determination is made after a period Tb has elapsed from time point A. This time point after the period Tb has elapsed from time point A is defined as the alert output determination time. In other words, it is determined at the alert output determination time whether or not to output an alert, and if an alert is to be output, the alert is output at the alert output determination time. The period Tb is a predetermined period that is shorter than the period Ta. The period Tb is, for example, 7 days, but is not limited to this. Furthermore, the period Tc is the period from the time when the alert output is determined to the next time A. Here, Tc=Ta-Tb.

[0039] At time A (first time point), the support device 100 predicts a patient task to be addressed during a period Ta (first period) from time A. That is, the support device 100 calculates a predicted patient task for the period Ta starting from time A. Meanwhile, at the alert output determination time, the support device 100 acquires patient tasks (actual patient tasks) that were actually addressed during a period Tb (second period) starting from time A. Then, at the alert output determination time, the support device 100 compares the predicted patient task with the actual patient task to calculate the degree of agreement (prediction agreement) between the actual patient task and the predicted patient task. Then, the support device 100 determines whether the prediction agreement is less than a predetermined threshold. If the prediction agreement is less than the threshold, the support device 100 performs control so as to output an alert indicating that the patient task may not be appropriately set and therefore the goal may not be achieved.

[0040] With this configuration, the support device 100 can periodically determine whether to output an alert. Therefore, it is possible to periodically and efficiently check whether the patient task has been set appropriately. Furthermore, since the alert output determination is performed after a period Tb, which is shorter than the period Ta, it is possible to check whether the patient task has been set appropriately before the period Ta has elapsed from the time point A. Therefore, when an evaluation is performed at each time point A to see whether the goal can be achieved as planned, it is possible to check whether the patient task has been set appropriately before the evaluation is performed.

[0041] Fig. 6 is a diagram illustrating predicted patient tasks and actual patient tasks according to the first embodiment. As illustrated in Fig. 6, the patient tasks (predicted patient tasks and actual patient tasks) are composed of multiple items. Here, as illustrated in Fig. 6, the items included in the patient tasks may be set for each of multiple categories.

[0042] The items included in the patient task may be classified into higher-level tasks and lower-level tasks. Here, the higher-level tasks and lower-level tasks are in a relationship of higher and lower concepts in the task. In other words, a lower-level task is a lower concept of a higher-level task, and a higher-level task is a higher concept of a lower-level task. The higher-level task indicates what movements are a challenge for the patient in daily life. In other words, the higher-level task indicates movements that the patient has difficulty with in daily life. Furthermore, the lower-level task relates to the cause of difficulty in performing the movements of daily life indicated by the higher-level task. The lower-level task indicates basic movements and functions of the body that are impaired (motor function, cognitive function, etc.). In other words, it is considered that the patient has difficulty fully performing the movements of the higher-level task because the movements and functions corresponding to the lower-level task have deteriorated.

[0043] Furthermore, the items included in the patient task may be classified into major and minor items. That is, the items included in the higher-level task and the lower-level task may be further classified into major and minor items. The major items indicate general tasks, and the minor items indicate more specific tasks. The major items indicate names of actions and names of bodily functions, and the minor items indicate specific cases in which the task indicated in the major item is caused.

[0044] The actual patient task illustrated in FIG. 6 includes, as higher-level task items, "Major Item: Walking, Subitem: Right Side," "Major Item: Walking, Subitem: Paralyzed Side," and "Major Item: IADL, Subitem: Housework." The actual patient task illustrated in FIG. 6 also includes, as lower-level task items, "Major Item: Balance, Subitem: Standing," "Major Item: Balance, Subitem: Sitting," and "Major Item: Upper Limb Function, Subitem: None." In this case, the patient (target patient) related to the actual patient task has difficulty with "walking (major item of the higher-level task)" on the "right side (subitem of the higher-level task)" and the "paralyzed side (subitem of the higher-level task)" in daily life. This is due to a decline in the function of "balance (major item of the lower-level task)" in "standing (subitem of the lower-level task)" and "sitting (subitem of the lower-level task)." The target patient also has difficulty with "IADL (major item of the higher-level task)" for "housework (subitem of the higher-level task)" in daily life. The cause of this is a decline in "upper limb function (a major item of the subtask)." As will be explained later, IADLs are instrumental activities of daily living, that is, applied daily activities, which require complex movements and judgment.

[0045] 6 includes, as higher-level task items, "major item: walking, minor item: right side," "major item: walking, minor item: paralyzed side," "major item: IADL, minor item: hobbies," "major item: IADL, minor item: housework," and "major item: IADL, minor item: transfers." The actual patient task illustrated in FIG. 6 includes, as lower-level task items, "major item: balance, minor item: standing position," "major item: balance, minor item: sitting position," "major item: cognitive function, minor item: cognitive function," "major item: standing up, minor item: none," and "major item: upper limb function, minor item: none."

[0046] FIG. 7 is a diagram illustrating patient information according to the first embodiment. The patient information may include basic information about the target patient, the patient's condition, and set goals. The patient information may also include set patient tasks and actual patient tasks. The basic information may indicate the attributes of the target patient. For example, the basic information may indicate the name, age, and gender of the target patient. The basic information may also include the expected length of hospitalization and the number of days of hospitalization to date.

[0047] The patient condition may also indicate the patient's current functional level, such as motor function and cognitive function. The functional level may be, for example, an ability score (ability level) for the patient's daily activities. The functional level may be, for example, a score (index) for ADL (Activities of Daily Living) or IADL (Instrumental Activities of Daily Living). The functional level may also be indicated by, for example, an evaluation score for each evaluation item in the FIM (Function Independence Measure). Alternatively, the functional level may be indicated by a score on the GCS (Glasgow Coma Scale) or JCS (Japan Coma Scale), or may be indicated in subjective words by the user.

[0048] In Figure 7, the goals include short-term goals and long-term goals. In the example of Figure 7, "improvement of pain in the left hip joint" and "independent walking within the ward" are set as short-term goals. In addition, "independent walking outdoors" and "ability to live independently" are set as long-term goals.

[0049] Here, the long-term goal refers to the patient's condition and the actions that the patient is capable of achieving immediately before discharge in the rehabilitation plan. In other words, the long-term goal represents the patient's desired condition at the time of discharge, as well as the patient's condition and the actions that the patient is capable of achieving in order to be able to be discharged to the patient's desired destination (e.g., home). On the other hand, the short-term goal refers to the patient's condition and the actions that the patient is capable of achieving at the midpoint of the planned hospitalization period in the rehabilitation plan. In other words, the short-term goal refers to the patient's condition and the actions that the patient is capable of achieving at the midpoint in order to achieve the long-term goal.

