State evaluation device, state evaluation method, and state evaluation program

Abstract

To provide a state evaluation device capable of suppressing the influence of variation in information granularity of record information due to individuality of a target person on the basis of the record information in an arbitrary text format, and performing quantitative evaluation without variation in a state of the target person.SOLUTION: A state evaluation device comprises: a classification section including a machine-learned language model that outputs by determining whether or not a predetermined label is appropriate for each piece of record information with respect to an input of one or more pieces of record information in a predetermined period generated in an arbitrary text format regarding a target person; a change amount calculation section that calculates a change amount of a state of the target person on the basis of a result of determination of whether or not the predetermined label is appropriate by the classification section in a plurality of predetermined periods; and a determination section that determines the state of the target person on the basis of the change amount.SELECTED DRAWING: Figure 4

Description

[Technical Field]

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

[0002] Japan has seen a remarkable increase in life expectancy due to improvements in living standards, sanitary conditions, and medical standards that came with the rapid economic growth after the war. This, combined with a declining birth rate, has led to an aging society with a high aging rate. In such an aging society, it is expected that the number of people who require care, such as nursing care, due to illness, injury, and aging, will increase.

[0003] In nursing care facilities, multiple care staff work in shifts to provide care to one patient. Care staff create care records to enable each patient's condition to be understood by the other care staff. Care records are created in any written format to accommodate the individuality of each patient. Nursing records may be required to include quantitative assessments of each patient's condition to demonstrate the effectiveness of the care plan.

[0004] However, because the quantitative assessment of each subject's condition is performed independently by the care staff, there is a possibility that variations will occur and the reliability of the assessment results may be low.

[0005] The following prior art is disclosed in Patent Document 1 listed below. Fixed phrase information that sets fixed phrases to be used for entering care information is stored in a storage unit. Fixed phrases are displayed in the care record field of the care record input screen so that they can be selected using a pull-down menu. The fixed phrase selected by the caregiver is displayed in the care record field. Input is accepted into variable input fields included in the fixed phrase. Then, the care record in which input has been made into the variable input fields is stored in the care record information.

[0006] Patent Document 2 below discloses the following prior art: A medical ontology and input data entered in natural language format in a free-form comment field of a medical report are acquired. The medical ontology is a database that stores relationships between medical languages ​​and relationships between medical languages ​​and medical classifications. Natural language processing is performed on the input data to generate analysis data of the input data. Structured data in which the medical ontology is associated with the analysis data is generated. Confirmation data that displays the structured data in natural language format is generated. The confirmation data is then displayed on a display. The confirmation data includes an assessment of the possibility of cerebral infarction, etc. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Publication No. 2022-103155 [Patent Document 2] Japanese Patent Publication No. 2022-54218 Summary of the Invention [Problem to be solved by the invention]

[0008] However, the prior art described in Patent Document 1 requires the nursing care record to be written using standard phrases, which may prevent flexible and appropriate nursing care records from being created. Furthermore, changes in the method of writing nursing care records incur training costs for care staff. Furthermore, past nursing care records cannot be effectively utilized, making it impossible to quantitatively evaluate the condition of the subject.

[0009] The prior art described in Patent Document 2 allows for the input of medical reports and the like in natural language format. However, it requires the preparation of a medical ontology. Creating a medical ontology requires high technical skills in both medical and statistical terms, and there is a possibility that it may not be sufficiently comprehensive in terms of operation.

[0010] The present invention has been made to solve the above-mentioned problems, and aims to provide a condition assessment device, a condition assessment method, and a condition assessment program that can suppress the influence of variations in the information granularity of the recorded information due to the individuality of the subject, based on recorded information in any text format, and enable a consistent quantitative assessment of the subject's condition. [Means for solving the problem]

[0011] The above-mentioned problems of the present invention are solved by the following means.

[0012] (1) A condition assessment device having: a classification unit including a machine-learned language model that, in response to input of one or more pieces of record information relating to a subject over a predetermined period, created in any sentence format, determines whether each piece of record information corresponds to a predetermined label and outputs the determined result; a change amount calculation unit that calculates the amount of change in the subject's condition based on the results of the classification unit's determination of whether each piece of record information corresponds to a predetermined label over a plurality of the predetermined periods; and a judgment unit that judges the subject's condition based on the amount of change.

[0013] (2) The condition evaluation device according to (1) above, wherein the change amount calculation unit calculates the amount of change between each of the plurality of predetermined periods from the number of times the predetermined label is associated with the predetermined period in each of the plurality of predetermined periods.

[0014] (3) A condition evaluation device as described in (2) above, further comprising an output unit that outputs the number of times the specified label corresponds to the specified period calculated by the change amount calculation unit, along with the condition of the subject determined by the determination unit.

[0015] (4) The condition assessment device described in (1) above, wherein the language model is machine-learned based on the recorded information for the specified period, created in any sentence format, and the results of known judgments of whether the specified label applies to the recorded information for the specified period.

[0016] (5) The state evaluation device according to (2), wherein the change amount calculation unit calculates the change amount using a threshold value set for the number of times the specified label is relevant.

[0017] (6) The condition evaluation device described in (1) above, wherein the change amount calculation unit calculates the change amount of the subject's condition by correcting the change amount of the subject's condition, estimated based on the results of the judgment of whether or not the condition is correct for multiple specified periods, based on the results of visual confirmation by a user or by using a learned LLM model.

[0018] (7) The condition assessment device described in (1) above, wherein the recorded information is a care record and the subject's condition is an assessment for each item of a dementia behavioral disorder scale.

[0019] (8) The state assessment device described in (1) above, wherein the language model further outputs a likelihood for each of the specified labels for each of the record information in response to input of one or more of the record information for the specified period regarding the subject, the record information being created in any sentence format.

[0020] (9) The change amount calculation unit weights the results of the judgment of whether or not the condition is correct for the specified period by the likelihood, and calculates the change amount of the subject's condition based on the weighted results of the judgment of whether or not the condition is correct for multiple specified periods.This is a condition evaluation device described in (8) above.

[0021] (10) A condition evaluation method comprising the steps of: (a) inputting one or more pieces of record information relating to a subject over a predetermined period, written in any sentence format, and using a machine-learned language model to make a judgment as to whether a predetermined label applies to each piece of record information and output the judgment; (b) calculating the amount of change in the subject's condition based on the results of the judgment as to whether a predetermined label applies to each piece of record information over a plurality of the predetermined periods in (a); and (c) judging the subject's condition based on the amount of change calculated in (b).

[0022] (11) The condition evaluation method according to (10) above, wherein in step (b), the amount of change between each of the plurality of predetermined periods is calculated from the number of times the predetermined label is relevant in each of the plurality of predetermined periods.

[0023] (12) A condition evaluation method as described in (11) above, further comprising a step (d) of outputting the number of occurrences of the specified label during the specified period output in step (b) along with the condition of the subject determined in step (c).

[0024] (13) A condition assessment method as described in (10) above, wherein the language model is machine-learned based on the recorded information for the specified period, created in any sentence format, and the results of known judgments of whether the specified label applies to the recorded information for the specified period.

[0025] (14) The state evaluation method according to (11) above, wherein in the step (b), the amount of change is calculated using a threshold value set for the number of occurrences of the predetermined label.

[0026] (15) In the step (b), the change in the subject's condition estimated based on the results of the judgment for multiple predetermined periods is corrected based on the results of visual confirmation by a user or by using a trained LLM model, thereby calculating the change in the subject's condition.

