Support system and support method

Abstract

We provide a support system that balances accident prevention with promoting proactive behavior among those receiving support. [Solution] The system detects when the person receiving support 1 approaches the location 3 where an accident is a concern. It will also be installed at location 3 requiring support. The detection device 10 and the notification device 32 are located in an area where the notification device 32 is placed. The notification device 32 notifies the area around the location where the person requiring support 1 is detected by the detection device 10 to be approaching the location requiring support 3. In this context, the support recipient 1 for each of the 3 support locations Based on the risk DB 52 which stores the level of accident risk, and the support recipient 1 that has approached the support location 3, the notification determination unit 53 determines whether notification by the notification device 32 is necessary based on the level of accident risk stored in the risk DB 52, and if it determines that notification is necessary, Attached to the location requiring assistance 3, which was approached by person 1 who needed assistance. The system includes a notification execution unit 54 that performs notification by a notification device 32.

Description

Technical Field

[0001] The present disclosure relates to a support system and a support method.

Background Art

[0002] In recent years, with the progress of an aging society, the demand for nursing services in elderly care facilities has been rapidly increasing. However, in response to this demand, the shortage of nursing staff has been worsening, and there are limitations to nursing solely by staff. To address this problem, it is preferable to provide high-quality services and appropriately manage the accident risks of users. In elderly care facilities, the physical and mental states of users may change rapidly, and it is required to respond quickly and appropriately to this. However, in conventional nursing methods, there is a lot of reliance on the experience and intuition of staff, and due to the shortage of personnel, the quality of accident risk management (such as accident management and activity promotion) may not be consistent. Therefore, technical support for realizing efficient and effective accident risk management is essential.

[0003] Claim 1 of Patent Document 1 describes "A monitoring system in which a user communication terminal device on the wheelchair user side and a monitor communication terminal device on the monitor side are connected to a cloud server via communication means, and on the cloud network, the wheelchair behavior conditions for the wheelchair user are pre-programmed and stored, and when the behavior conditions deviate or there is a risk of deviation in light of the wheelchair behavior information, an alert signal or behavior guideline signal is transmitted to the user communication terminal device and the monitor communication terminal device."

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the monitoring system described in Patent Document 1, for each user, such as a wheelchair user, an alert signal is sent to the user and the monitor when the user's actions deviate from pre-set behavioral conditions (paragraphs 0026, 0032, etc.). However, the range of movement of each user differs, and an accident such as a fall does not necessarily occur when the user deviates from that range of movement. Therefore, if an alert signal is always sent when a user's actions deviate from pre-set behavioral conditions, it may excessively restrict the user's actions. As a result, the user's spontaneous actions may be hindered, and the user's activities cannot be promoted. The problem that this disclosure aims to solve is to provide a support system and support method that balances accident prevention with promoting voluntary actions by those receiving support. [Means for solving the problem]

[0006] The support system of this disclosure comprises a detection device that detects when a person requiring support approaches a location requiring support where an accident is a concern, a notification device that notifies the area around the location where the notification device is installed, and a support device, wherein the support device comprises a risk database that stores the level of accident risk, which is the risk that the person requiring support detected by the detection device approaching the location requiring support will have an accident at the location requiring support, a notification determination unit that determines the necessity of notification by the notification device for the person requiring support who has approached the location requiring support, based on the level of accident risk stored in the risk database, and a notification execution unit that executes notification by the notification device when the notification determination unit determines that notification is necessary. Other solutions will be described later in the embodiments for carrying out the invention. [Effects of the Invention]

[0007] According to this disclosure, it is possible to provide a support system and support method that balances accident prevention with promoting the spontaneous actions of those receiving support. [Brief explanation of the drawing]

[0008] [Figure 1] This disclosure's support system is shown in the Block Table. [Figure 2] This is a schematic diagram showing a location database. [Figure 3] Block diagram showing the hardware configuration of the support device in this disclosure. [Figure 4] This is a schematic diagram showing a risk database. [Figure 5] This is a schematic diagram showing the target database. [Figure 6] This flowchart shows how to support this disclosure. [Modes for carrying out the invention]

[0009] Hereinafter, embodiments (referred to as "models") for implementing this disclosure will be described with reference to the drawings. Within the description of one embodiment below, other embodiments applicable to that embodiment will also be described as appropriate. This disclosure is not limited to the following embodiment, and different embodiments can be combined or modified as appropriate without significantly impairing the effects of this disclosure. In addition, the same reference numerals will be used for the same components, and redundant descriptions will be omitted. Furthermore, components having the same function will be given the same name. The illustrations are for illustrative purposes only, and for illustrative purposes, the actual configuration may be changed or some components may be omitted or modified between drawings without significantly impairing the effects of this disclosure. Also, the same embodiment does not necessarily need to have all the components.