[0050] In the example of Figure 7, the patient's condition is shown by the current FIM. Specifically, each item indicating the patient's condition (such as "transfer from bed to wheelchair") is associated with an FIM score (degree of independence) as shown below. This indicates the degree of independence the patient currently has for each item. 7 points: Complete independence (including time and safety) 6 points: Modified independence (independent with the use of assistive devices) 5 points: Supervision (direction, prompting, and preparation required) 4 points: Minimal assistance (less than 25% assistance required) 3 points: Moderate assistance required (25% to 50% assistance required) 2 points: Maximum assistance (50% to 75% assistance required) 1 point: Full assistance (75% or more assistance required)

[0051] As shown in FIG. 7, the patient information may include predicted patient tasks predicted by the patient task prediction unit 120 (described later) and actual patient tasks that have been addressed. In the example of FIG. 7, the patient information includes "patient tasks addressed in the previous period" and "predicted patient tasks for the next period." For example, if the current time is the 20th day since the start of rehabilitation, the "patient tasks addressed in the previous period" indicate the patient tasks addressed from the 10th to 19th days. The "predicted patient tasks for the next period" indicate the patient tasks predicted to be addressed from the 20th to 29th days. In FIG. 5, the current time corresponds to "time A(n)," and the "previous period" corresponds to the period Ta from "time A(n-1)" to "time A(n)." The "next period" corresponds to the period Ta from "time A(n)" to "time A(n+1)."

[0052] In the example shown in FIG. 7, the "predicted patient tasks for the next period" are the same as those illustrated in FIG. 6. In addition, in the example shown in FIG. 7, the "patient tasks addressed in the previous period" include, as higher-level task items, "major item: walking, minor item: right side," "major item: walking, minor item: paralyzed side," "major item: IADL, minor item: toileting," and "major item: walking, minor item: left side." In addition, the "patient tasks addressed in the previous period" include, as lower-level task items, "major item: balance, minor item: standing," "major item: balance, minor item: sitting," and "major item: standing up, minor item: none." In this case, the target patient has difficulty "walking (major item of higher-level task)" on the "right side (minor item of higher-level task)," "paralyzed side (minor item of higher-level task)," and "left side (minor item of higher-level task)." The cause of this is a decline in the ability to "stand" (a minor item in a minor task) and "balance" (a major item in a minor task) when "sitting" (a minor item in a minor task). Furthermore, the target patient has difficulty with "IADL" (a major item in a major task) for "toileting" (a minor item in a major task) in daily life. The cause of this is a decline in "standing" (a major item in a minor task).

[0053] The support device 100 may perform control so that the patient information exemplified in FIG. 7 is displayed. In this case, the support device 100 may perform control so that the patient information exemplified in FIG. 7 is displayed for each period Ta. In this case, the support device 100 may also display the patient information independently of an alert. The support device 100 may also perform control so that the patient information is displayed on the display device (user interface) of the user terminal 60. The support device 100 may also perform control so that the patient information is displayed on the interface unit 108.

[0054] FIG. 8 is a flowchart showing a support method executed by the support device 100 according to the first embodiment. The flowchart shown in FIG. 8 can be executed for each patient (target patient) to be processed. The patient information storage unit 110 stores patient information (step S100). Specifically, the patient information storage unit 110 stores the patient information using the memory unit 104. The patient information storage unit 110 may store, for example, patient information received from the user terminal 60. The patient information storage unit 110 may also store the patient information exemplified in FIG. 7.

[0055] The patient problem prediction unit 120 predicts a patient problem (step S102). Specifically, the patient problem prediction unit 120 calculates, as a predicted patient problem (predicted patient problem), a patient problem that was actually addressed in the past for a patient (past patient) whose patient information is similar to that of the target patient. The patient problem prediction unit 120 may determine that the patient information of the target patient and the patient information of the past patient are similar if the similarity between the two is equal to or greater than a predetermined threshold. The similarity may be calculated, for example, by comparing the basic information of the target patient with the basic information of the past patient, comparing the goals of the target patient with the goals of the past patient, and comparing the patient condition of the target patient with the patient condition of the past patient. The similarity may also be calculated by comparing the recovery status of the target patient with the recovery status of the past patient.

[0056] 5, the patient problem prediction unit 120 calculates the predicted patient problem for a period Ta (first period) starting from time A (first time point). Specifically, for a past patient related to patient information similar to the patient information of the target patient at time A, the patient problem prediction unit 120 calculates, as the predicted patient problem, the patient problem that was actually addressed in the period Ta starting from the time point when the patient information of that past patient was obtained.

[0057] It should be noted that the patient problem prediction section 120 does not necessarily have to predict the patient problem at time point A. The patient problem prediction section 120 may predict the patient problem to be addressed in the period Ta at the time point of determining whether to output an alert. Even in this case, the patient problem prediction section 120 predicts the patient problem using patient information obtained at time point A, which is prior to the time point of determining whether to output an alert.

[0058] The actual patient task acquisition unit 130 acquires the actual patient task of the target patient (step S104). Specifically, the actual patient task acquisition unit 130 acquires the actual patient task input by the therapist. The actual patient task acquisition unit 130 may acquire (receive) the actual patient task from the therapist's user terminal 60. Alternatively, the actual patient task acquisition unit 130 may acquire (extract) the actual patient task stored in the support device 100. In this case, the actual patient task may be included in the patient information stored in the patient information storage unit 110. As described with reference to FIG. 5 , the actual patient task acquisition unit 130 acquires, at the time of determining whether to output an alert, the patient task (actual patient task) that was actually addressed during the period Tb (second period) starting from time A. Note that, in the course of daily work, therapists actually address the preset patient tasks and input the addressed patient tasks into the user terminal 60, etc. Therefore, the actual patient task may correspond to the set patient task.

[0059] The prediction agreement calculation unit 140 calculates the prediction agreement Ca, which is the degree of agreement between the actual patient task and the predicted patient task (step S110). Specifically, the prediction agreement calculation unit 140 compares the actual patient task with the predicted patient task at the alert output determination time shown in FIG. 5, and calculates the prediction agreement Ca. The prediction agreement Ca indicates the degree of agreement between the actual patient task and the predicted patient task. In other words, the prediction agreement Ca indicates the degree to which the actual patient task and the predicted patient task match. In other words, the prediction agreement Ca becomes small when the difference between the actual patient task and the predicted patient task is large, and becomes large when the difference between the actual patient task and the predicted patient task is small. An example of a method for calculating the prediction agreement Ca will be described below, but the method for calculating the prediction agreement Ca is not limited to the following example.