[0027] (16) The condition assessment method described in (10) above, wherein the recorded information is a care record and the condition of the subject is an assessment for each item of the Dementia Behavioral Disorders Scale.

[0028] (17) The state assessment method described in (10) above, wherein the language model further outputs a likelihood for each of the predetermined labels for each of the record information in response to input of one or more of the record information for the specified period, the record information being created in any sentence format, regarding the subject.

[0029] (18) In the step (b), the results of the judgment of whether or not the subject is in a state of being determined for the specified period are weighted by the likelihood, and the amount of change in the subject's state is calculated based on the weighted results of the judgment of whether or not the subject is in a state of being determined for a plurality of the specified periods.

[0030] (19) A condition evaluation program for executing the condition evaluation method according to any one of (10) to (18) above by a computer. [Effects of the Invention]

[0031] Using a machine-learned language model, a judgment is made as to whether a predetermined label applies to each piece of recorded information input over a predetermined period in an arbitrary sentence format.The amount of change in the subject's condition is calculated based on the results of the judgments over multiple predetermined periods, and the subject's condition is judged based on this amount of change.This suppresses the influence of variations in the information granularity of the recorded information due to the individuality of the subject's condition based on the recorded information in an arbitrary sentence format, enabling a consistent quantitative evaluation of the subject's condition. [Brief explanation of the drawings]

[0032] Advantages and features provided by one or more embodiments of the present invention will be more fully understood from the following detailed description and the accompanying drawings, which are for purposes of illustration only and are not intended to be limiting. [Figure 1] FIG. 1 is a diagram illustrating an overall configuration of an information processing system. [Figure 2] FIG. 2 is a block diagram showing a schematic configuration of a server. [Figure 3] FIG. 2 is a block diagram showing the functions of a control unit. [Figure 4] FIG. 2 is an explanatory diagram showing information processing by a control unit. [Figure 5] FIG. 1 is a diagram showing the 13 types of items in DBD13. [Figure 6] FIG. 10 is an explanatory diagram showing an example of a determination of applicability of items in DBD13 for each care record. [Figure 7]FIG. 10 is a diagram showing an output screen of the judgment results of the subject's current evaluation of each item of DBD13, output by the output control unit. [Figure 8] FIG. 2 is a block diagram showing a schematic configuration of a mobile terminal. [Figure 9] FIG. 10 is a diagram illustrating an example of a flow of a periodic assessment. [Figure 10] 10 is a flowchart showing the operation of the server. [Figure 11] FIG. 10 is a diagram showing recorded information, a predetermined label, and the state of a subject to be evaluated in Modification 1. [Figure 12] FIG. 10 is a diagram showing recorded information, a predetermined label, and the state of a subject to be evaluated in Modification 2. [Figure 13] FIG. 10 is a diagram showing recorded information, a predetermined label, and the state of a subject to be evaluated in Modification 3. DETAILED DESCRIPTION OF THE INVENTION

[0033] Hereinafter, a condition assessment device, a condition assessment method, and a condition assessment program according to an embodiment of the present invention will be described with reference to the accompanying drawings. However, the scope of the present invention is not limited to the disclosed embodiments. In the drawings, identical elements are denoted by the same reference numerals, and duplicate explanations will be omitted. Furthermore, the dimensional proportions in the drawings are exaggerated for the sake of explanation and may differ from the actual proportions.

[0034] In this specification, the subject broadly includes care recipients who receive care such as nursing care or caregiving, people with disabilities who do not require care, healthy people who have previously experienced an injury or surgery, etc. The subject includes, for example, people who require care, people who require assistance, and patients. For simplicity of explanation, the following description will be given taking the case where the subject is a person who requires care as an example.

[0035] (First embodiment) FIG. 1 is a diagram showing the overall configuration of an information processing system 1. As shown in FIG. 1, the information processing system 1 includes a server 10, a fixed terminal 20, and one or more mobile terminals 30. These are connected to each other so as to be able to communicate with each other via a network 40, such as a LAN (Local Area Network), a telephone network, or a data communication network, either wired or wirelessly. The network 40 may include a repeater that relays communication signals. In the example shown in FIG. 1, the server 10, the mobile terminal 30, and the fixed terminal 20 are connected to each other so as to be able to communicate with each other via a network 40, such as a wireless LAN (for example, a LAN conforming to the IEEE 802.11 standard) that includes an access point 41.

[0036] The mobile terminal 30 may be carried by each of the care staff 50 who care for the subject.

[0037] The information processing system 1 may be configured by only the server 10. The information processing system 1 constitutes a condition evaluation device.

[0038] The server 10 and the fixed terminal 20 are preferably installed in buildings such as elderly care facilities and hospitals. The server 10 may be configured as an on-premise server or a cloud server. The server 10 may also be configured as a stand-alone PC (Personal Computer).

[0039] (Server 10) 2 is a block diagram showing a schematic configuration of the server 10. The server 10 includes a control unit 11, a communication unit 12, and a storage unit 13. These components are interconnected by a bus. The server 10 constitutes an information processing device.

[0040] The control unit 11 is configured with a CPU (Central Processing Unit) and memories such as RAM (Random Access Memory) and ROM (Read Only Memory), and controls each part of the server 10 and performs arithmetic processing according to a program.

[0041] The communication unit 12 is an interface for communicating with other devices including, for example, the fixed terminal 20 and the mobile terminal 30 via the network 40, and may be, for example, a LAN card.

[0042] The storage unit 13 is configured by a hard disk drive (HDD), a solid state device (SSD), etc. The storage unit 13 stores various programs and various data.

[0043] Fig. 3 is a block diagram showing the functions of the control unit 11. Fig. 4 is an explanatory diagram showing information processing by the control unit 11. By executing a program, the control unit 11 functions as a recorded information receiving unit 111, a classification unit 112, a change amount calculation unit 113, a state determination unit 114, a learning unit 115, and an output control unit 116. The state determination unit 114 constitutes the determination unit. The output control unit 116 constitutes the output unit.

[0044] The record information receiving unit 111 receives record information. The record information includes various records such as care records, diagnosis records, and follow-up observation records. For simplicity of explanation, the following description will be given taking the case where the record information is care records as an example.

[0045] The record information accepting unit 111 accepts, for example, the care record input by the care staff 50 at the mobile terminal 30 by receiving it from the mobile terminal 30. The accepted care record can be stored in the storage unit 13.

[0046] The record information receiving unit 111 receives care records created in any text format. That is, the record information receiving unit 111 receives care records created in non-standard sentences and standard sentences. The care records may be input by the care staff 50 into each mobile terminal 30 and received by the communication unit 12. The care records are input by the care staff 50 into the mobile terminal 30 for each subject at least once a day. That is, the record information receiving unit 111 receives one or more care records for one subject.

[0047] The classification unit 112 determines whether a specific label applies to each input care record and outputs the result. That is, the classification unit 112 classifies the care record into the corresponding label by determining whether the specific label applies to the care record. One care record can be classified into one or more labels. The classification unit 112 includes a machine-learned language model. Any language model can be used as long as it determines whether a specific label applies to each input care record and outputs the result. For example, BERT (Bidirectional Encoder Representations from Transformers) is used as the language model. The specific labels include any labels corresponding to the subject's evaluation. For simplicity, the following description will be given using an example in which the specific labels are evaluations of 13 items in the Dementia Behavior Scale (DBD) 13. The DBD 13, a dementia behavioral disorder scale, is an evaluation index that can concisely detect peripheral symptoms (behavioral and psychological symptoms) of dementia. As described below, the language model is machine-learned by the learning unit 115. Hereinafter, the machine-learned language model will also be referred to as a "classification model."