[0010] Figure 1 is a block diagram illustrating the support system 100 of this disclosure. The support system 100 is a system that grasps the varying levels of accident risk for each person requiring support, such as an elderly person, and for each location requiring support, such as a toilet. The support system 100 can assist in the decision-making of supporters 2 (which may include supervisors) who support person 1, specifically in determining whether or not support is necessary for person 1 (whether or not the accident risk is high). In addition, the support system 100 is a system that notifies via a notification device 32 when person 1 approaches a location requiring support that has been determined to have a high accident risk for that person 1.

[0011] Examples of people requiring support (Person 1) include the elderly and people with disabilities. Examples of places requiring support (Person 3) include toilets, bedrooms, washrooms, etc. Accident risk is the risk (possibility) of accidents such as falls, tripping, and bumping into structures. In the examples disclosed here, accidents include minor accidents, serious accidents (e.g., accidents that threaten life), and near misses that could lead to accidents.

[0012] The level of accident risk varies depending on physical ability (level of care needed, level of support needed, etc.) and also on the location where the person requiring support 1 is acting (location requiring support 3). For this reason, as will be explained in detail later, location DB55 and risk DB52 are used, which store the level of accident risk associated with each location requiring support 3. For example, a person who uses a wheelchair needs to get out of their wheelchair when using the toilet. Therefore, the risk of an accident in the toilet is extremely high. However, when traveling along straight, wide, and flat corridors inside a facility such as a hospital 30, there is no need to get out of the wheelchair, and there are usually no obstacles in the corridors, so the risk of an accident is extremely low.

[0013] Furthermore, even when a user gets out of their wheelchair to use the toilet, the risk of an accident is not very high for a user with a care level of 1, but it is extremely high for a user with a care level of 4. For this reason, as will be explained in detail later, the user database 56 and risk database 52, which store the level of accident risk associated with each person being supported, are also used.

[0014] In particular, some individuals receiving support (Person 1) try to do everything themselves to avoid bothering Supporter (Person 2). Therefore, while accidents can be prevented (i.e., the risk of accidents is low) with the support of Supporter (Person 2) assisting Person 1, accidents can occur (i.e., the risk of accidents increases) when they try to do everything themselves.

[0015] Therefore, in this disclosure, the level of accident risk is determined differently for each person requiring support 1 and each location requiring support 3, and a risk database 52 is constructed. Based on the risk database 52, if the accident risk is determined to be high, a notification is issued, for example, by a notification device 32 located at location requiring support 3. This alerts person requiring support 1 at location requiring support 3, thereby reducing the accident risk. In addition, through notifications from a mobile communication terminal 21 (an example of a notification device 32) carried by the supporter 2, the supporter 2 can find out which location requiring support 3 and which person requiring support 1 needs support from the supporter 2. As a result, even if the supporter 2 is not at location requiring support 3 at the time of notification, the supporter 2 can notice the notification from the notification device 32 and rush to location requiring support 3. Thus, the occurrence of an accident can be prevented.

[0016] On the other hand, if notifications are constantly issued whenever any person requiring assistance 1 approaches the location requiring assistance 3, notifications will be issued even when the person requiring assistance 1 has a low risk of causing an accident, as described above. This may cause the person requiring assistance 1 to feel annoyed and their actions may be inhibited. Also, if a support worker 2 notices the notification from the notification device 32 and rushes to the location requiring assistance 3, the support worker 2 will realize that the person requiring assistance 1 has a low risk of causing an accident and does not require assistance. As a result, the support worker 2 may feel that the notifications are excessive and may shut down the notification device 32 or the support system 100.

[0017] Therefore, as described above, notification is made in the case of the support target person 1 who needs support at the support required location 3 according to the level of the accident risk, and no notification is made in the case of the support target person 1 who does not need support at the support required location 3. Thereby, it is possible to achieve both prevention of accidents and promotion of the spontaneous action of the support target person 1. Also, even when notified that the accident risk is high, there may be some support target persons 1 who continue their actions. However, even in this case, it is assumed that the support target person 1 who notices the notification will act with hesitation more than when not notified. For this reason, it takes time for the support target person 1 to act, and time until an accident occurs can be gained. Thereby, the supporter 2 can arrive at the support required location 3 during that time, and the occurrence of an accident can be suppressed.