[0060] 9 is a flowchart showing a first example of a method (S110) for calculating the prediction agreement Ca executed by the prediction agreement calculation unit 140 according to the first embodiment. The prediction agreement calculation unit 140 calculates the total number Na1 of items included in the actual patient task (step S114A). In the example of FIG. 6, for example, the prediction agreement calculation unit 140 calculates the total number of items included in the actual patient task as Na1=6. Note that in this example, the patient task items to be compared when calculating the prediction agreement may be a combination of major and minor items. In other words, the patient task items to be compared when calculating the prediction agreement may correspond to a combination of major and minor items.

[0061] Next, the prediction agreement calculation unit 140 calculates the number Nb1 of items included in the actual patient task that are included in the predicted patient task (step S116A). In the example of FIG. 6, all items included in the actual patient task are included in the predicted patient task, so Nb1 = 6 (items). Next, the prediction agreement calculation unit 140 calculates Nb1 / Na1 as the prediction agreement Ca (step S118A). That is, in the first example, Ca = Nb1 / Na1. In other words, the prediction agreement calculation unit 140 calculates the ratio of the number Nb1 of items included in the actual patient task that are included in the predicted patient task to the total number Na1 of items included in the actual patient task as the prediction agreement Ca. In the example of FIG. 6, for example, the prediction agreement calculation unit 140 calculates the prediction agreement Ca as Ca = 6 / 6 = 1.0. In the first example, the prediction agreement Ca can be calculated using a simpler method than in the second example described below.

[0062] 10 is a flowchart showing a second example of the method (S110) for calculating the prediction agreement Ca executed by the prediction agreement calculation unit 140 according to the first embodiment. The prediction agreement calculation unit 140 calculates the frequency with which each item included in the actual patient task was addressed during the period Tb (step S112B). The frequency of each item increases as it is addressed more frequently during the period Tb.

[0063] There are various methods for calculating the frequency of any item A included in the actual patient task. For example, frequency = (number of days item A was addressed in period Tb) / (number of days in period Tb). In other words, the frequency of item A may be the ratio of the number of days item A was addressed to the total number of days in period Tb. In this case, for example, if period Tb is 7 days long and item A was addressed for 4 days in period Tb, the frequency of item A is calculated as 4 / 7 ≒ 0.57.

[0064] Alternatively, the frequency of item A may be calculated as follows: Frequency = (number of times item A was addressed in period Tb) / (total number of times in period Tb). In other words, the frequency of item A may be the ratio of the number of times item A was addressed to the total number of times in period Tb. Note that the "total number of times in period Tb" is the total number of times any item was addressed in period Tb. For example, if period Tb has 7 days, and some patient task is addressed a total of 9 times each day over the 7 days, the "total number of times in period Tb" is 9 x 7 = 63 times. Then, suppose item A was addressed 21 times in period Tb. In this case, the frequency of item A is calculated as 21 / 63 ≒ 0.33.

[0065] Next, the prediction coincidence calculation unit 140 calculates the total number Na2 of high-frequency items (step S114B). Here, high-frequency items are items whose frequency is equal to or greater than a predetermined threshold Th2. For example, if the total number of items in the actual patient problem is 20 and the frequency of five of those items is equal to or greater than the threshold Th2, the total number of high-frequency items (first items) is calculated as Na2=5 (items). Here, the threshold Th2 (second threshold) is, for example, Th2=0.5, but is not limited to this. Furthermore, in the above example of the frequency calculation method, if frequency=(number of days item A was addressed in period Tb) / (number of days in period Tb), Th2=0.5 may be used. On the other hand, if frequency=(number of times item A was addressed in period Tb) / (total number of times in period Tb), Th2=0.3 may be used.

[0066] In the above example, the frequency of item A is defined as the ratio of the number of days (or number of times) that item A was worked on to the total number of days (or number of times) in period Tb, but this is not limited to this. If the number of days in period Tb is predetermined, the frequency may simply be the number of days that item A was worked on in period Tb. In this case, for example, threshold Th2 may be Th2=4 (days). Alternatively, if the total number of times in period Tb is predetermined, the frequency may simply be the number of times that item A was worked on in period Tb. In this case, for example, threshold Th2 may be Th2=20 (times).

[0067] Next, the prediction agreement calculation unit 140 calculates the number Nb2 of the high-frequency items included in the predicted patient task (step S116B). For example, if two of the five high-frequency items are included in the predicted patient task, Nb2 = 2 (items). Next, the prediction agreement calculation unit 140 calculates Nb2 / Na2 as the prediction agreement Ca (step S118B). That is, in the second example, Ca = Nb2 / Na2. In other words, the prediction agreement calculation unit 140 calculates the ratio of the number Nb2 of high-frequency items included in the predicted patient task to the total number Na2 of high-frequency items included in the actual patient task whose frequency of being addressed in the period Tb is equal to or greater than the threshold Th2, as the prediction agreement Ca. In the above example, the prediction agreement calculation unit 140 calculates the prediction agreement as Ca = 2 / 5 = 0.4.

[0068] Returning to the description of the flowchart in Fig. 8, the alert output determination unit 150 determines whether the predicted matching score Ca calculated in the process of S110 is less than a threshold value Th1 (step S120). Here, the threshold value Th1 (first threshold value) is a predetermined value. For example, Th1 = 0.5, but is not limited to this.

[0069] If the predicted agreement Ca is not less than the threshold value Th1 (NO in S120), the alert output determination unit 150 determines not to output an alert (step S122). In this case, the processing flow may end. This alert corresponds to a warning indicating that the patient task may not be appropriately set and therefore the goal may not be achieved. On the other hand, if the predicted agreement Ca is less than the threshold value Th1 (YES in S120), the alert output determination unit 150 determines to output an alert (step S124).

[0070] Here, the predicted patient tasks are those previously addressed by previous patients with similar conditions to the target patient. Therefore, addressing the items included in the predicted patient tasks is expected to result in a high probability of achieving the target patient's goals. Therefore, the predicted patient tasks can be considered model patient tasks. In contrast, if the degree of agreement (prediction agreement) between the actual patient tasks and the predicted patient tasks is low, the actual patient tasks deviate from the predicted patient tasks, and the actual patient tasks are likely to be far removed from the model predicted patient tasks. As described above, since the therapist is actually addressing the pre-set patient tasks, the actual patient tasks correspond to the set patient tasks. Therefore, if the prediction agreement is low, the patient tasks may not be set appropriately, and the goal may not be achieved. Therefore, if the prediction agreement is low, the support device 100 according to the first embodiment outputs an alert indicating that the goal may not be achieved because the patient tasks may not be set appropriately.