[0048] Figure 5 shows the 13 types of items in DBD13. The items in DBD13 include, for example, 4. "Waking up in the middle of the night for no particular reason" (night waking) and 7. "Walking around excessively" (Walking around). Other items in DBD13 include 8. "Repeating the same action over and over" (repetitive behavior) and 9. "Using abusive language" (abusive language). Another item in DBD13 is 10. "Wearing inappropriate clothing that is out of place or not appropriate for the season" (Inappropriate clothing).

[0049] The classification unit 112 determines whether or not "asking the same thing over and over again" applies to item 1 of the DBD 13 for each care record and outputs the result. For example, if "asking the same thing over and over again" applies, it outputs "1", and if not, it outputs "0". The classification unit 112 similarly determines whether or not items 2 to 13 of the DBD 13 apply and outputs the result.

[0050] FIG. 6 is an explanatory diagram showing an example of the applicability determination of the items of the DBD 13 for each care record.

[0051] In the example shown in Figure 6, the nursing care record includes the time the nursing care record was entered, the record type, and the recorded text. The nursing care record in the upper row of Figure 6 includes a non-standard sentence: "The patient rested well except for going to the toilet until 11:00 PM, but from 11:00 PM, he wandered around the hallway and peeked out from the door of his room for about 30 minutes." The nursing care record also includes a non-standard sentence: "I tried to encourage him to fall asleep repeatedly, but he refused to listen." Based on these sentences, the nursing care record was classified as falling under items 4, 7, and 8 of DBD13. The classification model is thought to have classified the nursing care record into items 4, 7, and 8 of DBD13 by specifically responding to the bolded parts of the recorded text shown in Figure 6. The nursing care record in the lower row of Figure 6 includes a non-standard sentence: "The patient tried to remove his jacket and pants while on the floor." The nursing care record also includes a non-standard sentence: "I watched over him to make sure he didn't remove his clothes while on the floor." From these records, the nursing care record has been classified as falling under item 10 of DBD13. It is believed that the classification model reacted particularly to the parts written in bold in the records shown in Figure 6, and classified the nursing care record as item 10 of DBD13.

[0052] The classification unit 112 determines whether each item in the DBD 13 is applicable to each of one or more care records for a predetermined period. Specifically, the classification unit 112 determines whether each item in the DBD 13 is applicable to each of one or more care records for a predetermined period using a classification model that receives one or more care records for a predetermined period and outputs an applicability determination for each item in the DBD 13 for each care record. The predetermined period is a period of fixed length, such as three months, six months, or one year, but may also be an indefinite period with an indeterminate length. For simplicity of explanation, the following description will be given using an example where the predetermined period is three months.

[0053] The determination of applicability of each item of the DBD 13 by the classification unit 112 will be described with reference to FIG. 4 . The classification unit 112 determines the applicability of each item of the DBD 13 of the subject A for the care records of the subject A for the three months from June to March. The classification unit 112 can use a classification model generated by machine learning by the learning unit 115 to determine the applicability of each item of the DBD 13 of the subject A for the care records of the subject A for the three months from June to March. The classification unit 112 determines the applicability of each item of the DBD 13 of the subject A for the care records of the subject A for the three months from March to the present. The classification unit 112 can use a classification model generated by machine learning by the learning unit 115 to determine the applicability of each item of the DBD 13 of the subject A for the care records of the subject A for the three months from March to the present.

[0054] The change amount calculation unit 113 calculates the amount of change in the condition of subject A based on the results of the suitability determination for subject A over multiple predetermined periods. Specifically, the change amount calculation unit 113 can calculate the amount of change in the condition of the subject based on, for example, the suitability determination results for each of two consecutive predetermined periods. The condition of the subject is, for example, the evaluation of each item in the subject's DBD 13. The amount of change in the condition of the subject is, for example, the amount of change in the evaluation of each item in the subject's DBD 13. For ease of explanation, the following description will be given taking as an example a case where the condition of the subject is the evaluation of each item in the subject's DBD 13.

[0055] More specifically, the change amount calculation unit 113 calculates the amount of change in the condition of the subject from the number of times each item in the DBD 13 applies during each of a plurality of predetermined periods for the subject. The number of times each item in the DBD 13 applies during each period corresponds to the number of times a predetermined label applies. The change amount calculation unit 113 can calculate the number of times each item in the DBD 13 applies during the predetermined period of three months based on the result of the classification unit 112 determining whether each item in the DBD 13 applies during the predetermined period of three months. That is, the change amount calculation unit 113 calculates, for example, the number of times each item in the DBD 13 applies for subject A during each three-month period. The change amount calculation unit 113 can calculate the number of times each item in the DBD 13 applies for each subject during each three-month period.

[0056] The calculation of the amount of change in the subject's condition by the change amount calculation unit 113 will be specifically described. The change amount calculation unit 113 calculates the number of times each item in the DBD 13 of subject A has occurred during the three months from June to March, based on the applicability determination results by the classification unit 112 for each item in the DBD 13 of subject A during the three months from June to March. The change amount calculation unit 113 calculates the number of times each item in the DBD 13 has occurred during the three months from March to the present, based on the applicability determination results by the classification unit 112 for each item in the DBD 13 of subject A during the three months from March to the present. Hereinafter, the number of times each item in the DBD 13 has occurred during the three months from June to March is also referred to as the "number of times an item in the DBD 13 has occurred during the previous predetermined period." Furthermore, the number of times each item in the DBD 13 has occurred during the three months from March to the present is also referred to as the "number of times an item in the DBD 13 has occurred during the three months from March to the present."

[0057] The change amount calculation unit 113 estimates the amount of change in the evaluation of each item in DBD 13 of subject A based on the number of times each item in DBD 13 occurred in the previous predetermined period and the number of times each item in DBD 13 occurred in the current predetermined period. This estimates the amount of change in the evaluation of each item in DBD 13 of subject A from three months ago. Hereinafter, the amount of change in state estimated by the change amount calculation unit 113 will also be simply referred to as the "estimated amount of change in state."

[0058] The change amount calculation unit 113 estimates the amount of change in the evaluation of each item of the DBD 13, for example, as follows: As a method of estimating the amount of change in the evaluation of each item of the DBD 13, there are a case where a threshold is not used and a case where a threshold is used.

[0059] (1) When no threshold is used (1-1) If there is no difference between the number of times that item a in DBD13 occurred in the three months from June to March and the number of times that item a in DBD13 occurred in the three months from March to the present, the change in the evaluation of item a is estimated to be 0.

[0060] (1-2) If the number of times that item a in DBD13 occurred in the three months from March to the present is greater than the number of times that item a in DBD13 occurred in the three months from June to March, the change in the evaluation of item a is estimated to be +1. If the number of times that item a in DBD13 occurred in the three months from March to the present is less than the number of times that item a in DBD13 occurred in the three months from June to March, the change in the evaluation of item a is estimated to be -1.

