JP7760920B2 - Excretion management system, excretion management method and program - Google Patents

Description

The present invention relates to an excretion management system, an excretion management method, and a program.

In recent years, the accelerating effects of the declining birthrate and aging population have led to a shortage of caregivers, and a high turnover rate due to the increased burden of care per person. In the field of caregiving, so-called "excretion care" is one of the most important care tasks for maintaining the quality of life of those receiving care. According to a 2013 Cabinet Office survey, 62.5% of 696 people with caregiving experience responded that they had difficulty accompanying patients during excretion or changing diapers. In excretion care in nursing facilities, urine-absorbing pads or sheets are typically attached to disposable diapers, and the pads or sheets are changed during regular rounds.

In nursing care, while it is important to partially reduce the burden on caregivers, it is also important to protect the dignity of the care recipient. Of course, depending on the care recipient, their ADL (Activities of Daily Living) may decline due to the effects of a serious illness, making the use of pads and sheets essential. However, toileting is an act that does not originally require the intervention of others. Having someone else handle the disposal of excrement associated with toileting not only makes the care recipient feel embarrassed, but can also result in damage to their self-esteem. For nursing care facilities, the costs of continually purchasing pads and sheets can also be a burden. Furthermore, delays in regular visits to change pads and sheets can lead to secondary problems such as bedsores or defecation. To resolve these issues, it is essential to create an invention that enables toileting prediction.

In order to reduce the burden of such caregiving work, technologies that enable the detection of excretory behavior using sensors have been developed in recent years. One such technology for detecting excretory behavior using sensors uses a thin moisture detection sensor that can be attached to diaper pads to detect the excretory behavior of the care recipient early on, thereby preventing secondary problems such as stool manipulation by users with pressure sores or dementia, and accumulating and analyzing excretion detection data to understand the user's excretory rhythm (see, for example, Patent Document 1).

In addition, in recent years, there has been gradual progress in the development of technology that uses artificial intelligence (AI), which has rapidly become more prevalent in everyday life, to predict excretion. One such AI-based technology has been disclosed, which uses nursing care records, such as meals, as training data to train AI to predict excretion (see, for example, Patent Document 2).

However, while the technology described in Patent Document 1 can detect excretory behavior using a sensor, it cannot predict excretion in advance, which poses the problem of not being able to carry out care work at the predicted time.

Furthermore, building an AI model that accurately predicts excretion requires accurate and massive amounts of training data. However, the technology described in Patent Document 2 uses nursing care records as training data, making it difficult to obtain accurate time-series data on when excretion occurred from the nursing care records. For example, even if a nursing care record is created by checking the excretion status of a care recipient during regular rounds to change pad sheets, the record often differs from the actual time of excretion. Furthermore, when highly absorbent pad sheets are used, the system is unable to recognize when the care recipient has excreted, making it difficult to obtain the accurate time of excretion. Furthermore, the technology described in Patent Document 2 uses a regression-based machine learning method that is not suitable for long-term storage of time-series data, so it can only predict the time of excretion approximately one hour before the actual excretion. In many nursing care facilities, the day's care plan is communicated at the start of work, and it is desirable for the excretion care plan to be presented at that time. Therefore, there is a problem in that it is difficult to predict excretion early and accurately.

The present invention has been made in light of the above, and aims to provide an excretion management system, excretion management method, and program that can perform excretion predictions early and accurately.

In order to solve the above-mentioned problems and achieve the object, the present invention is characterized by comprising a first acquisition unit that acquires excretion information of a care recipient detected by a measurement device as time-series data; a second acquisition unit that acquires care records of the care recipient; a first prediction unit that predicts a first excretion time of the care recipient using a first excretion prediction model generated based on the time-series data of the excretion information of the care recipient obtained from the measurement device during a first reference period; a first creation unit that creates a first excretion care plan that is an excretion care plan for the care recipient based on the first excretion time; a second prediction unit that predicts a second excretion time of the care recipient by inputting the excretion information acquired by the first acquisition unit and the care records acquired by the second acquisition unit into a second excretion prediction model generated based on the time-series data of the excretion information of the care recipient during the first reference period and the care records; and a second creation unit that creates a second excretion care plan based on the first excretion care plan and the second excretion time.

The present invention makes it possible to predict excretion early and accurately.

FIG. 1 is a diagram illustrating an example of the configuration of an excretion management system according to an embodiment. FIG. 2 is a diagram illustrating an example of a hardware configuration of the information processing apparatus according to the embodiment. FIG. 3 is a diagram illustrating an example of a functional block configuration of the information processing apparatus according to the embodiment. FIG. 4 is a diagram illustrating the operation of generating the first excretion prediction model. FIG. 5 is a diagram illustrating the operation of generating the second excretion prediction model. FIG. 6 is a diagram showing an example of a display of an excretion care plan. FIG. 7 is a flowchart showing an example of the flow of operations of the excretion management system according to the embodiment. FIG. 8 is a diagram showing an example of the configuration of an excretion management system according to a modified example.

The excretion management system, excretion management method, and program according to the present invention will be described in detail below with reference to the accompanying drawings. Furthermore, the present invention is not limited to the following embodiments, and the components in the following embodiments include those that would be easily conceived by a person skilled in the art, those that are substantially identical, and those that are within the scope of what is known as equivalents. Furthermore, various omissions, substitutions, modifications, and combinations of the components can be made without departing from the spirit of the following embodiments.

Computer software refers to programs related to computer operation and other information used for computer processing that is equivalent to a program (hereinafter, computer software will be referred to as software). Application software is a general term for software used to perform specific tasks. Meanwhile, an operating system (OS) is software that controls a computer and allows application software and other software to use computer resources. The operating system performs basic computer management and control, such as input/output control, hardware management such as memory and hard disk, and process management. Application software operates using the functions provided by the operating system. A program is a set of instructions for a computer, combined to produce a single result. Equivalents to programs are those that cannot be called programs because they are not direct instructions to a computer, but have properties similar to programs in that they define computer processing. For example, a data structure (the logical structure of data expressed by the interrelationships between data elements) is equivalent to a program.