[0071] When it is determined that an alert should be output (S124), the alert type determination unit 160 determines the type of alert (step S130). Here, the "type of alert" indicates why an alert was output. In other words, the type of alert may correspond to the cause of a discrepancy between the actual patient task and the predicted patient task. Furthermore, the type of alert indicates the situation under which an alert was output. In other words, the type of alert may correspond to the type of situation under which a discrepancy occurred between the actual patient task and the predicted patient task.

[0072] 11 is a flowchart showing an alert type determination method (S130) executed by the alert type determination unit 160 according to the first embodiment. The past comparison unit 162 of the alert type determination unit 160 determines the type of alert based on a comparison between an actual patient task and a patient task that was actually addressed in a period prior to the period in which the actual patient task was addressed (past actual patient task) (S132 to S140). Furthermore, the goal comparison unit 164 of the alert type determination unit 160 determines the type of alert based on a comparison between a character string of an item included in the actual patient task and a character string of a word included in the goal (S142 to S148).

[0073] The past comparison unit 162 acquires past actual patient tasks for the period Ta going back from time point A (step S132). That is, the past comparison unit 162 acquires patient tasks that were actually tackled (past actual patient tasks) from time point A(n) to time point A(n-1), going back the period Ta in Fig. 5. The method for acquiring past actual patient tasks may be substantially the same as the method for acquiring actual patient tasks in S104.

[0074] The past comparison unit 162 calculates the degree of agreement (past agreement Cb) between the actual patient task and the past actual patient task (step S134). The method for calculating the past agreement Cb may be, for example, a method substantially similar to the first example of the method for calculating the predicted agreement described above. In other words, the past agreement Cb may be the degree of agreement between the actual patient task and the past actual patient task. Specifically, the past comparison unit 162 calculates the total number Na3 (=Na1) of items included in the actual patient task. The past comparison unit 162 calculates the number Nb3 of items included in the past actual patient task among the items included in the actual patient task. The past comparison unit 162 calculates Nb3 / Na3 as the past agreement Cb. In other words, the past comparison unit 162 calculates the ratio of the number of items in the actual patient task that are included in the past actual patient task to the total number of items included in the actual patient task as the past agreement Cb.

[0075] The past comparison unit 162 determines whether the past agreement Cb is equal to or greater than a threshold value Th3 (step S136). Here, the threshold value Th3 (third threshold value) is a predetermined value. For example, Th3=0.5, but is not limited to this. If the past agreement Cb is equal to or greater than the threshold value Th3 (YES in S136), the deviation between the actual patient task and the past actual patient task is small. Therefore, in this case, the past comparison unit 162 determines that the therapist is "continuing the same patient task" for the target patient (step S138). This case in S138 is referred to as "Case A1." On the other hand, if the past agreement Cb is less than the threshold value Th3 (NO in S136), the deviation between the actual patient task and the past actual patient task is large. Therefore, in this case, the past comparison unit 162 determines that the therapist has "intentionally changed the patient task" for the target patient (step S140). This case in S140 is referred to as "Case A2."

[0076] In this way, the past comparison unit 162 is configured to determine the type of alert based on a comparison between the actual patient task and the past actual patient task, which allows the therapist or the like to easily understand whether an alert is being output because the same patient task is being continued or because the patient task has been intentionally changed.

[0077] The goal comparison unit 164 calculates the degree of match (goal match Cc) between the character string of the item in the actual patient task and the character string included in the goal (step S142). The method for calculating the goal match Cc may be, for example, a method substantially similar to the first example of the method for calculating the predicted match described above. In other words, the goal match Cc may be the degree of match (similarity) between the character string of the item included in the actual patient task and the character string of the word included in the goal.

[0078] The goal comparison unit 164 may also determine whether a character string of an item included in the actual patient task is included in a word of the goal. For example, in the examples of Figures 6 and 7, the character string of the actual patient task item "walking" is included in the goals (short-term goal and long-term goal) illustrated in Figure 7. Therefore, the goal comparison unit 164 determines that the actual patient task item "walking" matches a word of the goal.

[0079] The goal comparison unit 164 may also determine whether the words in the goal contain a character string similar to a character string in an item included in the actual patient task. For example, in the examples of FIGS. 6 and 7, the character string "IADL" in the actual patient task is similar to the word "lifestyle" included in the goals (short-term and long-term goals) illustrated in FIG. 7. Therefore, the goal comparison unit 164 determines that the item "IADL" in the actual patient task is similar to the words in the goal. The determination of whether words are similar can be made using an existing word similarity determination method. For example, the goal comparison unit 164 performs morphological analysis on the goal to break it down into words. The goal comparison unit 164 may then vectorize each word in the goal and the actual patient task and determine the similarity (cosine similarity) of the obtained word vectors to determine whether the words are similar. Alternatively, the goal comparison unit 164 may determine whether words are similar using a pre-created synonym dictionary.

[0080] The target comparison unit 164 calculates the total number of items included in the actual patient task, Na4 (=Na1). The target comparison unit 164 also calculates the number Nb4 of items in the actual patient task whose character strings match (or are similar to) character strings of words included in the target. The target comparison unit 164 calculates Nb4 / Na4 as the target coincidence Cc. In other words, the target comparison unit 164 calculates the ratio of the number of character strings of words included in the target among the items in the actual patient task to the total number of items included in the actual patient task as the target coincidence Cc.

[0081] The goal comparison unit 164 determines whether the goal coincidence Cc is equal to or greater than a threshold value Th4 (step S144). Here, the threshold value Th4 (fourth threshold value) is a predetermined value. For example, Th4=0.3, but is not limited to this. If the goal coincidence Cc is equal to or greater than the threshold value Th4 (YES in S144), the goal contains many character strings of items included in the actual patient task or similar character strings. Therefore, in this case, the goal comparison unit 164 determines that the therapist is "attempting to address the patient task in line with the goal" for the target patient (step S146). This case in S146 is referred to as "Case B1." On the other hand, if the goal coincidence Cc is less than the threshold value Th4 (NO in S144), the goal does not contain many character strings of items included in the actual patient task or similar character strings. Therefore, the therapist's intention in addressing the patient task is unclear. Therefore, the goal comparison unit 164 determines that the intention is "unclear" (step S148). This case of S148 will be referred to as "Case B2."