[0061] (2) When using a threshold The change amount calculation unit 113 may estimate the amount of change in the evaluation of each item of the DBD 13 using a predetermined threshold set for the number of times each item of the DBD 13 corresponds. The change amount calculation unit 113 estimates the amount of change in the evaluation of each item of the DBD 13 using the predetermined threshold, for example, as follows. The predetermined threshold can be appropriately set by experiment or the like from the viewpoint of the accuracy of the determination result of the evaluation of the DBD 13.

[0062] (2-1) When the threshold is set to a value other than 0 for each item of DBD13 When the rate of change in the number of times item a in DBD13 occurred in the three months from March to the present relative to the number of times item a in DBD13 occurred in the three months from June to March is equal to or less than a predetermined threshold, the change amount calculation unit 113 estimates the amount of change in the evaluation of item a to be 0. This is because, when the change in the number of times a subject occurred in a certain item in DBD13 is relatively small, it is considered appropriate not to change the evaluation of that item.

[0063] (2-2) When the threshold is a threshold for increasing the amount of change in each item of DBD13 If the increase in the number of occurrences of item a in DBD13 in the three months from March to the present relative to the number of occurrences of item a in DBD13 in the three months from June to March is A times a predetermined threshold, the change amount calculation unit 113 estimates the change in the evaluation of item a as +A. If the decrease in the number of occurrences of item a in DBD13 in the three months from March to the present relative to the number of occurrences of item a in DBD13 in the three months from June to March is B times a predetermined threshold, the change amount calculation unit 113 estimates the change in the evaluation of item a as -B. For example, assume that the increase in the number of occurrences of item 5 in DBD13 in the three months from March to the present relative to the number of occurrences of item 5 in DBD13 in the three months from June to March is 40. If the predetermined threshold is 10, the increase in the number of occurrences of item 5 in DBD13 in the three months from March to the present relative to the number of occurrences of item 5 in DBD13 in the three months from June to March is 4 times the predetermined threshold. Therefore, in this case, the change in the evaluation of item 5 is estimated to be +4.

[0064] The predetermined threshold may be registered by being associated with an item of DBD 13 and stored in storage unit 13. The predetermined threshold may be changed according to the difference between the number of times an item of DBD 13 applies in a previous predetermined period and the number of times an item of DBD 13 applies in a current predetermined period. In this case, the predetermined threshold may be registered using a relational expression that indicates the relationship between the difference between the number of times an item of DBD 13 applies in a current predetermined period and the number of times an item of DBD 13 applies in a previous predetermined period and the predetermined threshold.

[0065] The change amount calculation unit 113 receives a visual inspection result by an administrator or the like of the estimated state change amount. The change amount calculation unit 113 transmits, for example, the estimated state change amount, the number of times the item in DBD13 occurred during the previous predetermined period, the number of times the item in DBD13 occurred during the current predetermined period, and the determination result of the evaluation of the item in DBD13 from the previous time to the mobile terminal 30 or the fixed terminal 20. As a result, the estimated state change amount, the number of times the item in DBD13 occurred during the previous predetermined period, the number of times the item in DBD13 occurred during the current predetermined period, and the determination result of the evaluation of the item in DBD13 from the previous time on the display unit of the mobile terminal 30 or the fixed terminal 20. The visual inspection result can be received by receiving it input to the mobile terminal 30 or the fixed terminal 20. The visual inspection result can be the state change amount resulting from the visual inspection. In the visual inspection, the estimated state change amount, the number of times an item in DBD13 occurred during the previous predetermined period, the number of times an item in DBD13 occurred during the current predetermined period, and the judgment result of the evaluation of the item in DBD13 from the previous time may be taken into consideration. The state change amount based on the visual inspection result is the state change amount that is deemed appropriate by the visual inspection. When the change amount calculation unit 113 receives a visual inspection result, it corrects the estimated state change amount to the state change amount that is the received visual inspection result. In this way, the change amount calculation unit 113 calculates the state change amount. On the other hand, when the visual inspection result is not received, the change amount calculation unit 113 calculates the state change amount by using the estimated state change amount as is.

[0066] The change amount calculation unit 113 may calculate the change amount of the subject's state by correcting the estimated change amount of the subject's state using learned LLMs (Large Language Models). In this case, the estimated state change amount, the number of times that an item in DBD13 occurred during the previous predetermined period, the number of times that an item in DBD13 occurred during the current predetermined period, and the evaluation result of the item in DBD13 from the previous period may be input to the LLM. As a result, the LLM may output the corrected change amount of the subject's state.

[0067] Note that the change amount calculation unit 113 may estimate the amount of state change by taking into consideration the previous evaluation result among the evaluation results of each item of the DBD 13 of the subject A by the state determination unit 114, which will be described later. Specifically, for example, if the evaluation of each item of the DBD 13 of the subject A was last performed three months ago and the evaluation of a certain item was the maximum value of 4, estimating the amount of state change of the item to be 1 or more would result in the evaluation value exceeding the maximum value. Therefore, in this case, in order to avoid estimating an unreasonable amount of state change, the change amount calculation unit 113 estimates the amount of state change to be a value of 0 or less.

[0068] The state determination unit 114 determines the state of the subject based on the amount of state change calculated by the change amount calculation unit 113. Specifically, the state determination unit 114 determines the evaluation of each item of the DBD 13 of the subject based on the calculated amount of change in the evaluation of each item of the DBD 13 of the subject. The state determination unit 114 determines the score of each item of the DBD 13 as the state of the subject.

[0069] 4, the determination of the subject's condition by the condition determination unit 114 will be specifically described. The condition determination unit 114 determines the current evaluation of each item of the DBD 13 of the subject A based on the amount of change in the evaluation of each item of the DBD 13 of the subject A calculated by the change amount calculation unit 113. Specifically, the condition determination unit 114 adds the amount of change in the evaluation of each item of the DBD 13 of the subject A from three months ago, calculated by the change amount calculation unit 113, to the evaluation of each item of the DBD 13 of the subject A three months ago. In this way, the condition determination unit 114 determines the evaluation of each item of the DBD 13 of the subject A currently.

[0070] Returning to Fig. 3, the description will be continued. The learning unit 115 generates a classification model for the classification unit 112 by machine learning a language model. The learning unit 115 can machine learn the language model using a combination of care records of a plurality of subjects and correct answers to the applicability information for each item in the DBD 13 of each care record corresponding to the care record.

[0071] The output control unit 116 outputs the evaluation results of each item in the subject's current DBD 13, as determined by the status determination unit 114. Specifically, the output control unit 116 transmits the evaluation results of each item in the subject's current DBD 13 to the mobile terminal 30 or the fixed terminal 20 in association with information identifying the subject. The information identifying the subject may include, for example, the subject's name, room number, etc. The output control unit 116 then displays the subject's status for each subject on the mobile terminal 30 or the fixed terminal 20. The output control unit 116 may output the number of times each item in the DBD 13 occurred during a predetermined period of three months, along with the subject's status determined by the status determination unit 116. In this case, the output control unit 116 may output the number of times each item in the DBD 13 occurred during the previous predetermined period and the number of times each item in the DBD 13 occurred during the current predetermined period. The output control unit 116 may further output the evaluation results of each item in the subject's DBD 13 from three months ago.

[0072] 7 is a diagram showing an output screen of the evaluation determination results of each item in the subject's current DBD 13, output by the output control unit 116. Hereinafter, the output screen of the evaluation determination results of each item in the subject's current DBD 13 will also be referred to as the "evaluation result output screen."