(Configuration of excretion management system)
1 is a diagram showing an example of the configuration of an excretion management system according to an embodiment. The configuration of an excretion management system 1 according to this embodiment will be described with reference to FIG.

The excretion management system 1 shown in Figure 1 is a system that constructs an excretion prediction model for predicting excretion times based on excretion information, etc., of the care recipient, and uses the excretion prediction model to predict excretion times early and accurately. As shown in Figure 1, the excretion management system 1 has an information processing device 10 and a measurement device 20.

The information processing device 10 is an information processing device such as a PC (Personal Computer), smartphone, smartwatch, tablet device, or server device that constructs an excretion prediction model for predicting excretion times based on time-series data such as excretion information detected by the measuring device 20, predicts excretion times using the excretion prediction model, and creates an excretion care plan.

The measuring device 20 is a device that detects at least one of the defecation or urination of the care recipient being monitored as excretion information, an example of biological information. The measuring device 20 is equipped with, for example, a sensor that detects at least one of the defecation or urination of the care recipient. The sensor may be, for example, a film-type semiconductor sensor attached to a pad or sheet, or a sheet-type odor sensor placed on a bed mattress. The measuring device 20 may, for example, detect excretion information at predetermined intervals and detect the excretion information in response to user operation. The measuring device 20 also acquires the time (detection time) at which at least one of the defecation or urination of the care recipient is detected, and outputs the excretion information together with information indicating the detection and the time. This makes it possible to handle the excretion information as time-series data. The excretion information may also include the type of excretion, either defecation or urination. The measuring device 20 outputs the detected excretion information to the information processing device 10. The method and configuration for detecting excretion information in the measuring device 20 are not limited to the above configuration, and known technologies or sensing devices can be used, and the measuring device 20 may be equipped with, for example, a contact or non-contact sensor.

The measuring device 20 is not limited to directly detecting excretion information, but may also estimate at least one of the care recipient's defecation or urination by detecting one or more other types of biometric information. For example, the measuring device 20 may measure the care recipient's blood pressure and body movements as biometric information, and from this biometric information estimate at least one of the care recipient's defecation or urination, which is acquired as excretion information. The measuring device 20 may also capture an image of the care recipient and, from the image, obtain biometric information such as skeletal movement or autonomic nervous fluctuations during defecation or urination, thereby estimating at least one of defecation or urination, which is acquired as excretion information. The measuring device 20 may also obtain biometric information such as thermal displacement of the lower abdomen obtained from a thermographic image of the care recipient, and estimate at least one of defecation or urination, which is acquired as excretion information. Additionally, the measurement device 20 may measure the depth of sleep of the care recipient as biological information, and estimate at least one of the care recipient's defecation and urination from the biological information to obtain excretion information.

Furthermore, the measuring device 20 is not limited to being composed of a single device, but may be composed of multiple devices. In this case, if the measuring device 20 is composed of multiple devices, the excretion information may be transmitted individually from each device, or the excretion information may be collected by one of the devices and then transmitted from that device.

The communication method between the information processing device 10 and the measurement device 20 is not particularly limited, and various methods can be used. For example, data communication may be performed via Bluetooth (registered trademark), a wireless LAN (Local Area Network), the Internet, or the like.

As shown in FIG. 1, the excretion management system 1 includes an information processing device 10 and a measurement device 20, but the information processing device 10 alone may be considered to constitute the excretion management system 1.

(Hardware configuration of information processing device)
2 is a diagram showing an example of the hardware configuration of the information processing apparatus 10 according to the embodiment, with reference to FIG.

As shown in FIG. 2, the information processing device 10 includes a CPU (Central Processing Unit) 501, a ROM (Read Only Memory) 502, a RAM (Random Access Memory) 503, an auxiliary storage device 505, a media drive 507, a display 508 (an example of a display device), a network I/F 509, a keyboard 511, a mouse 512, a DVD (Digital Versatile Disc) drive 514, and a sensor I/F 515.

The CPU 501 is a computing device that controls the overall operation of the information processing device 10. The ROM 502 is a non-volatile storage device that stores programs for the information processing device 10. The RAM 503 is a volatile storage device used as a work area for the CPU 501.

The auxiliary storage device 505 is a storage device such as an HDD (Hard Disk Drive) or SSD (Solid State Drive) that stores various information, such as excretion information detected by the measuring device 20, as well as programs. The media drive 507 is a device that controls the reading and writing of data from and to a recording medium 506, such as a flash memory, under the control of the CPU 501.

Display 508 is a display device composed of a liquid crystal or organic EL (Electro-Luminescence) display that displays various information such as a cursor, menu, window, text, or image.

The network I/F 509 is an interface for communicating data with external devices such as the information processing device 10 via a network. The network I/F 509 is, for example, a NIC (Network Interface Card) that supports Ethernet (registered trademark) and is capable of communication in accordance with TCP (Transmission Control Protocol)/IP (Internet Protocol), etc.

The keyboard 511 is an input device for selecting letters, numbers, and various instructions, and for moving the cursor. The mouse 512 is an input device for selecting and executing various instructions, selecting processing targets, and for moving the cursor.

The DVD drive 514 is a device that controls the reading and writing of data from and to the DVD 513, such as a DVD-ROM or DVD-R (Digital Versatile Disk Recordable), which is an example of a removable storage medium.

The sensor I/F 515 is an interface for connecting to the measuring device 20 and receiving excretion information. Note that when the information processing device 10 performs data communication with the measuring device 20 via a wireless LAN, the Internet, or the like, the information processing device 10 may receive the excretion information via the network I/F 509.