[0018] Note that the accident risk may be, for example, a numerical value of the accident risk, a conceptual one such as "high", "medium", "low", etc., an either-or such as an accident actually occurred or not occurred, etc. In the example of the present disclosure, mainly, conceptual accident risks such as "high", "medium", "low", etc. are exemplified as the accident risk.

[0019] The support system 100 includes a detection device 10, a notification device 32, a support device 50, a display device 60, and an input device 61.

[0020] The detection device 10 is a device that detects that the support target person 1 has approached the support required location 3 where an accident is feared. The approach includes the case where the support target person 1 is within a predetermined distance from the support required location 3, and also includes the case where the support target person 1 has reached the support required location 3 (the distance between the support target person 1 and the support required location 3 is 0). Also, being concerned about an accident conceptually means that there is an accident risk. However, strictly speaking, an accident can occur at any place. However, it is preferable that the support required location 3 where an accident is feared does not include a place where it is considered that "an accident hardly occurs in the common sense of society". Thereby, the labor for constructing the location DB55 described later can be reduced.

[0021] Location requiring support 3 is, for example, a place where there is a risk of accident if person receiving support 1 acts alone, but where accidents can be avoided if supporter 2 acts with support. Location requiring support 3 is a place in each of the facilities 30, such as a hospital, a nursing home, or a residence for the elderly or people with disabilities. Because facilities 30 have elderly and people with disabilities, the risk of accidents is relatively higher compared to other general facilities (for example, entertainment facilities, restaurants, residences for young people, etc.). Therefore, being a place in a facility 30 can reduce the risk of accidents. Specifically, locations requiring support 3 in each of the facilities 30 are places where person receiving support 1 may change their posture, such as toilets, beds (bedrooms, etc.), and washrooms.

[0022] The detection device 10 is a device that performs at least one of the following: transmitting or receiving radio waves, or imaging. By using such a device, it is possible to detect when a person requiring support 1 approaches the location requiring support 3. Examples of transmitting and receiving devices include devices that perform either one-way communication or two-way communication. However, in the example of this disclosure, the detection device 10 used is a device that can identify person requiring support 1, that is, a device that can identify who approached the location requiring support 3. Examples of imaging devices include cameras and image sensors. When using an imaging device, it is possible to detect which person requiring support 1 approached by recognizing, for example, the face, body shape, etc., of person requiring support 1 in the image or video.

[0023] In the example of this disclosure, the detection device 10 is a transceiver that performs one-way communication and includes identification devices 11 and 31. The identification device 11 is, for example, placed on an item (e.g., a wheelchair, cane, etc.) that is carried by the person being supported 1 or moves together with the person being supported 1. The identification device 11 is, for example, a device that emits radio waves that only transmit over short distances, such as a proximity beacon or Bluetooth®, an RFID driven by induced current, or a distance sensor. On the other hand, the identification device 31 is, for example, a receiving device that receives radio waves generated from the identification device 11. If the identification device 11 is a device that emits a proximity beacon, the identification device 31 is a device that receives the transmitted proximity beacon. The identification device 31 is, for example, placed at the location requiring assistance 3. This allows the detection device 10 to detect which person being supported 1 is approaching the location requiring assistance 3.

[0024] The notification device 32 is a device that provides notifications to the area around where it is located. Preferably, the notification is given in a way that is easy for the person being notified (easy to recognize and understand) to the person being notified 1 and the supporter 2. As described above, the notification device 32 provides a notification when the person being notified 1 is at high risk of accident as they approach the area requiring support 3. This allows people around the notification device 32, including the person being notified 1, to understand that the person being notified 1 is at high risk of accident as they approach the area requiring support 3. For example, this allows the person being notified 1 to be encouraged to act with sufficient caution at the area requiring support 3. It also allows the supporter 2, who is around the notification device 32, to understand that support is needed and to provide support to the person being notified 1 at the area requiring support 3.

[0025] The notification device 32 is a device that emits at least one (or both) of sound or light towards the area surrounding the notification device 32. Sound and light travel long distances. Therefore, even if a supporter 2 (caregiver, nurse, staff (supervisor, etc.)) is not near the location where the notification device 32 is placed (for example, the location requiring support 3), a supporter 2 at a distance can recognize the sound or light. This allows a supporter 2 at a distance to rush to the location requiring support 3 and support the actions of the person requiring support 1 at the location requiring support 3. This reduces the risk of accidents. In particular, if sound is emitted, even if the location requiring support 3 is hidden by a structure and cannot be seen, the sound can still be recognized by hearing.