[0073] In the example shown in FIG. 7 , the current assessment result of item 9 in DBD13 for each subject is displayed together with the name of each subject as the assessment result of the current assessment of item 9 in DBD13. Here, in the column for the assessment result of the current assessment of item 9 in DBD13, the amount of change in the assessment of item 9 in DBD13 is displayed in parentheses. Furthermore, the number of times item 9 in DBD13 was met in the three months from June to March is displayed as the number of times item 9 in DBD13 was met in the previous DBD13. The number of times each item in DBD13 was met in the three months from March to the present is displayed as the number of times item 9 in DBD13 was met in the current DBD13. Furthermore, the assessment result of item 9 in DBD13 for each subject from three months ago is displayed as the assessment result of item 9 in the previous DBD13.

[0074] For subject A, the change in the evaluation of DBD13 item 9 is -3. The change in the evaluation of DBD13 item 9 is thought to have been calculated as follows: The decrease in the number of times DBD13 item 9 occurred in the three months from March to the present, compared to the number of times DBD13 item 9 occurred in the three months from June to March, is more than three times but less than four times the specified threshold of 10. Therefore, the change in the evaluation of DBD13 item 9 is calculated as -3. The current evaluation result of DBD13 item 9 is 1, which is the result of adding the change in the evaluation of DBD13 item 9, -3, to the previous evaluation result of DBD13 item 9, 4.

[0075] For subject B, the change in the evaluation of DBD13 item 9 is ±0. The change in the evaluation of DBD13 item 9 is thought to have been calculated as follows: The rate of change in the number of times DBD13 item 9 occurred in the three months from March to the present compared to the number of times DBD13 item 9 occurred in the three months from June to March is below the specified threshold of 20%, so the change in the evaluation of DBD13 item 9 is calculated as 0. The current evaluation result of DBD13 item 9 is 0 because the change in the evaluation of DBD13 item 9, 0, is added to the previous evaluation result of DBD13 item 9, 0.

[0076] For Subject C, the change in the assessment of DBD13 item 9 is ±0. The change in the assessment of DBD13 item 9 is thought to have been calculated as follows: The difference between the number of times that DBD13 item 9 occurred in the three months from June to March compared to the number of times that DBD13 item 9 occurred in the three months from March to the present is below the specified threshold of 20%, so the change in the assessment of DBD13 item 9 is calculated as 0. The current assessment result for DBD13 item 9 is 4, which is the sum of the previous assessment result for DBD13 item 9, 4, and the change in the assessment of DBD13 item 9, 0. Note that both the previous and current assessment results for DBD13 item 9 were the highest rank of 4, but the number of times that DBD13 item 9 occurred in the three months was 0. This is because, as the note states, "abusive language has become normalized" and Subject C's abusive language has become normalized, no records of abusive language are being made in the nursing care records. The text displayed in the remarks column may be registered by being input into the mobile terminal 30 by the care staff member 50, transmitted from the mobile terminal 30 to the server 10, associated with the subject, and stored in the storage unit 13. The text displayed in the remarks column may be taken into consideration when the manager, care staff member 50, or the like visually checks the amount of change in condition estimated by the change amount calculation unit 113. As a result, the information in the text displayed in the remarks column may be reflected in the visual check result.

[0077] For subject D, the change in the evaluation of item 9 of DBD13 is +2. The change in the evaluation of item 9 of DBD13 is thought to have been calculated as follows: The increase in the number of times item 9 of DBD13 occurred in the three months from March to the present compared to the number of times item 9 of DBD13 occurred in the three months from June to March is more than three times but less than four times the predetermined threshold of 10. However, the evaluation result of item 9 of DBD13 in the previous evaluation was 2. Therefore, when the change in the evaluation of item 9 of DBD13, +3, is added to the evaluation result of item 9 of DBD13 in the previous evaluation, it exceeds the maximum value of 4 for item 9 of DBD13. Therefore, to prevent the evaluation result of item 9 of DBD13 this time from exceeding 4, the change in the evaluation of item 9 of DBD13 is calculated as +2. The evaluation result of item 9 of DBD13 this time is 4, which is the result of adding the change in the evaluation of item 9 of DBD13, +2, to the evaluation result of item 9 of DBD13 in the previous evaluation, 2.

[0078] (Fixed terminal 20) The fixed terminal 20 is a PC (Personal Computer) and includes a control unit, a storage unit, a display unit, a communication unit, an input unit, etc. The functions of the control unit, storage unit, and communication unit are similar to those of the corresponding components of the server 10, and therefore description thereof will be omitted.

[0079] The input unit accepts input of various instructions and information from the administrator. The control unit performs various registrations in the server 10 based on the instructions and information accepted by the input unit. The control unit can register the subject in the server 10 in association with the care staff member 50 in charge of the subject.

[0080] (Mobile terminal 30) 8 is a block diagram showing a schematic configuration of the mobile terminal 30. The mobile terminal 30 includes a control unit 31, a wireless communication unit 32, a memory unit 33, a display unit 34, an input unit 35, and an audio input / output unit 36. These components are connected to each other via a bus. The mobile terminal 30 may be configured by a communication terminal device that can be carried by the care staff 50, such as a tablet computer, a smartphone, or a mobile phone. The mobile terminal 30 may be replaced by a fixed terminal.

[0081] The control unit 31 has the same basic components as the control unit 11 of the server 10, such as a CPU, RAM, and ROM.

[0082] The wireless communication unit 32 has a function of performing wireless communication according to standards such as Wi-Fi and Bluetooth (registered trademark), and performs wireless communication with each device directly or via an access point 41.

[0083] The storage unit 33 stores various data and programs and is configured by, for example, a flash memory.

[0084] The display unit 34 and the input unit 35 are touch panels, and a touch sensor as the input unit 35 is provided on the display surface of the display unit 34 which is made up of a liquid crystal or the like.

[0085] The voice input / output unit 36 ​​is configured with, for example, a speaker and a microphone. The voice input / output unit 36 ​​enables voice communication between the care staff 50 and other mobile terminals 30 via the wireless communication unit 32.

[0086] The control unit 31 can transmit the care record input by the care staff 50 to the input unit 35 to the server 10 via the wireless communication unit 32.

[0087] The control unit 31 receives the evaluation result output screen from the server 10 via the wireless communication unit 32. The control unit 31 causes the display unit 34 to display the received evaluation result output screen. As a result, the evaluation of each item in the DBD 13 of the subject is displayed as the subject's condition. In addition to the subject's condition, the number of times each item in the previous DBD 13 was applicable and the number of times each item in the current DBD 13 was applicable are also displayed.

[0088] Fig. 9 is a diagram showing an example of the flow of a periodic assessment to which this embodiment is applied. In Fig. 9, white circles indicate that the condition of the subject is determined manually by the care staff 50 or the like without applying this embodiment. Black circles indicate that the condition of the subject is determined according to this embodiment.

[0089] In the first periodic assessment conducted after a subject enters a nursing facility or the like, this embodiment is not applied, and the condition of the subject may be determined manually by the care staff 50 or the like. This is because, at the time of the first periodic assessment, there are no past assessment results of the subject's condition that can be used as a reference for calculating the amount of change in the subject's condition.

[0090] This embodiment can be applied to the second and subsequent periodic assessments after the subject enters a nursing home or the like.