The above-mentioned CPU 501, ROM 502, RAM 503, auxiliary storage device 505, media drive 507, display 508, network I/F 509, keyboard 511, mouse 512, DVD drive 514, and sensor I/F 515 are communicatively connected to each other via a bus 510 such as an address bus and a data bus.

Note that the hardware configuration of the information processing device 10 shown in Figure 2 is an example, and it is not necessary to include all of the components shown in Figure 2, or it may include other components. Furthermore, the information processing device 10 is not limited to being configured as a single information processing device as shown in Figure 2, but may also be configured from multiple network devices such as multiple information processing devices.

(Configuration and operation of functional blocks of information processing device)
Fig. 3 is a diagram showing an example of the configuration of functional blocks of an information processing device according to an embodiment. Fig. 4 is a diagram explaining the operation of generating a first excretion prediction model. Fig. 5 is a diagram explaining the operation of generating a second excretion prediction model. Fig. 6 is a diagram showing an example of the display of an excretion care plan. The configuration and operation of functional blocks of the information processing device 10 according to this embodiment will be described with reference to Figs. 3 to 6.

As shown in FIG. 3, the information processing device 10 has a first acquisition unit 101, a second acquisition unit 102, a first generation unit 103, a first prediction unit 104, a first creation unit 105, a second generation unit 106, a second prediction unit 107, a determination unit 108, a second creation unit 109, a display control unit 110, a display unit 111, and a memory unit 112.

The first acquisition unit 101 is a functional unit that acquires excretion information detected by the measuring device 20 via the sensor I/F 515. The first acquisition unit 101 stores the acquired excretion information in the memory unit 112. The excretion information sequentially acquired from the measuring device 20 by the first acquisition unit 101 can be treated as time-series data based on the detection time included in the excretion information. Note that the excretion information may also be treated as time-series data based on the time at which the excretion information was acquired by the first acquisition unit 101.

The second acquisition unit 102 is a functional unit that acquires care records for the care recipient. The care records are not limited to records of excretion, but preferably include comprehensive records such as meals and bathing. The second acquisition unit 102 may also acquire the care records by inputting data using the keyboard 511 and mouse 512, or may acquire the care records from an external device via the network I/F 509. The second acquisition unit 102 stores the acquired care records in the memory unit 112.

The first generation unit 103 is a functional unit that generates a first excretion prediction model for predicting the excretion time of the care recipient using time series data of excretion information acquired by the first acquisition unit 101 during a first reference period, which is a predetermined period. For example, the first generation unit 103 generates the first excretion prediction model by using a recursive deep learning algorithm that can store past time series data and process new events, and adjusting the depth of the network depending on the number of data points in the time series data. Examples of recursive deep learning algorithms include RNN (Recurrent Neural Network) and LSTM (Long Short-Term Memory). However, it is preferable to use LSTM, which has an advantage over RNN in long-term memory. By adopting LSTM, it is possible to efficiently learn the long-term dependency of the time series data of excretion information, enabling accurate prediction of excretion times. Here, the first reference period may be any long-term period determined in units of, for example, one month, six months, or one year, but it is preferable that it can be set arbitrarily by the user. However, if the care recipient develops a serious illness or other such condition that causes a significant change in their ADL or health condition, it is preferable not to use time series data from the period prior to that.

Specifically, as shown in FIG. 4, the first generation unit 103 inputs the time series data of excretion information for the first reference period acquired by the first acquisition unit 101, and generates a first excretion prediction model that can predict and output excretion times using a recursive deep learning algorithm such as LSTM.

Note that although the first excretion prediction model outputs the excretion time, the first excretion prediction model may also output the predicted elapsed time of excretion from the time at which the prediction is made. However, even in this case, the predicted excretion time can be obtained by adding the predicted elapsed time to the time at which the prediction is made, so it can be considered as predicting the excretion time.

The first generation unit 103 stores the generated first excretion prediction model in the storage unit 112.

In the past, excretion information based on care records was used to generate an excretion prediction model for predicting excretion times. However, in the case of excretion, it is difficult to create a care record that reflects the exact excretion time. For example, even if the excretion status of the care recipient is confirmed during regular rounds to change the pad sheet, the actual excretion time is often different. Furthermore, when a highly absorbent pad sheet is used, the care recipient may not recognize that they have excreted, making it difficult to obtain the exact excretion time. In contrast, in this embodiment, the first generation unit 103 generates the first excretion prediction model using time-series data of excretion information that reflects the exact excretion time acquired from the measurement device 20 by the first acquisition unit 101, thereby making it possible to predict the excretion time with high accuracy.

It is preferable that the time series data of the excretion information described above has already been corrected. For example, if the amount of urine excreted by the care recipient exceeds the allowable absorption capacity of the pad/sheet, it becomes difficult to obtain the exact time of excretion. Therefore, excretion information obtained after the allowable absorption capacity is exceeded will have an impact as disturbance information that does not reflect the accurate time of excretion. In such cases, when generating an excretion prediction model, it is necessary to correct the time series data of excretion information, such as by deleting the disturbance information. One method of correcting, such as deleting, disturbance information is to register the allowable absorption capacity of the pad/sheet of the care recipient in advance, provide a sensor that can detect the amount of urine excreted, and delete the time series data from the time the detected amount of urine exceeds the allowable absorption capacity to the time the diaper is changed. This correction process may be performed by the measurement device 20 or the information processing device 10. By performing such a correction process on the excretion information, accurate time series data of the excretion information can be obtained.

The first prediction unit 104 is a functional unit that predicts the excretion time (first excretion time) of the care recipient using the first excretion prediction model generated by the first generation unit 103 and stored in the memory unit 112. However, while prediction of excretion time using the first excretion prediction model is possible for a relatively long future period, it may lack prediction accuracy because it does not reflect the excretion information and care records immediately prior to the prediction time (such as the day of the care work or the day before). Therefore, in this embodiment, as described below, a second excretion prediction model is used that is generated by the second generation unit 106 and can predict the excretion time reflecting the excretion information and care records immediately prior to the prediction time. The first prediction unit 104 sends the predicted excretion time to the first creation unit 105.