[0026] As for sound, it could be an alarm sound to warn or draw attention, but it could also be a voice message calling out the name of the person being assisted as they approach. For example, calling out the name of the person being assisted, such as "Please be careful, ~," or "Please wait until the assistant arrives, ~," can directly draw the attention of the person being assisted. As for light, for example, the same color could be emitted each time a warning is issued, but for example, different colors of light could be emitted depending on the level of accident risk.

[0027] The notification device 32 is placed in locations determined to have a high accident risk in the location DB 55 (DB stands for database, and the same applies hereinafter; this is an example of a database). The location DB 55 is a database that stores at least one of the following for each support-requiring location 3: information that affects the level of accident risk (hereinafter referred to as location information), or the level of accident risk. First, the location DB 55 will be explained.

[0028] Figure 2 is a schematic diagram of the location database 55. Note that the contents of the location database 55 shown in Figure 2 (especially the recorded contents) may differ from the actual contents; therefore, the contents of the location database 55 are not limited to those shown in Figure 2. The same applies to the risk database 52 and its related data, which will be explained below.

[0029] Location DB55 is a database that stores (records) information relating the level of accident risk to the locations requiring support 3. Furthermore, for each location requiring support 3, the above location information is stored in Location DB55. The location information includes, for example, the hazard level of location requiring support 3 and the number of accidents that actually occurred due to that hazard (accident count). Generally, the higher the accident count, the higher the accident risk. Therefore, in the example disclosed here, Location DB55 stores the accident count (part of the location information).

[0030] For example, the accident risk is high for support location 3, where the number of accidents is relatively high; the accident risk is also moderate for support location 3, where the number of accidents is relatively moderate; and the accident risk is low for support location 3, where the number of accidents is relatively low. The accident risk assigned to each support location 3 in location DB55 may be determined by, for example, support worker 2, or by using a predetermined machine learning model, as will be described in detail later.

[0031] For example, if the designated location for assistance (location 3) is a toilet, one of the potential hazards is a fall by the person receiving assistance (location 1). For instance, 10 incidents of falls in the toilet are stored. Another potential hazard is that the person receiving assistance (location 1) might hit their body against the toilet when they fall. For example, 7 incidents of this type of accident are stored. Therefore, the accident risk in location 3 is considered high.

[0032] Similarly, if, for example, the location requiring support 3 is a bed, a potential risk could be, for example, an accident where the person receiving support 1 falls and hits their body on the bed. For example, 5 such accidents might be stored. Therefore, the accident risk in location requiring support 3 can be said to be moderate. Furthermore, if, for example, the location requiring support 3 is a bathroom, a potential risk could be, for example, an accident where the person receiving support 1 falls and hits their body on the sink. For example, 1 such accident might be stored. Therefore, the accident risk in location requiring support 3 can be said to be low.

[0033] The level of accident risk varies depending on the actual conditions, such as the design of the support location 3. Therefore, it is preferable to determine the location DB 55 for each facility 30 that has a support location 3. Specifically, for example, it is preferable for support staff 2, etc., to extract support locations 3 that are empirically expected to have a high accident risk in facility 30, and to determine the specific dangers for each support location 3. Furthermore, if an accident occurs at one of the determined support locations 3, it is preferable to store it as an accident count, as shown in Figure 2. This improves the reliability of the location DB 55.

[0034] For example, support staff 2, for instance, inputs location information (including the number of accidents) through the input unit 57 using the input device 61, for example, at predetermined intervals or each time an accident occurs. This updates the location database 55.

[0035] Furthermore, when a new facility 30 is established, the number of accidents at the support-requiring locations 3 is always zero. For this reason, it is preferable to determine the accident risk by referring to the number of accidents at support-requiring locations 3 in other facilities 30. For example, if a support-requiring location 3 has a high number of accidents at other facilities 30, it can be inferred that the accident risk will also be high at the facility 30 where the support system 100 is newly installed. For this reason, supporters 2, etc., can input location information and accident risk for each support-requiring location 3 into the support device 50 via the input unit 57 using the input device 61. This allows for the creation of a new location database 55.

[0036] The level of accident risk can be objectively determined by a supporter, for example, based on the number of incidents. For instance, if the number of incidents is above a predetermined value, the accident risk is considered high, and if it is below that value, the accident risk is considered low. Accident risk can also be determined using a machine learning model, for example, with location information as the explanatory variable and accident risk as the dependent variable. When performing machine learning, it is preferable to use training data in which location information is labeled (annotated) with the level of accident risk. Any artificial intelligence model (e.g., convolutional neural networks (CNNs), deep neural networks (DNNs), recurrent neural networks (RNNs), etc.) can be used for machine learning.