[0091] 10 is a flowchart showing the operation of the server 10. This flowchart can be executed by the control unit 11 of the server 10 in accordance with a program.

[0092] The control unit 11 determines whether or not input of a care record has been accepted (S101). That is, the control unit 11 determines whether or not the care record input in the mobile terminal 30 has been received from the mobile terminal 30. If the control unit 11 determines that input of a care record has not been accepted (S101: NO), it executes step S101 again.

[0093] When the control unit 11 determines that input of a care record has been accepted (S101: YES), the control unit 11 stores the accepted care record in the storage unit 13 (S102).

[0094] The control unit 11 uses the classification model to determine whether each item of the DBD 13 is applicable to each care record (S103).

[0095] The control unit 11 calculates the number of times each subject applies to each item in DBD13 for each period of the multiple predetermined periods based on the results of the applicability determination for each item in DBD13 for each subject for each period of the multiple predetermined periods (S104).

[0096] The control unit 11 calculates the amount of change in the state of each subject from the number of times that each item in the DBD 13 corresponds to each subject in each of a plurality of predetermined periods (S105).

[0097] The control unit 11 determines the state of each subject based on the amount of change in the state of each subject (S106).

[0098] The control unit 11 outputs the number of times each item in the DBD 13 has been hit for a plurality of predetermined periods, along with the condition of the subject (S107).

[0099] (Variation 1) A modified example will be described. In the above-described embodiment, the recorded information is care information, the predetermined labels are the items in the DBD 13, and the subject's condition to be evaluated is the score for each item in the DBD 13. In this modified example, the recorded information is a medical report, the predetermined labels are the rank of hospital visit frequency, and the subject's condition to be evaluated is the recovery level of motor function after surgery.

[0100] 11 is a diagram showing the recorded information, the predetermined label, and the subject's condition to be evaluated in this modified example. In this modified example, the recorded information is a medical report, the predetermined label is a rank of the frequency of hospital visits, and the subject's condition to be evaluated is the level of recovery of motor function after surgery.

[0101] The record information receiving unit 111 receives, as record information, a medical report created in any text format. The medical report may be input by a doctor to each mobile terminal 30 and received by the communication unit 12. The record information receiving unit 111 receives one or more medical reports for one subject.

[0102] The classification unit 112 determines whether each rank of hospital visit frequency applies to multiple medical reports of the subject after surgery and outputs the results. That is, the classification unit 112 determines the rank of hospital visit frequency based on multiple medical reports of the subject. The rank of hospital visit frequency can be set arbitrarily. For example, the rank of hospital visit frequency can be set to 0 to 10. In this case, rank 0 can correspond to a hospital visit frequency of less than 0.1 times per 6 months. Rank 10 can correspond to a hospital visit frequency of 6 times or more per 6 months. Ranks 1 to 9 can be set according to the rank number, corresponding to a hospital visit frequency of 0.1 times or more to less than 6 times per 6 months. The classification unit 112 can use a language model such as BERT to determine whether each rank of hospital visit frequency applies to multiple medical reports of the subject.

[0103] The change amount calculation unit 113 calculates the amount of change in the subject's condition based on the result of the appropriateness determination. Specifically, the change amount calculation unit 113 may calculate the amount of change in the subject's condition based on, for example, the appropriateness determination results for each rank of the hospital visit frequency for each of two consecutive predetermined periods. If the rank of the hospital visit frequency for the six months from June to the present is lower than the rank of the hospital visit frequency for the six months from December to June, the change amount calculation unit 113 may calculate the amount of change in the subject's condition as +1. If the rank of the hospital visit frequency for the six months from June to the present is higher than the rank of the hospital visit frequency for the six months from December to June, the change amount calculation unit 113 may calculate the amount of change in the subject's condition as -1. Therefore, for example, if the rank of the hospital visit frequency for the six months from December to June is 5 and the rank of the hospital visit frequency for the six months from June to the present is 3, the change amount calculation unit 113 may calculate the amount of change in the subject's condition as +1. Note that the state determination unit 114 may calculate the amount of change in the state of the subject using a predetermined threshold value, as in the above-described embodiment.

[0104] The condition determination unit 114 determines the post-operative motor function recovery level as the condition of the subject based on the amount of change in the condition of the subject. For example, the post-operative motor function recovery level may be set to levels 0 to 5. In this case, level 5 may correspond to the highest post-operative motor function recovery level. Level 0 may correspond to the lowest post-operative motor function recovery level. Levels 2 to 4 may be set as motor function recovery levels intermediate between levels 0 and 5, depending on the level number. Specifically, the condition determination unit 114 may determine the current motor function recovery level as the value obtained by adding the amount of change in the condition of the subject to the motor function recovery level six months ago. That is, if the motor function recovery level six months ago was 3 and the calculated amount of change in the condition of the subject is +1, the condition determination unit 114 may determine the post-operative motor function recovery level as 4, obtained by adding the amount of change in the condition of the subject to the recovery level six months ago.

[0105] (Variation 2) 12 is a diagram showing the recorded information, the predetermined labels, and the condition of the subject to be evaluated in this modified example. In this modified example, the recorded information is a medical report, the predetermined labels are the ranks of motor function for each body part, and the condition of the subject to be evaluated is the disability level.

[0106] The record information receiving unit 111 receives, as record information, a medical report created in any text format. The medical report may be input by a doctor into each mobile terminal 30 and received by the communication unit 12. The record information receiving unit 111 receives one or more medical reports for one subject. The medical report is considered to include details of treatment, test results, etc.

[0107] The classification unit 112 determines whether each rank of motor function for each body part is appropriate for the multiple medical reports of the subject and outputs the result. That is, the classification unit 112 determines the rank of motor function for each body part based on the multiple medical reports of the subject. The rank of motor function for each body part can be set arbitrarily. For example, the rank of motor function for the arm can be set to ranks 0 to 3. In this case, rank 3 is the relatively highest level of motor function for the arm, and rank 0 is the relatively lowest level of motor function for the arm. The classification unit 112 can use a language model such as BERT to determine whether each rank of motor function for each body part is appropriate for the multiple medical reports of the subject.

[0108] The change amount calculation unit 113 calculates the amount of change in the subject's condition based on the result of the corresponding determination. Specifically, the change amount calculation unit 113 may calculate the amount of change in the subject's condition based on, for example, the result of the corresponding determination of each rank of motor function for the arm for each of two consecutive predetermined periods. If the rank of motor function for the arm for the six months from June to the present is lower than the rank of motor function for the arm for the six months from December to June, the change amount calculation unit 113 may calculate the amount of change in the subject's condition as +1. If the rank of motor function for the arm for the six months from June to the present is higher than the rank of motor function for the arm for the six months from December to June, the change amount calculation unit 113 may calculate the amount of change in the subject's condition as -1. Therefore, for example, if the rank of the motor function of the arm for the six months from December before to June before is 1 and the rank of the motor function of the arm for the six months from June before to the present is 2, the change amount calculation unit 113 can calculate the amount of change in the subject's condition as −1. Note that the condition determination unit 114 may calculate the amount of change in the subject's condition using a predetermined threshold value, as in the above-described embodiment.