[0082] In this way, the goal comparison unit 164 is configured to determine the type of alert based on a comparison between the character strings of the items included in the actual patient task and the character strings of the words included in the goal. This allows the therapist or the like to easily understand whether an alert will be output even though they are trying to work on the patient task in line with the goal.

[0083] Returning to the description of the flowchart in Fig. 8, the alert output unit 170 performs control so that an alert is output (step S150). Here, the alert output unit 170 may perform control so that the interface unit 108 displays the alert. Alternatively, the alert output unit 170 may perform control so that the user terminal 60 displays the alert. In this case, the alert output unit 170 may transmit a command (alert output command) to the user terminal 60 to display the alert. As a result, the user terminal 60 displays the alert corresponding to the alert output command on an output device (user interface), such as a display device, provided in the user terminal 60.

[0084] Alternatively, the alert output unit 170 may perform control so that an alert is output by voice or the like. Even in this case, the alert output unit 170 may perform control so that the interface unit 108 or the user terminal 60 outputs the alert by voice or the like. In this case, the alert output unit 170 may transmit a command (alert output command) to the user terminal 60 to output the alert by voice. As a result, the user terminal 60 outputs the alert corresponding to the alert output command by activating a speaker provided in the user terminal 60.

[0085] Alternatively, the alert output unit 170 may perform control so that the alert is output by vibration or the like. In this case, the alert output unit 170 may transmit a command (alert output command) to the user terminal 60 to output the alert by vibration. As a result, the user terminal 60 activates a vibration function provided in the user terminal 60, thereby outputting an alert corresponding to the alert output command.

[0086] The alert output unit 170 may also perform control so that an alert is output for each patient. The alert output unit 170 may also perform control so that an alert is output in a different format for each type of alert determined in the processing of S130. For example, the alert output unit 170 may perform control so that an alert is output that is displayed differently for each type of alert.

[0087] FIG. 12 is a diagram illustrating an example of an alert output under the control of the alert output unit 170 according to the first embodiment. FIG. 12 shows an example in which an alert is output using a color display. That is, in the example of FIG. 12, the alert is displayed using a color on a display, a lamp, or the like. In addition, in the example of FIG. 12, the alert is displayed in a different color depending on the type of alert. Note that, as illustrated in FIG. 12, an alert may be output for each patient. That is, an alert may be output in association with a patient. In this case, the alert may be output together with the patient information illustrated in FIG. 7. For example, when patient information is displayed on a screen rendered as in FIG. 7, an image indicating the alert may be displayed in a blank area (e.g., the upper left area) of the screen displaying the patient information.

[0088] In the example of Fig. 12, an alert image Im0 indicating "no alert" is displayed for patient A. The alert image Im0 is shown in blue, for example. In this case, the patient task set for patient A may have been set appropriately.

[0089] Additionally, an alert image Im1 is displayed for patient B, indicating an alert for "Case A1 and Case B1." The alert image Im1 is shown in red, for example. In this case, the patient task set for patient B may not have been set appropriately in the situation of "Case A1 and Case B1."

[0090] Additionally, an alert image Im2 is displayed for patient C, indicating an alert for "Case A1 and Case B2." The alert image Im2 is shown in orange, for example. In this case, the patient task set for patient C may not have been set appropriately in the situation of "Case A1 and Case B2."

[0091] Additionally, an alert image Im3 is displayed for patient D, indicating an alert for "Case A2 and Case B1." The alert image Im3 is shown in dark red, for example. In this case, the patient task set for patient D may not have been set appropriately in the situation of "Case A2 and Case B1."

[0092] Additionally, an alert image Im4 is displayed for patient E, indicating an alert for "Case A2 and Case B2." The alert image Im4 is shown in purple, for example. In this case, the patient task set for patient E may not have been set appropriately in the situation of "Case A2 and Case B2."

[0093] In the example of FIG. 12, alerts are output in different colors for each type of alert, but this is not a limitation. Alerts may be displayed as messages such as text. In this case, for patient B in the example of FIG. 12, a message such as "There is an alert for case A1 and case B1 for patient B" may be displayed. When an alert is output as audio, the above message may be output as audio. When an alert is output as an audible alarm, different audible alarms may be output for different types of alerts. When an alert is output as a vibration, different vibration patterns may be generated for different types of alerts. Alternatively, an alert for "case A1 and case B1" may be output as audio, and an alert for "case A1 and case B2" may be output as vibration.

[0094] As described above, the support device 100 according to the first embodiment is configured to output an alert when the degree of agreement between the actual patient task and the predicted patient task is low. As a result, an alert is output when there is a difference between the predicted patient task that can serve as a model and the actual patient task, and therefore, there is a possibility that the patient task has not been set appropriately. Therefore, it is possible to efficiently check whether the patient task has been set appropriately.

[0095] It is preferable that the therapist in charge of the patient for whom an alert has been output checks the output alert, but rather that the therapist's supervisor (e.g., team leader) checks the alert. In other words, it is preferable that the alert is output to the team leader's user terminal 60. This allows the team leader to provide appropriate advice and guidance to the therapist in charge. Furthermore, by outputting alerts in different formats for each type of alert, the team leader can more easily understand the type of alert, allowing them to provide more appropriate advice, etc.

[0096] For example, if an alert is output for Case A1, the team leader can point out to the therapist in charge of the patient for whom the alert was output, "You're repeating the same patient tasks over and over. Don't you need to work on other patient tasks?" Also, if an alert is output for Case A2, the team leader can point out to the therapist in charge of the patient for whom the alert was output, "It seems like you're working on a different patient task than last time. Why?" Also, if an alert is output for Case B1, the team leader can understand that even though they are trying to work on patient tasks that are in line with the goal, the patient tasks may not be appropriate.

[0097] Furthermore, if an alert is output to the user terminal 60 of the therapist in charge (especially an inexperienced therapist), there is a risk that the therapist in charge will blindly believe the alert and easily revise the patient's tasks. On the other hand, therapists often set patient tasks by taking into account the patient's circumstances (patient information, external factors, etc.) that are not reflected in data through conversations with the patient and observations of the patient's condition. In this way, it can be said that patient tasks that therapists, for example, under the direction of a doctor, consider necessary for the patient to achieve goals should be respected as much as possible. Therefore, by outputting an alert to the user terminal 60 of the team leader, it is expected that the team leader will be able to discuss with the therapist in charge and set more appropriate patient tasks.