The first creation unit 105 is a functional unit that creates a first excretion care plan, which is a plan for excretion care for the care recipient, based on the excretion time predicted by the first prediction unit 104. This makes it possible to create a first excretion care plan in advance before work begins, based on the excretion time predicted using the first excretion prediction model generated based on excretion information for a first reference period, which is a long-term period. This makes it possible to incorporate plans for excretion care and task adjustment by the caregiver, as well as support for the care recipient's independence, into work, for example, before the caregiver starts work. The first creation unit 105 sends the created first excretion care plan to the determination unit 108.

The second generation unit 106 is a functional unit that generates a second excretion prediction model for predicting the excretion time of the care recipient using the time series data of excretion information acquired by the first acquisition unit 101 during the above-mentioned first reference period and the care records corresponding to the first reference period acquired by the second acquisition unit 102. For example, the second generation unit 106 generates the second excretion prediction model using a statistical method capable of predicting the relationship between multiple explanatory variables and a target variable, such as multiple regression analysis, or an ensemble learning algorithm, such as random forest.

Specifically, as shown in FIG. 5, the second generation unit 106 uses the time-series data of excretion information for the first reference period acquired by the first acquisition unit 101 and the care records corresponding to the first reference period acquired by the second acquisition unit 102 to generate a second excretion prediction model that can predict and output excretion times using the statistical method or ensemble learning algorithm described above. The example of the second excretion prediction model shown in FIG. 5 is a model generated using an ensemble learning algorithm such as random forest. That is, combinations of excretion information and care records corresponding to each time are input as dataset (1), dataset (2), ..., dataset (n), and the learning process generates corresponding decision trees (1), decision tree (2), ..., decision tree (n), respectively.

Note that although the second excretion prediction model outputs the excretion time, the second excretion prediction model may also output the predicted elapsed time of excretion from the time at which the prediction is made. However, even in this case, the predicted excretion time can be obtained by adding the predicted elapsed time to the time at which the prediction is made, so it can be considered as predicting the excretion time.

The second generation unit 106 stores the generated second excretion prediction model in the memory unit 112.

In this way, in this embodiment, the second generation unit 106 generates the second excretion prediction model using time series data of excretion information that reflects the exact excretion time acquired from the measurement device 20 by the first acquisition unit 101, making it possible to predict the excretion time with high accuracy.

Furthermore, by generating a second excretion prediction model using an ensemble learning algorithm such as random forest, it is possible to utilize the causal relationships in the time-series data of care records and excretion information, making it possible to accurately predict excretion times.

It is preferable that at least the meal contents, calorie intake, and water intake are used as explanatory variables based on the care record used by the second generation unit 106 to generate the second excretion prediction model. Furthermore, it is preferable that the meal contents, calorie intake, and water intake are recorded and accumulated by the caregiver or the care recipient promptly after the care recipient eats and drinks. This makes it possible to use the second excretion prediction model to predict the excretion time according to the condition of the care recipient.

In addition, it is preferable that explanatory variables based on biological information used in the generation of the second excretion prediction model by the second generation unit 106 include weight, activity level, blood pressure variation, and body temperature, and it is also preferable that they include variation in sleep depth during the night. In this case, the measurement device 20 can be used to measure these. It is preferable that these data used as explanatory variables be obtained at least within three hours before and after the excretion time predicted by the first prediction unit 104. This makes it possible to use the second excretion prediction model to predict the excretion time according to the condition of the care recipient.

It is also known that sleep depth changes periodically during the night. In many cases, light sleep (REM sleep) and deep sleep (non-REM sleep) alternate in a 1.5-hour cycle. The relationship between these sleep cycles and excretion is evident in, for example, bedwetting. While bedwetting is often thought to occur while dreaming (during REM sleep), it is known that it most often occurs when transitioning from deep sleep to light sleep. Therefore, changes in sleep depth as biological information may be used as an explanatory variable to generate the second excretion prediction model by the second generation unit 106, making it possible to construct a more accurate second excretion prediction model.

The second prediction unit 107 is a functional unit that predicts the excretion time (second excretion time) of the care recipient by inputting the time series data of excretion information acquired by the first acquisition unit 101 during a second reference period, which is a predetermined period, and the care records corresponding to the second reference period acquired by the second acquisition unit 102, into the second excretion prediction model generated by the second generation unit 106. Here, the second reference period may be any period determined in units of, for example, one hour, half a day, or one day, but it is a period that includes at least one working day and is shorter than the above-mentioned first reference period. However, if the care recipient develops a serious illness or the like, and there is a significant change in the care recipient's ADL or health condition, it is preferable not to use the time series data from the period prior to that.

In this way, the second prediction unit 107 predicts the excretion time using the second excretion prediction model, using the excretion information and care records from the second reference period, which is a period shorter than the first reference period (the period immediately before the prediction), and therefore it is possible to predict the excretion time while reflecting the most recent condition of the care recipient.

The second prediction unit 107 sends the predicted excretion time to the determination unit 108.

The determination unit 108 is a functional unit that determines whether or not it is necessary to create an excretion care plan for the care recipient based on the first excretion care plan created by the first creation unit 105 and the excretion time predicted by the second prediction unit 107. The determination unit 108, for example, determines whether or not the time difference between the excretion time indicated in the first excretion care plan and the excretion time predicted by the second prediction unit 107 is equal to or greater than a predetermined threshold, and if it is equal to or greater than the threshold, determines that it is necessary to create an excretion care plan for the care recipient. The determination unit 108 sends the determination result to the second creation unit 109.