[0109] The condition determination unit 114 determines the disability level as the condition of the subject based on the amount of change in the subject's condition. For example, the disability level may be set to levels 0 to 5. In this case, level 5 may correspond to the highest degree of disability. Level 0 may correspond to a condition equivalent to that of a healthy person. Specifically, the condition determination unit 114 may determine the current disability level as a value obtained by adding the amount of change in the subject's condition to the disability level from six months ago. That is, if the disability level from six months ago was 3 and the calculated amount of change in the subject's condition is +1, the condition determination unit 114 may determine the current disability level as 4, obtained by adding the amount of change in the subject's condition to the disability level from six months ago.

[0110] (Variation 3) 13 is a diagram showing the recorded information, predetermined labels, and the subject's condition to be evaluated in this modified example. In this modified example, the recorded information is the execution history of the training menu, the predetermined labels are the ranks of accumulated physical strain, and the subject's condition to be evaluated is the risk of injury.

[0111] The record information receiving unit 111 receives, as record information, the implementation history of a practice menu created in any text format. The implementation history of the practice menu may be input into each mobile terminal 30 by the trainer of the subject baseball player, for example, and received by the communication unit 12. In this case, the practice menu may be a baseball practice menu. The record information receiving unit 111 receives the implementation history of one or more practice menus for one subject.

[0112] The classification unit 112 determines whether each rank of accumulated physical strain is appropriate for the subject's implementation history of multiple training menus and outputs the result. That is, the classification unit 112 determines the rank of accumulated physical strain based on the subject's implementation history of multiple training menus. Each rank of accumulated physical strain can be set arbitrarily. For example, each rank of accumulated physical strain can be set to ranks 0 to 5. In this case, rank 0 can correspond to no accumulated physical strain. Rank 5 can correspond to the most severe accumulated physical strain. Ranks 1 to 4 are intermediate ranks between rank 0 and rank 5, and can correspond to the degree of accumulated physical strain depending on the rank number. The classification unit 112 can use a language model such as BERT to determine whether each rank of accumulated physical strain is appropriate for the subject's implementation history of multiple training menus.

[0113] The change amount calculation unit 113 calculates the amount of change in the subject's condition based on the result of the corresponding determination. Specifically, the change amount calculation unit 113 can calculate the amount of change in the subject's condition based on, for example, the result of the corresponding determination of each rank of the accumulated physical strain for each of two consecutive predetermined periods. If the rank of the accumulated physical strain for the six months from June to the present is lower than the rank of the accumulated physical strain for the six months from December to June, the change amount calculation unit 113 can calculate the amount of change in the subject's condition as -1. If the rank of the accumulated physical strain for the six months from June to the present is higher than the rank of the accumulated physical strain for the six months from December to June, the change amount calculation unit 113 can calculate the amount of change in the subject's condition as +1. Therefore, for example, if the rank of accumulated physical strain for the six months from December to June is 1, and the rank of accumulated physical strain for the arms for the six months from June to the present is 2, the change amount calculation unit 113 can calculate the change amount of the subject's condition as +1.

[0114] The condition determination unit 114 determines the subject's injury risk as the subject's condition based on the amount of change in the subject's condition. For example, the subject's injury risk may be leveled from 0 to 5. In this case, level 5 may correspond to the highest injury risk. Level 0 may correspond to the lowest injury risk. Levels 2 to 4 may be set according to the level number as injury risk levels intermediate between levels 0 and 5. Specifically, the condition determination unit 114 may determine the current motor function recovery level as a value obtained by adding the amount of change in the subject's condition to the subject's injury risk level from six months ago. That is, if the subject's injury risk level from six months ago was 3 and the calculated amount of change in the subject's condition is +1, the condition determination unit 114 may determine the subject's current injury risk as 4, obtained by adding the amount of change in the subject's condition to the recovery level from six months ago.

[0115] (Second embodiment) A second embodiment will be described. The present embodiment differs from the first embodiment in the following points. In the first embodiment, in order to determine whether a predetermined label applies, a language model is used that outputs a determination of whether a predetermined label applies to each nursing record input. The language model determines whether a predetermined label applies to each nursing record input and outputs the determination. On the other hand, in the present embodiment, a language model is used that outputs a determination of whether a predetermined label applies to each nursing record input and a likelihood for each nursing record input. The likelihood is then weighted to the determination of whether a predetermined label applies to each nursing record, and the amount of change in the subject's condition is calculated based on the weighted determinations of whether a predetermined label applies to a plurality of predetermined periods. In other respects, this embodiment is similar to the first embodiment, and therefore, redundant explanations will be omitted or simplified.

[0116] The classification unit 112 outputs a determination of whether a predetermined label applies to each care record and a likelihood for each care record when a care record is input. That is, the classification unit 112 classifies the care record into the corresponding label by determining whether a predetermined label applies to the care record, and calculates the likelihood corresponding to the classified label. The likelihood is a value indicating the degree of certainty that the care record corresponds to the classified label, and is a value of 1 or less. The classification unit 112 includes a machine-learned language model. Any language model can be used as long as it is a model that outputs a determination of whether a predetermined label applies to each care record when any care record is input and a likelihood for each care record. For example, BERT is used as the language model.

[0117] The change amount calculation unit 114 weights the applicability determinations made by the classification unit 112 for a predetermined period by a likelihood. The change amount calculation unit 114 calculates the amount of change in the subject's condition based on the weighted applicability determinations. Specifically, the change amount calculation unit 114 calculates the amount of change in the subject's condition, for example, as follows: Assume that a care record is determined to correspond to item 9 of DBD13, and the likelihood is calculated to be 0.9. In this case, in calculating the number of times item 9 of DBD13 corresponds to item 9 of DBD13 for the above care record, the applicability determination for the number of times 1 corresponds to item 9 of DBD13 for the above care record is weighted by 0.9, resulting in 0.9. As a result, the number of times item 9 of DBD13 corresponds to the above care record for the predetermined period can be reduced by the weighting. This allows the amount of change in the subject's condition to be calculated more accurately. As a result, the subject's condition can be determined more accurately.

[0118] The embodiment has the following advantages.

[0119] Using a machine-learned language model, a determination is made as to whether a predetermined label applies to each record information input over a predetermined period, created in an arbitrary text format. Then, based on the results of the determinations over multiple predetermined periods, the amount of change in the subject's condition is calculated, and the subject's condition is determined based on this amount of change. This reduces the impact of variations in the information granularity of the record information due to the individuality of the subject's condition based on the record information in an arbitrary text format, allowing for a consistent quantitative assessment of the subject's condition. In other words, it is possible to prevent a decrease in the accuracy of the quantitative assessment of the subject's condition due to the subject's normalized condition no longer being recorded in the record information. It is also possible to reduce variations in the quantitative assessment of the subject's condition that may occur when the quantitative assessment of the subject's condition is based on the care staff's subjective judgment.

[0120] Furthermore, in calculating the amount of change in the subject's condition, the amount of change between each period is calculated from the number of times a predetermined label is associated with each period over a plurality of predetermined periods, thereby enabling the amount of change in the subject's condition to be calculated simply and with high accuracy.

[0121] In addition, the calculated number of occurrences of a predetermined label during a predetermined period is output together with the subject's condition to be determined, thereby enabling accurate understanding of information that serves as the basis for calculating the subject's condition.

[0122] Furthermore, the language model used is a model trained by machine learning based on recorded information for a predetermined period, created in any sentence format, and the results of known classifications of predetermined labels for the recorded information for the predetermined period, enabling a simple and highly accurate quantitative assessment of the subject's condition.