[0098] In this embodiment, the patient task is predicted based on past data, but it is preferable that the prediction results not be presented to the therapist in charge for the following reasons. As described above, the patient task that the therapist has considered, taking into account the patient's circumstances, for example, under the direction of a doctor, should be respected as much as possible, so it is undesirable for the therapist in charge to blindly believe the prediction results and easily change the patient task. Furthermore, if the prediction results are presented to the therapist in charge, there is a risk that the therapist will set the patient task in accordance with the prediction results, and will not carefully consider the patient task. Therefore, from an educational perspective, it is preferable that the prediction results not be presented, especially to inexperienced therapists.

[0099] As described above, in the second example of the method for calculating the degree of prediction agreement, the degree of prediction agreement is calculated for high-frequency items included in the actual patient tasks that were frequently addressed during the period Tb. For example, suppose the actual patient tasks include five items, three of which are high-frequency items. Furthermore, suppose that of the five items included in the actual patient tasks, the items included in the predicted patient tasks are two of the high-frequency items. Furthermore, let the threshold value Th1 be 0.5. In this case, in the first example, the degree of prediction agreement is 2 / 5 = 0.4, which is below the threshold value Th1, and therefore an alert is output. On the other hand, in the second example, the degree of prediction agreement is 2 / 3 ≒ 0.67, which is above the threshold value Th1, and therefore no alert is output. As such, the results may differ between the first and second examples even if the same actual patient tasks and predicted patient tasks are used.

[0100] Here, the high frequency items are items that the therapist frequently attempts to address, and therefore may reflect the therapist's intentions. Therefore, by calculating the degree of agreement between the actual patient tasks and the predicted patient tasks for the high frequency items, it is possible to more effectively determine whether the therapist's intentions are appropriate for achieving the goal. That is, if the predicted agreement calculated for the high frequency items is low, it is likely that the therapist is repeating ineffective patient tasks multiple times, and therefore the therapist's intentions are likely to be inappropriate for achieving the goal. On the other hand, if the predicted agreement calculated for the high frequency items is high, it is likely that the therapist is repeating effective patient tasks, and therefore the therapist's intentions are likely to be appropriate for achieving the goal. Therefore, by calculating the predicted agreement as in the second example, it is possible to determine whether to output an alert while reflecting the therapist's intentions. That is, it is possible to output an alert when the therapist's intentions may be inappropriate for achieving the goal.

[0101] (Embodiment 2) Next, a second embodiment will be described with reference to the drawings. For clarity of explanation, the following description and drawings have been omitted and simplified as appropriate. In addition, in each drawing, the same elements are given the same reference numerals, and repeated explanations are omitted as necessary. Note that the system configuration according to the second embodiment is substantially the same as that shown in FIG. 3, and therefore explanations thereof will be omitted. In addition, the configuration of the support device 100 according to the second embodiment is substantially the same as that shown in FIG. 4, and therefore explanations thereof will be omitted.

[0102] As illustrated in FIG. 6, items included in a patient task can be set for each of a plurality of categories. The support device 100 according to the second embodiment calculates the degree of agreement for each category of patient task and performs control so that an alert is output for each category of patient task. That is, the predicted agreement calculation unit 140 according to the second embodiment calculates the degree of agreement for each category of patient task. Furthermore, the alert output unit 170 according to the second embodiment performs control so that an alert is output for each category of patient task. Here, the "major item of a higher-level task" is "category A," the "minor item of a higher-level task" is "category B," the "major item of a lower-level task" is "category C," and the "minor item of a lower-level task" is "category D."

[0103] 13 is a flowchart showing a support method executed by the support device 100 according to the second embodiment. As in S100 of FIG. 8, the patient information storage unit 110 stores patient information (step S200). As in S102 of FIG. 8, the patient problem prediction unit 120 predicts a patient problem (step S202). As in S104 of FIG. 8, the actual patient problem acquisition unit 130 acquires the actual patient problem of the target patient (step S204).

[0104] The support device 100 executes S110 to S150 for the major item (category A) of the higher-level task (step S212). Specifically, the predicted agreement calculation unit 140 calculates the predicted agreement Ca for category A (S110). That is, the predicted agreement calculation unit 140 calculates the agreement between the items included in category A of the actual patient task and the items included in category A of the predicted patient task as the predicted agreement Ca. Furthermore, the alert output determination unit 150 determines whether the predicted agreement Ca for category A is less than a threshold Th1 (S120). Furthermore, when it is determined that an alert should be output for category A (S124), the alert type determination unit 160 determines the type of alert for category A (S130). The past comparison unit 162 determines the type of alert based on a comparison between the items included in category A of the actual patient task and the items included in category A of the past actual patient task. Furthermore, the target comparison unit 164 determines the type of alert based on a comparison between the character strings of the items included in category A of the actual patient task and the character strings of the words included in the target. Then, the alert output unit 170 performs control so that an alert for category A is output (S150). As will be described below, similar processing is performed for categories B, C, and D.

[0105] The support device 100 executes S110 to S150 for a minor item (category B) of the higher-level task (step S214). Specifically, the predicted coincidence calculation unit 140 calculates the predicted coincidence Ca for category B (S110). Furthermore, the alert output determination unit 150 determines whether the predicted coincidence Ca for category B is less than a threshold Th1 (S120). Furthermore, if it is determined that an alert should be output for category B (S124), the alert type determination unit 160 determines the type of alert for category B (S130). Then, the alert output unit 170 performs control so that an alert for category B is output (S150).

[0106] The support device 100 executes S110 to S150 for the major item (category C) of the sub-task (step S216). Specifically, the predicted coincidence calculation unit 140 calculates the predicted coincidence Ca for category C (S110). Furthermore, the alert output determination unit 150 determines whether the predicted coincidence Ca for category C is less than a threshold Th1 (S120). Furthermore, if it is determined that an alert should be output for category C (S124), the alert type determination unit 160 determines the type of alert for category C (S130). Then, the alert output unit 170 performs control so that an alert for category C is output (S150).

[0107] The support device 100 executes S110 to S150 for a minor item (category D) of a sub-task (step S218). Specifically, the predicted coincidence calculation unit 140 calculates the predicted coincidence Ca for category D (S110). Furthermore, the alert output determination unit 150 determines whether the predicted coincidence Ca for category D is less than a threshold Th1 (S120). Furthermore, if it is determined that an alert should be output for category D (S124), the alert type determination unit 160 determines the type of alert for category D (S130). Then, the alert output unit 170 performs control so that an alert for category D is output (S150).