The second creation unit 109 is a functional unit that creates a second excretion care plan, which is a plan for excretion care for the care recipient, based on the first excretion care plan and the excretion time predicted by the second prediction unit 107, when the determination unit 108 determines that a plan for excretion care for the care recipient is necessary. In this case, the second creation unit 109 may create the second excretion care plan by changing, correcting, or editing the first excretion care plan. That is, as described above, it is possible to create a second excretion care plan by simply changing, correcting, or editing the first excretion care plan created in advance by the first creation unit 105, using the excretion time accurately predicted by the second excretion prediction model using the excretion information and care records for the second reference period, which is the shortest period immediately prior to the prediction.

The display control unit 110 is a functional unit that causes the display unit 111 to display the first excretion care plan created by the first creation unit 105 or the second excretion care plan created by the second creation unit 109. The display unit 111 is a functional unit that displays the first excretion care plan or the second excretion care plan in accordance with the control of the display control unit 110. The display unit 111 is realized by the display 508 shown in Figure 2.

6(b) shows an example in which the display 508, which is the display unit 111, displays the first excretion care plan or the second excretion care plan. Note that the display unit 111 is not limited to being realized by the display 508, and may be, for example, a smart watch 508a (an example of a display device) carried by the care recipient, as shown in FIG. 6(a). In this case, the display unit 111 displays the first excretion care plan or the second excretion care plan on the smart watch 508a, as shown in FIG. 6(a). Furthermore, the first excretion care plan and the second excretion care plan may be intended for the caregiver and the care recipient, and the plan intended for the caregiver may be displayed on the display 508, and the plan intended for the care recipient may be displayed on the smart watch 508a. This not only enables caregivers to provide efficient toileting support, but also encourages the care recipient to toilet voluntarily. Furthermore, even when an individual fails to toilet, whereas in the past it was not possible for the care recipient to even recognize that they had actually toileted, the care recipient can now recognize that they have failed to toilet, which is expected to lead to the care recipient becoming more independent with toileting.

The memory unit 112 is a functional unit that stores the excretion information acquired by the first acquisition unit 101, the care records acquired by the second acquisition unit 102, the first excretion prediction model generated by the first generation unit 103, and the second excretion prediction model generated by the second generation unit 106. The memory unit 112 may also store the excretion times predicted by the first prediction unit 104 and the second prediction unit 107, as well as the first excretion care plan created by the first creation unit 105 and the second excretion care plan created by the second creation unit 109. The memory unit 112 is realized by the RAM 503 or the auxiliary storage device 505 shown in FIG. 2.

The above-mentioned first acquisition unit 101, second acquisition unit 102, first generation unit 103, first prediction unit 104, first creation unit 105, second generation unit 106, second prediction unit 107, determination unit 108, second creation unit 109, and display control unit 110 are realized, for example, by the CPU 501 shown in FIG. 2 executing a program. Note that some or all of these functional units may be realized not by a software program, but by a hardware circuit (integrated circuit) such as an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit).

Furthermore, the functional units of the information processing device 10 shown in Figure 3 are conceptual representations of functions, and are not limited to this configuration. For example, the multiple functional units shown as independent functional units in the information processing device 10 shown in Figure 3 may be configured as a single functional unit. On the other hand, the functions of a single functional unit in the information processing device 10 shown in Figure 3 may be divided into multiple units and configured as multiple functional units.

In addition, while the first and second excretion prediction models are generated by the information processing device 10 in the above description, this is not limited to this. These models may be generated by an external device in the cloud or the like, and the information processing device 10 may use the models to predict excretion times. Specifically, the function of the first acquisition unit 101 to acquire excretion information for the first and second reference periods, the function of the second acquisition unit 102 to acquire care records corresponding to the second reference period, the function of the first generation unit 103 to generate the first excretion prediction model, and the function of the second generation unit 106 to generate the second excretion prediction model may be realized by an external device.

(Flow of operations using the excretion management system)
7 is a flowchart showing an example of the flow of operations of the excretion management system 1 according to the embodiment. The flow of operations of the excretion management system 1 according to the embodiment will be described with reference to FIG.

<Step S11>
The second acquisition unit 102 of the information processing device 10 acquires the care record for the care recipient corresponding to the first reference period. The second acquisition unit 102 stores the acquired care record in the storage unit 112. Then, the process proceeds to step S12.

<Step S12>
The first acquisition unit 101 of the information processing device 10 acquires excretion information of the care recipient detected by the measurement device 20 during the first reference period via the sensor I/F 515. The first acquisition unit 101 stores the acquired excretion information in the storage unit 112. Then, the process proceeds to step S13.

Note that steps S11 and S12 do not have to be processed in the order mentioned, and may be processed in parallel.

<Step S13>
The first generating unit 103 of the information processing device 10 generates a first excretion prediction model for predicting the excretion time of the care recipient, using time series data of excretion information for a first reference period acquired by the first acquiring unit 101 and stored in the storage unit 112. For example, the first generating unit 103 generates the first excretion prediction model by adjusting the depth of the network according to the number of data points in the time series data, using a recursive deep learning algorithm that can store past time series data and process new events.

The second generation unit 106 of the information processing device 10 generates a second excretion prediction model for predicting the excretion time of the care recipient, using the time-series data of excretion information for the first reference period acquired by the first acquisition unit 101 and the care records corresponding to the first reference period acquired by the second acquisition unit 102, which are stored in the memory unit 112. For example, the second generation unit 106 generates the second excretion prediction model using a statistical method capable of predicting the relationship between multiple explanatory variables and a target variable, such as multiple regression analysis, or an ensemble learning algorithm, such as random forest. Then, the process proceeds to step S14.

<Step S14>
The first prediction unit 104 of the information processing device 10 predicts the excretion time of the care recipient by using the first excretion prediction model generated by the first generation unit 103 and stored in the storage unit 112. Then, the process proceeds to step S15.