[0123] Furthermore, the amount of change in the subject's state is calculated using a threshold value set for the number of times a predetermined label is associated with the subject, which allows the amount of change in the subject's state to be calculated with higher accuracy.

[0124] Furthermore, the amount of change in the subject's condition estimated based on the results of the judgments over multiple predetermined periods is corrected based on the results of visual confirmation by the user or by using a trained LLM model to calculate the amount of change in the subject's condition, thereby ensuring the validity of the amount of change in the subject's condition.

[0125] In addition, the recorded information will be used as a care record, and the subject's condition will be evaluated for each item on the Dementia Behavioral Disorders Scale, which will enable a more appropriate and accurate assessment of the subject's condition.

[0126] In addition, a language model is used that outputs the likelihood of each predetermined label for each record information input, in response to one or more record information about a subject created in any sentence format over a predetermined period of time. This allows the accuracy of determining whether a predetermined label applies to each record information to be confirmed.

[0127] Furthermore, the results of the suitability determination for a predetermined period are weighted by likelihood, and the amount of change in the subject's condition is calculated based on the weighted suitability determination results for multiple predetermined periods, thereby further improving the accuracy of determining the subject's condition.

[0128] The configuration of the information processing system 1 described above is a main configuration for explaining the features of the above embodiment, but is not limited to the above configuration and can be modified in various ways within the scope of the claims. Furthermore, configurations that are included in general information processing systems are not excluded.

[0129] For example, the functions of the server 10 may be provided in the fixed terminal 20 or any of the mobile terminals 30.

[0130] Furthermore, the server 10, the fixed terminal 20, and the mobile terminal 30 may each be configured by a plurality of devices, or any of a plurality of devices may be configured as a single device.

[0131] In addition, some steps may be omitted from the above-described flowcharts, other steps may be added, some of the steps may be executed simultaneously, or one step may be divided into multiple steps and executed.

[0132] The means and methods for performing various processes in the information processing system 1 described above can be realized by either a dedicated hardware circuit or a programmed computer. The programs may be provided by a computer-readable recording medium such as a USB memory or a DVD (Digital Versatile Disc)-ROM, or may be provided online via a network such as the Internet. In this case, the programs recorded on the computer-readable recording medium are typically transferred to and stored in a storage unit such as a hard disk. The programs may be provided as standalone application software or may be incorporated as a function into the software of a device such as a detection unit.

[0133] While embodiments of the present invention have been described and illustrated in detail, the disclosed embodiments are made for purposes of illustration and example only, and not limitation, and the scope of the present invention should be construed by the language of the appended claims. [Explanation of symbols]

[0134] 1 Information processing system, 10 servers, 11 control section, 111 Record Information Reception Department, 112 Classification Department, 113 change amount calculation unit, 114 state determination unit, 115 Learning Department, 116 output control section, 12 Communications Department, 13 storage section, 20 fixed terminals, 30 mobile devices, 31 control section, 32 Radio Communication Department, 33 Memory section, 34 Display section, 35 input section, 36 Audio input / output unit, 40 networks, 41 access points, 50 care staff.

Claims

1. a classification unit including a machine-learned language model that, in response to input of one or more pieces of record information relating to a subject over a predetermined period, created in an arbitrary sentence format, determines whether each piece of record information corresponds to a predetermined label and outputs the result; a change amount calculation unit that calculates a change amount of the subject's condition based on the result of the determination by the classification unit for a plurality of the predetermined periods; a determination unit that determines a state of the subject based on the amount of change; A condition assessment device comprising:

2. The condition evaluation device according to claim 1 , wherein the change amount calculation unit calculates the change amount between each of a plurality of the predetermined periods from the number of times the predetermined label corresponds in each of the plurality of predetermined periods.

3. 3. The condition evaluation device according to claim 2, further comprising an output unit that outputs the number of times the specified label corresponds to the specified period calculated by the change amount calculation unit, along with the condition of the subject determined by the determination unit.

4. 2. The condition assessment device of claim 1, wherein the language model is machine-learned based on the recorded information for the specified period, created in any sentence format, and a known result of a corresponding / non-corresponding determination of the specified label for the recorded information for the specified period.

5. The state evaluation device according to claim 2 , wherein the change amount calculation unit calculates the change amount using a threshold value set for the number of times the predetermined label is relevant.

6. The condition evaluation device of claim 1, wherein the change amount calculation unit calculates the change amount of the subject's condition by correcting the change amount of the subject's condition, estimated based on the results of the judgments for multiple predetermined periods, based on the results of visual confirmation by a user or by using a learned LLM model.

7. The condition evaluation device according to claim 1 , wherein the record information is a care record, and the subject's condition is an evaluation for each item of a dementia behavioral disorder scale.

8. 2. The condition assessment device according to claim 1, wherein the language model further outputs a likelihood for each of the predetermined labels for each of the record information in response to input of one or more of the record information for the predetermined period, the record information being created in any sentence format, regarding the subject.

9. The condition evaluation device of claim 8, wherein the change amount calculation unit weights the results of the determination of whether or not the condition has changed for the specified period by the likelihood, and calculates the amount of change in the condition of the subject based on the weighted results of the determination of whether or not the condition has changed for multiple specified periods.

10. a step (a) of inputting one or more pieces of record information relating to a subject over a predetermined period, written in any sentence format, and outputting a determination as to whether a predetermined label applies to each piece of record information, using a machine-learned language model to make the determination; a step (b) of calculating a change amount of the subject's condition based on the results of the determination of whether or not the subject is in a state corresponding to a plurality of the predetermined periods in the step (a); a step (c) of determining a state of the subject based on the amount of change calculated in step (b); A condition assessment method comprising:

11. 11. The condition evaluation method according to claim 10, wherein in said step (b), said amount of change between each of a plurality of said predetermined periods is calculated from the number of times said predetermined label corresponds in each of said predetermined periods.

12. 12. The condition evaluation method according to claim 11, further comprising a step (d) of outputting the number of occurrences of the specified label during the specified period output in step (b) together with the condition of the subject determined in step (c).

13. The condition evaluation method according to claim 10, wherein the language model is machine-learned based on the recorded information for the predetermined period created in an arbitrary sentence format and a known result of a corresponding or non-corresponding judgment of the predetermined label for the recorded information for the predetermined period.

14. 12. The condition evaluation method according to claim 11, wherein in the step (b), the amount of change is calculated using a threshold value set for the number of occurrences of the predetermined label.

15. 11. The condition evaluation method of claim 10, wherein in step (b), the amount of change in the subject's condition estimated based on the results of the judgments for multiple predetermined periods is calculated by correcting the amount of change in the subject's condition based on the results of visual confirmation by a user or by using a learned LLM model.

16. The condition evaluation method according to claim 10 , wherein the record information is a care record, and the condition of the subject is an evaluation for each item of a dementia behavioral disorder scale.

17. 11. The condition assessment method according to claim 10, wherein the language model further outputs a likelihood for each of the predetermined labels for each of the record information in response to input of one or more of the record information for the predetermined period that is created in any sentence format and relates to the subject.

18. 18. A condition evaluation method as described in claim 17, wherein in step (b), the results of the judgment of whether or not the subject is in a state for the specified period are weighted by the likelihood, and the amount of change in the subject's condition is calculated based on the weighted results of the judgment of whether or not the subject is in a state for a plurality of the specified periods.

19. A condition evaluation program for executing the condition evaluation method according to any one of claims 10 to 18 by a computer.

Images