[0108] Fig. 14 is a diagram illustrating an example of an alert output under the control of the alert output unit 170 according to the second embodiment. As in Fig. 12, Fig. 14 shows an example in which an alert is output using a different color. As in Fig. 12, in the example of Fig. 14, alerts are displayed in different colors depending on the type of alert.

[0109] In the example of FIG. 14, alerts are displayed for patient A for each of categories A, B, C, and D. In the example of FIG. 14, an alert image Im0 indicating "no alert" is displayed for category A. Furthermore, an alert image Im1 indicating an alert for "case A1 and case B1" is displayed for category B. Furthermore, an alert image Im3 indicating an alert for "case A2 and case B1" is displayed for category C. Furthermore, an alert image Im1 indicating an alert for "case A1 and case B1" is displayed for category D.

[0110] The support device 100 according to the second embodiment is configured to output an alert for each of a plurality of categories of patient tasks. This makes it possible to check which of the plurality of categories of patient tasks contains an item that has not been set appropriately. This makes it possible to check in more detail whether the patient tasks have been set appropriately. In other words, it becomes possible to check more efficiently and effectively whether the patient tasks have been set appropriately.

[0111] In the example of FIG. 14, an alert is output for all categories of patient tasks, but this is not the only possible configuration. That is, it is not necessary to output an alert for all categories of patient tasks. In other words, an alert may be output for only some combinations of multiple categories of patient tasks. For example, an alert may be output for only categories A and C. In this case, the processes of S214 and S218 in FIG. 13 may be omitted. Furthermore, for example, an alert may be output for categories A, B, and C. In this case, the process of S218 in FIG. 13 may be omitted.

[0112] (Variation) The present invention is not limited to the above-described embodiment, and can be modified as appropriate without departing from the spirit of the present invention. For example, in the above-described flowchart, the order of each process (step) can be modified as appropriate. Furthermore, one or more of the multiple processes (steps) may be omitted. For example, the process of S130 in FIG. 8 may be omitted. In other words, in this embodiment, it is not necessary to determine the type of alert.

[0113] Furthermore, although the assistance device 100 according to the above-described embodiment provides assistance in setting a patient task that can achieve the rehabilitation goal, the activity to which this embodiment is applied is not limited to rehabilitation. This embodiment can be applied to any activity aimed at improving abilities. For example, this embodiment can also be applied to habilitation. Furthermore, this embodiment can also be applied to activities to improve sports abilities.

[0114] Furthermore, in the above-described embodiment, the items included in the patient task are classified into higher-level tasks and lower-level tasks, and into major items and minor items, but this configuration is not limited to this. The items included in the patient task do not have to be classified into higher-level tasks and lower-level tasks. Similarly, the items included in the patient task do not have to be classified into major items and minor items. Furthermore, in the above-described first embodiment, the items included in the patient task do not have to be set for multiple categories. Furthermore, in the second embodiment, it is sufficient that the items included in the patient task are set for any multiple categories, and they do not have to be classified into higher-level tasks and lower-level tasks, or into major items and minor items.

[0115] Furthermore, the patient task items to be compared when calculating the predicted match may be a combination of major and minor tasks. In this case, in the example of FIG. 6, the actual patient task may include "Walking - Right Side," "Walking - Paralyzed Side," and "IADL - Housework" as higher-level task items. Furthermore, the actual patient task illustrated in FIG. 6 may include "Balance - Standing," "Balance - Sitting," and "Upper Limb Function" as lower-level task items. Alternatively, the patient task items to be compared when calculating the predicted match may be separated into major and minor tasks. In this case, in the example of FIG. 6, the actual patient task may include "Walking," "IADL," "Right Side," "Parly Side," and "Housework" as higher-level task items. Furthermore, the actual patient task illustrated in FIG. 6 may include "Balance," "Upper Limb Function," "Standing," and "Sitting" as lower-level task items. The same applies to predicted patient tasks. The same applies to calculating the past match.

[0116] Furthermore, the prediction matching degree may be the number of items included in the actual patient task that are also included in the predicted patient task. In other words, the prediction matching degree may be the number of elements in the intersection of the set of items included in the actual patient task and the set of items included in the predicted patient task. The same applies to the past matching degree and the target matching degree. In other words, the past matching degree may be the number of items included in the actual patient task that are also included in the past actual patient task. Furthermore, the target matching degree may be the number of items included in the actual patient task whose character strings match or are similar to character strings of words included in the goal.

[0117] The prediction agreement may also be the degree of agreement between the predicted patient task and the actual patient task. In this case, in the first example of the method for calculating the prediction agreement described above, the prediction agreement may be the ratio of the number of items in the predicted patient task that are included in the actual patient task to the total number of items included in the predicted patient task. The same applies to the second example. When calculating the prediction agreement as described above, even if many items in the actual patient task are also included in the predicted patient task, as in the example of Figure 6, the prediction agreement may be low if the number of items in the predicted patient task is greater than the number of items in the actual patient task. In other words, if the number of items in the predicted patient task is greater than the number of items in the actual patient task and there are items that are not included in the actual patient task but are included in the predicted patient task, the prediction agreement may be low. On the other hand, items that are not included in the actual patient task but are included in the predicted patient task may be addressed during the period Tc in Figure 5. Therefore, it should be noted that a low prediction agreement calculated as described above does not necessarily mean that the patient task is not set appropriately. In other words, as in the first and second examples, by defining the degree of predicted agreement as the ratio of the number of items common to the actual patient tasks to the number of items in the actual patient tasks, it is possible to more effectively evaluate whether the patient tasks actually addressed are appropriate.

[0118] The above-mentioned program includes a set of instructions (or software code) that, when loaded into a computer, causes the computer to perform one or more functions described in the embodiments. The program may be stored in a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable media or tangible storage media include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technologies, CD-ROM, digital versatile disk (DVD), Blu-ray® disk or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable media or communication media include electrical, optical, acoustic, or other forms of propagated signals.