<Step S15>
The first creation unit 105 of the information processing device 10 creates a first excretion care plan, which is a plan for excretion care for the care recipient, based on the excretion time predicted by the first prediction unit 104. Then, the display control unit 110 of the information processing device 10 causes the display unit 111 to display the first excretion care plan created by the first creation unit 105. Then, the process proceeds to step S16.

<Step S16>
The second acquisition unit 102 acquires the care record for the care recipient corresponding to the second reference period. The second acquisition unit 102 stores the acquired care record in the storage unit 112. Then, the process proceeds to step S17.

<Step S17>
The first acquisition unit 101 acquires the excretion information detected by the measuring device 20 during the second reference period via the sensor I/F 515. The first acquisition unit 101 stores the acquired excretion information in the storage unit 112. Then, the process proceeds to step S18.

Note that steps S16 and S17 do not have to be processed in the order mentioned above and may be processed in parallel.

<Step S18>
The second prediction unit 107 of the information processing device 10 predicts the excretion time of the care recipient by inputting the time-series data of excretion information for the second reference period acquired by the first acquisition unit 101 and the care records corresponding to the second reference period acquired by the second acquisition unit 102, which are stored in the storage unit 112, into the second excretion prediction model generated by the second generation unit 106. Then, the process proceeds to step S19.

<Step S19>
The determination unit 108 of the information processing device 10 determines whether or not it is necessary to create an excretion care plan for the care recipient, based on the first excretion care plan created by the first creation unit 105 and the excretion time predicted by the second prediction unit 107. For example, the determination unit 108 determines whether or not the time difference between the excretion time indicated in the first excretion care plan and the excretion time predicted by the second prediction unit 107 is equal to or greater than a predetermined threshold, and if it is equal to or greater than the threshold, determines that the first excretion care plan for the care recipient needs to be revised. If it is determined that the first excretion care plan needs to be revised (step S19: Yes), the process proceeds to step S20. If it is determined that the first excretion care plan does not need to be revised (step S19: No), the process returns to step S14.

<Step S20>
When the determination unit 108 determines that the first excretion care plan needs to be revised, the second creation unit 109 of the information processing device 10 creates a second excretion care plan, which is a plan for excretion care for the care recipient, based on the first excretion care plan and the excretion time predicted by the second prediction unit 107. In this case, the second creation unit 109 may create the second excretion care plan by changing, correcting, or editing the first excretion care plan. Then, the display control unit 110 displays the second excretion care plan created by the second creation unit 109 on the display unit 111. Then, the process returns to step S14.

Instead of displaying the first excretion care plan in step S15, if it is determined in step S19 that no modification of the first excretion care plan is necessary, the display control unit 110 may display the first excretion care plan on the display unit 111.

As described above, in the information processing device 10 according to this embodiment, the first acquisition unit 101 acquires the excretion information of the care recipient detected by the measurement device 20 as time-series data, the second acquisition unit 102 acquires the care record of the care recipient, the first prediction unit 104 predicts the first excretion time of the care recipient using a first excretion prediction model generated based on the time-series data of the excretion information of the care recipient obtained from the measurement device 20 during a first reference period, and the first creation unit 105 creates the excretion record of the care recipient based on the first excretion time. A first excretion care plan is created, which is a plan of A, and the second prediction unit 107 predicts the second excretion time of the care recipient by inputting the excretion information acquired by the first acquisition unit 101 and the care records acquired by the second acquisition unit 102 into a second excretion prediction model generated based on the time-series data of the care recipient's excretion information for the first reference period and the care records, and the second creation unit 109 creates a second excretion care plan based on the first excretion care plan and the second excretion time. This allows excretion prediction to be performed early and accurately.

Furthermore, the first acquisition unit 101 acquires time series data of the care recipient's excretion information detected by the measurement device 20 during a first reference period, and the first generation unit 103 generates a first excretion prediction model using a recursive deep learning algorithm, using the time series data of excretion information during the first reference period acquired by the first acquisition unit 101. This makes it possible to efficiently learn the long-term dependency of the time series data of excretion information, and to accurately predict excretion times.

Furthermore, the first acquisition unit 101 acquires time series data of the excretion information of the care recipient detected by the measurement device 20 during the first reference period, the second acquisition unit 102 acquires the care record of the care recipient corresponding to the first reference period, and the second generation unit 106 generates a second excretion prediction model using an ensemble learning algorithm, using the time series data of the excretion information for the first reference period acquired by the first acquisition unit 101 and the care record corresponding to the first reference period acquired by the second acquisition unit 102. This makes it possible to utilize the causal relationship between the care record and the time series data of the excretion information, making it possible to accurately predict the time of excretion.

Furthermore, the first acquisition unit 101 acquires time series data of the care recipient's excretion information detected by the measurement device 20 during a second reference period that is shorter than the first reference period, the second acquisition unit 102 acquires the care record of the care recipient corresponding to the second reference period, and the second prediction unit 107 predicts the care recipient's second excretion time by inputting the time series data of the excretion information during the second reference period acquired by the first acquisition unit 101 and the care record corresponding to the second reference period acquired by the second acquisition unit 102 into a second excretion prediction model. This makes it possible to predict the excretion time while reflecting the care recipient's most recent condition.

The determination unit 108 determines whether or not a second excretion care plan needs to be created based on the first excretion care plan created by the first creation unit 105 and the second excretion time predicted by the second prediction unit 107. If the determination unit 108 determines that a second excretion care plan needs to be created, the second creation unit 109 creates the second excretion care plan based on the first excretion care plan and the second excretion time. As a result, if there is a discrepancy between the first excretion time included in the first excretion care plan and the predicted second excretion time, an excretion care plan can be created using the second excretion time that reflects the care recipient's most recent condition.

Furthermore, when the determination unit 108 determines that a second excretion care plan needs to be created, the display control unit 110 displays the second excretion care plan created by the second creation unit 109 on the display 508, smart watch 508a, etc. This not only enables the caregiver to provide efficient excretion support, but also makes it possible to encourage the care recipient to excrete voluntarily. Furthermore, even when an excretion failure occurs, the care recipient can recognize the failure, whereas in the past it was not possible to recognize that the excretion had actually occurred. This is expected to lead to the care recipient becoming more independent in excretion.

In the above-described embodiment, the excretion care plan for the caregiver and the excretion care plan for the care recipient are displayed on the display 508 of the information processing device 10 or the smart watch 508a, but this is not limited to this, and the excretion care plan may be displayed on the measuring device 20 or another device. Furthermore, the excretion care plan may be connected to a network camera or speaker via the network I/F 509 to notify the care recipient of the excretion time with a specific sound, etc.

(Modification)
The excretion management system according to this modification will be described, focusing on the differences from the excretion management system 1 according to the above-described embodiment. In the above-described embodiment, the information processing device 10 acquires excretion information directly from the measurement device 20 and generates a first excretion prediction model and a second excretion prediction model. In this modification, a configuration will be described in which the excretion information detected by the measurement device 20 is uploaded to the cloud via an on-premise server. The hardware configuration of the information processing device 10 according to this modification is the same as the configuration described in the above-described embodiment.

Figure 8 is a diagram showing an example of the configuration of an excretion management system according to a modified example. The configuration of the excretion management system 1a according to this modified example will be described with reference to Figure 8.

The excretion management system 1a shown in Figure 8 is a system that constructs an excretion prediction model for predicting excretion times based on excretion information, etc., of the care recipient, and uses the excretion prediction model to predict excretion times early and accurately. As shown in Figure 8, the excretion management system 1a has an information processing device 10a, a measuring device 20, a cloud 30, and an on-premises server 40.

The on-premise server 40 is a server device that collects excretion information detected by the measurement device 20. For example, the on-premise server 40 collects excretion information for a first reference period. The on-premise server 40 transmits (uploads) the collected excretion information to the cloud 30 via the cloud 30's API (Application Program Interface). The on-premise server 40 also transmits (uploads) care records corresponding to the first reference period to the cloud 30 via the cloud 30's API.

The cloud 30 generates a first excretion prediction model and a second excretion prediction model based on the excretion information and care records for the first reference period received from the on-premise server 40.

The information processing device 10a is an information processing device such as a PC, smartphone, smartwatch, tablet terminal, or server device that acquires, from the on-premises server 40, excretion information detected by the measurement device 20 during the second reference period and care records, acquires predicted excretion times using a first excretion prediction model and a second excretion prediction model, and creates and displays a first excretion care plan and a second excretion care plan based on the respective excretion times. That is, the information processing device 10a first acquires, from the on-premises server 40, the excretion information detected by the measurement device 20 during the second reference period and care records, and transmits (uploads) them to the cloud 30 via the cloud 30's API. Next, the cloud 30 predicts the care recipient's excretion time using the generated first excretion prediction model, and predicts the care recipient's excretion time by inputting the received time series data of excretion information for the second reference period and care records corresponding to the second reference period into the generated second excretion prediction model. Next, the cloud 30 transmits the excretion times predicted by the first excretion prediction model and the second excretion prediction model to the information processing device 10a. The information processing device 10a then creates a first excretion care plan, which is a plan for excretion care for the care recipient, based on the excretion times predicted by the first prediction unit 104, and creates a second excretion care plan based on the first excretion care plan and the excretion times predicted by the second excretion prediction model.

In the configuration of the excretion management system 1a according to this modified example, the on-premise server 40 performs the functions of the first acquisition unit 101 and second acquisition unit 102 of the information processing device 10 of the above-described embodiment, the cloud 30 performs the functions of the first generation unit 103, first prediction unit 104, second generation unit 106, second prediction unit 107, determination unit 108 and storage unit 112, and the information processing device 10a performs the functions of the first creation unit 105, second creation unit 109, display control unit 110 and display unit 111. This configuration of the excretion management system 1a according to this modified example not only achieves the same effects as the above-described embodiment, but also prevents the leakage of personal information and provides a desirable configuration from the standpoint of security measures by transmitting and managing private information such as excretion information to the cloud 30 without having it be acquired by the local information processing device 10a.

Note that the allocation of the functional units of the information processing device 10 shown in FIG. 3 to the information processing device 10a, cloud 30, and on-premise server 40 is not limited to the above. For example, the functions of the first creation unit 105 and second creation unit 109 of the information processing device 10 may also be performed by the cloud 30.

Furthermore, the program that realizes each function performed by the information processing device 10a may be a native application (native app), as shown in FIG. 8, or may be a web browser that runs in cooperation with a web app running on the cloud 30 and on-premise server 40.

Furthermore, each function of the above-described embodiments and variations can be realized by one or more processing circuits. Here, "processing circuit" includes a processor programmed to perform each function by software, such as a processor implemented in an electronic circuit, as well as devices such as an ASIC (Application Specific Integrated Circuit), DSP (Digital Signal Processor), FPGA (Field-Programmable Gate Array), SoC (System on a Chip), GPU (Graphics Processing Unit), and conventional circuit modules designed to perform each of the above-described functions.