[0119] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes. (Appendix 1) a predictive means for predicting patient challenges that should be addressed to achieve the patient's goals; a calculation means for calculating a degree of agreement between an actual patient task, which is a patient task that has actually been addressed for the patient, and a predicted patient task, which is a predicted patient task; an output means for controlling the device so that an alert is output when the degree of match is less than a predetermined first threshold; An assist device having: (Appendix 2) The prediction means calculates the predicted patient task for a predetermined first period starting from a first time point, the calculation means calculates a degree of agreement between the actual patient task that has been actually addressed during a predetermined second period that is shorter than the first period and that starts from the first time point, and the predicted patient task; 10. The assistive device of claim 1. (Appendix 3) The calculation means calculates, as the degree of agreement, a ratio of the number of items included in the predicted patient task among the items included in the actual patient task to the total number of items included in the actual patient task. 10. The assistive device of claim 2. (Appendix 4) the calculation means calculates, as the degree of agreement, a ratio of the number of first items included in the predicted patient task to the total number of first items, among the items included in the actual patient task, whose frequency of being addressed in the second period is equal to or greater than a predetermined second threshold; 10. The assistive device of claim 2. (Appendix 5) The items included in the patient task are set for each of a plurality of predetermined categories, the calculation means calculates the degree of match for each of the categories; The output means performs control so that the alert is output for each of the categories. 5. The assistance device according to any one of claims 1 to 4. (Appendix 6) a type determination means for determining the type of the alert based on the actual patient problem when the degree of coincidence is less than the first threshold value; and The output means performs control so that the alert is output in a different format depending on the determined type of alert. 6. The assistance device according to any one of appendices 1 to 5. (Appendix 7) the type determination means determines the type of the alert based on a comparison between the actual patient task and a patient task that was actually addressed in a period prior to the period in which the actual patient task was addressed. 10. The support device according to claim 6. (Appendix 8) the type determination means determines the type of the alert based on a comparison between a character string of an item included in the actual patient task and a character string of a word included in the goal. 8. The assistance device according to claim 6 or 7. (Appendix 9) Anticipate patient challenges that need to be addressed to achieve patient goals; Calculating the degree of agreement between actual patient tasks, which are patient tasks that have actually been addressed for the patient, and predicted patient tasks, which are predicted patient tasks; performing control so that an alert is output when the degree of match is less than a predetermined first threshold; How to help. (Appendix 10) Calculating the predicted patient task for a predetermined first period starting from a first time point; calculating a degree of agreement between the actual patient task that has been actually addressed during a predetermined second period starting from the first time point and being shorter than the first period, and the predicted patient task; Support methods described in Appendix 9. (Appendix 11) Calculating the degree of agreement as a ratio of the number of items included in the predicted patient task among the items included in the actual patient task to the total number of items included in the actual patient task; Support methods described in Appendix 10. (Appendix 12) calculating, as the degree of agreement, a ratio of the number of first items included in the predicted patient task to the total number of first items, among the items included in the actual patient task, whose frequency of being addressed in the second period is equal to or greater than a predetermined second threshold value; Support methods described in Appendix 10. (Appendix 13) The items included in the patient task are set for each of a plurality of predetermined categories, calculating the degree of match for each of the categories; performing control so that the alert is output for each of the classifications; 13. The method of support described in any one of appendices 9 to 12. (Appendix 14) If the degree of match is less than the first threshold, determining the type of alert based on the actual patient issue; performing control so that the alert is output in a different format depending on the determined type of alert; 14. The method of support described in any one of appendices 9 to 13. (Appendix 15) determining the type of the alert based on a comparison between the actual patient problem and a patient problem that was actually addressed in a period prior to the period in which the actual patient problem was addressed; Support methods described in Appendix 14. (Appendix 16) determining the type of the alert based on a comparison between a character string of an item included in the actual patient task and a character string of a word included in the goal; The support method described in Appendix 14 or 15. (Appendix 17) predicting patient challenges to be addressed to achieve the patient's goals; A step of calculating a degree of agreement between an actual patient task, which is a patient task actually addressed for the patient, and a predicted patient task, which is a predicted patient task; performing control such that an alert is output when the degree of match is less than a predetermined first threshold; A program that causes a computer to execute the following. [Explanation of symbols]

[0120] 1 Support equipment 2. Prediction Department 4 Calculation section 6 Output section 50 Support System 52 Network 60 User terminals 100 Support equipment 110 Patient information storage unit 120 Patient Issue Prediction Department 130 Actual Patient Issue Acquisition Department 140 Prediction matching calculation unit 150 Alert output determination unit 160 Alert type determination unit 162 Past Comparison Section 164 Target comparison section 170 Alert Output Unit

Claims

1. a predictive means for predicting patient challenges that should be addressed to achieve the patient's goals; a calculation means for calculating a degree of agreement between an actual patient task, which is a patient task that has actually been addressed for the patient, and a predicted patient task, which is a predicted patient task; an output means for controlling the device so that an alert is output when the degree of match is less than a predetermined first threshold; An assist device having:

2. The prediction means calculates the predicted patient task for a predetermined first period starting from a first time point, the calculation means calculates a degree of agreement between the actual patient task that has been actually addressed during a predetermined second period that is shorter than the first period and that starts from the first time point, and the predicted patient task; The support device according to claim 1 .

3. The calculation means calculates, as the degree of agreement, a ratio of the number of items included in the predicted patient task among the items included in the actual patient task to the total number of items included in the actual patient task. The support device according to claim 2 .

4. the calculation means calculates, as the degree of agreement, a ratio of the number of first items included in the predicted patient task to the total number of first items, among the items included in the actual patient task, whose frequency of being addressed in the second period is equal to or greater than a predetermined second threshold; The support device according to claim 2 .

5. The items included in the patient task are set for each of a plurality of predetermined categories, the calculation means calculates the degree of match for each of the categories; The output means performs control so that the alert is output for each of the categories. The assistance device according to any one of claims 1 to 4.

6. a type determination means for determining a type of the alert based on the actual patient problem when the degree of coincidence is less than the first threshold; and The output means performs control so that the alert is output in a different format depending on the determined type of alert. The assistance device according to any one of claims 1 to 5.

7. the type determination means determines the type of the alert based on a comparison between the actual patient task and a patient task that was actually addressed in a period prior to the period in which the actual patient task was addressed. The support device according to claim 6.

8. the type determination means determines the type of the alert based on a comparison between a character string of an item included in the actual patient task and a character string of a word included in the goal.

8. The assistance device according to claim 6 or 7.

9. Anticipate patient challenges that need to be addressed to achieve patient goals; Calculating the degree of agreement between actual patient tasks, which are patient tasks that have actually been addressed for the patient, and predicted patient tasks, which are predicted patient tasks; performing control such that an alert is output when the degree of match is less than a predetermined first threshold; How to help.

10. predicting patient challenges to be addressed to achieve the patient's goals; A step of calculating a degree of agreement between an actual patient task, which is a patient task actually addressed for the patient, and a predicted patient task, which is a predicted patient task; performing control such that an alert is output when the degree of match is less than a predetermined first threshold; A program that causes a computer to execute the following.

Images