In addition, in the above-described embodiments and variations, when at least one of the functional units of the information processing device 10, 10a, cloud 30, and on-premise server 40 is implemented by the execution of a program, the program is provided pre-installed in a ROM or the like. In addition, in the above-described embodiments and variations, the programs executed by the information processing device 10, 10a, cloud 30, and on-premise server 40 may be provided by being recorded in an installable or executable file format on a computer-readable recording medium such as a CD-ROM (Compact Disc Read Only Memory), a flexible disk (FD), a CD-R (Compact Disk-Recordable), a DVD, or an SD (Secure Digital) card. In addition, in the above-described embodiments and variations, the programs executed by the information processing device 10, 10a, cloud 30, and on-premise server 40 may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network. Furthermore, in the above-described embodiments and variations, the programs executed by the information processing devices 10, 10a, cloud 30, and on-premise server 40 may be configured to be provided or distributed via a network such as the Internet. Furthermore, in the above-described embodiments and variations, the programs executed by the information processing devices 10, 10a, cloud 30, and on-premise server 40 have a modular configuration including at least one of the above-described functional units, and in actual hardware, the CPU reads and executes the programs from the above-described storage devices, thereby loading and generating the above-described functional units on the main storage device.

REFERENCE SIGNS LIST 1, 1a Excretion management system 10, 10a Information processing device 20 Measuring device 30 Cloud 40 On-premise server 101 First acquisition unit 102 Second acquisition unit 103 First generation unit 104 First prediction unit 105 First creation unit 106 Second creation unit 107 Second prediction unit 108 Determination unit 109 Second creation unit 110 Display control unit 111 Display unit 112 Storage unit 501 CPU
502 ROM
503 RAM
505 Auxiliary storage device 506 Recording media 507 Media drive 508 Display 508a Smart watch 509 Network I/F
510 Bus 511 Keyboard 512 Mouse 513 DVD
514 DVD drive 515 Sensor I/F

JP 2011-019726 A Japanese Patent Application Laid-Open No. 2019-139693

Claims (9)

a first acquisition unit that acquires excretion information of the care recipient detected by the measurement device as time-series data;
a second acquisition unit that acquires a care record of the care recipient;
a first prediction unit that predicts a first excretion time of the care recipient using a first excretion prediction model generated based on time series data of excretion information of the care recipient obtained from the measurement device during a first reference period;
a first creation unit that creates a first excretion care plan, which is a plan for excretion care for the care recipient, based on the first excretion time;
a second prediction unit that predicts a second excretion time of the care recipient by inputting the excretion information acquired by the first acquisition unit and the care record acquired by the second acquisition unit into a second excretion prediction model that is generated based on the time-series data of the excretion information of the care recipient for the first reference period and the care record;
a second creation unit that creates a second excretion care plan based on the first excretion care plan and the second excretion time;
An excretion management system equipped with the first acquisition unit acquires time-series data of excretion information of the care recipient detected by the measuring device during the first reference period;
The excretion management system described in claim 1 further comprises a first generation unit that generates the first excretion prediction model using a recursive deep learning algorithm using time series data of excretion information for the first reference period acquired by the first acquisition unit. the first acquisition unit acquires time-series data of excretion information of the care recipient detected by the measuring device during the first reference period;
The second acquisition unit acquires a care record of the care recipient corresponding to the first reference period;
The excretion management system described in claim 1 or 2, further comprising a second generation unit that generates the second excretion prediction model using an ensemble learning algorithm, using time series data of excretion information for the first reference period acquired by the first acquisition unit and care records corresponding to the first reference period acquired by the second acquisition unit. the first acquisition unit acquires time-series data of excretion information of the care recipient detected by the measuring device during a second reference period that is shorter than the first reference period;
The second acquisition unit acquires a care record of the care recipient corresponding to the second reference period;
An excretion management system as described in any one of claims 1 to 3, wherein the second prediction unit predicts the second excretion time of the care recipient by inputting time series data of excretion information for the second reference period acquired by the first acquisition unit and care records corresponding to the second reference period acquired by the second acquisition unit into the second excretion prediction model. The excretion management system according to any one of claims 1 to 4, further comprising a display control unit that displays the first excretion care plan created by the first creation unit on a display device. a determination unit that determines whether or not the second excretion care plan needs to be created based on the first excretion care plan created by the first creation unit and the second excretion time predicted by the second prediction unit,
An excretion management system as described in any one of claims 1 to 5, wherein the second creation unit creates the second excretion care plan based on the first excretion care plan and the second excretion time when the judgment unit determines that the second excretion care plan needs to be created. The excretion management system of claim 6, further comprising a display control unit that, when the determination unit determines that the second excretion care plan needs to be created, displays the second excretion care plan created by the second creation unit on a display device. a first acquisition step in which a first acquisition unit acquires excretion information of the care recipient detected by a measurement device as time-series data;
a second acquisition step in which a second acquisition unit acquires a care record of the care recipient;
a first prediction step in which a first prediction unit predicts a first excretion time of the care recipient using a first excretion prediction model generated based on time series data of excretion information of the care recipient obtained from the measurement device during a first reference period;
a first creation step in which a first creation unit creates a first excretion care plan, which is a plan for excretion care for the care recipient, based on the first excretion time;
a second prediction step in which a second prediction unit predicts a second excretion time of the care recipient by inputting the excretion information acquired in the first acquisition step and the care record acquired in the second acquisition step into a second excretion prediction model generated based on the time-series data of the excretion information of the care recipient for the first reference period and the care record;
a second creation step in which a second creation unit creates a second excretion care plan based on the first excretion care plan and the second excretion time;
An excretion management method comprising: On the computer,
a first acquisition step of acquiring excretion information of the care recipient detected by a measurement device as time-series data;
a second acquisition step of acquiring a care record of the care recipient;
a first prediction step of predicting a first excretion time of the care recipient using a first excretion prediction model generated based on time series data of excretion information of the care recipient obtained from the measurement device during a first reference period;
a first creation step of creating a first excretion care plan, which is a plan for excretion care for the care recipient, based on the first excretion time;
a second prediction step of predicting a second excretion time of the care recipient by inputting the excretion information acquired in the first acquisition step and the care record acquired in the second acquisition step into a second excretion prediction model generated based on the time-series data of the excretion information of the care recipient for the first reference period and the care record;
a second creation step of creating a second excretion care plan based on the first excretion care plan and the second excretion time;
A program to execute.