[0037] The level of accident risk varies depending on the person being supported 1, but in places where supporters 2 etc. consider the accident risk to be high, the accident risk is relatively high for all people being supported 1, to varying degrees. For example, in a toilet, the interior of the toilet is relatively narrow, and there are many places where one could hit their head or other body parts if they fall, such as the rim of the toilet bowl or the rim of the sink. Also, in the toilet, people being supported 1 take off and put on their clothes (pants, etc.). Therefore, people being supported 1 who use a wheelchair need to stand up from their wheelchair, and people being supported 1 who use a cane need to let go of their cane. For this reason, the interior of a toilet is an environment where falls are likely to occur. Therefore, as described above, by placing the notification device 32 in such high-risk locations, the occurrence of accidents in high-risk locations can be effectively suppressed. In addition, since the notification device 32 only needs to be placed along predetermined locations DB55, the effort required to decide on the placement of the notification device 32 can be reduced.

[0038] Furthermore, the information stored in the support device 50, including the contents of the location DB 55, and the information obtained by the support system 100, are displayed on the display device 60 (an example of an output device). Display on the display device 60 is performed by the output unit 59. The display device 60 is, for example, a monitor, display, touch panel, etc. Therefore, for example, a support worker 2 can position the notification device 32 based on the contents of the location DB 55 displayed on the display device 60. Also, when installing based on the location DB 55, for example, it may be installed at locations requiring support 3 that are recorded as having a high accident risk, or at locations requiring support 3 where the number of accidents exceeds the predetermined value (threshold) for the number of accidents that can be evaluated as having a high accident risk. Furthermore, the location DB 55 may differ for each facility 30 that has locations requiring support 3, or it may be common to all facilities 30.

[0039] Returning to Figure 1, it is preferable that the notification device 32 be placed at the location requiring assistance 3 as described above. This allows a caregiver 2, who is located some distance from the notification device 32 but notices the notification from the device 32, to intuitively head towards the location requiring assistance 3 equipped with the notification device 32.

[0040] The notification device 32 notifies at least the person receiving support 1. Therefore, the notification device 32 is placed in a location where the person receiving support 1, who is at the location requiring support 3, will notice it. This allows the person receiving support 1 to be alerted as described above.

[0041] The location where the person receiving support 1 notices the device will vary depending on the location requiring support 3, but for example, it will be within the range of movement of the person receiving support 1 in location requiring support 3. In other words, the notification device 32 is placed in a location where the person receiving support 1 will notice it (a location that is easy to notice). For example, if location requiring support 3 is a toilet with a door, it can be attached around the handle or doorknob used to open and close the door (for example, on the door itself).

[0042] The notification device 32 also notifies the supporter 2. As described above, the supporter 2 is the person who provides support to the person receiving support 1 who has been notified. It is preferable that the notification device 32 be placed in a location within the support-requiring area 3 that is easily noticed by the supporter 2, who may be at a distance. Specifically, for example, it is preferable to place it in a location that is easily noticed by the supporter 2, such as on the wall or door at the entrance of the support-requiring area 3, such as a toilet, bedroom, or washroom, for example, on the side facing a corridor. By doing so, the supporter 2, who may be at a distance, can go to the support-requiring area 3, and accidents that may occur due to the person receiving support 1 acting alone in the support-requiring area 3 can be suppressed.

[0043] As will be explained in more detail later, when person 1 approaches location 3, in addition to notification from the notification device 32 located at location 3, a notification is also sent to the mobile communication terminal 21, which serves as a notification device 32 carried by caregiver 2. This means that even if caregiver 2 does not notice the notification from the notification device 32 at location 3, caregiver 2 can become aware that person 1 is approaching location 3 through their mobile communication terminal 21. The mobile communication terminal 21 can be, for example, a mobile phone, PHS, smartphone, or tablet.

[0044] In addition to being placed at the location requiring assistance 3 as described above, and being carried by the supporter 2, the notification device 32 may also be placed in, for example, the supporter 2's office (e.g., an office, nurse's station, or security guard's room in a nursing home or hospital). Since someone is usually always present in the office, the notification being issued in the office will allow someone to notice the approach to the location requiring assistance 3. Furthermore, similar to the identification device 11, the notification device 32 may be carried by the person receiving assistance 1, or placed on an item that moves together with the person receiving assistance 1 (e.g., a wheelchair or cane). This will help prevent the person receiving assistance 1 from failing to notice the notification from the notification device 32.

[0045] The support device 50 is a device that, for example, determines the level of accident risk for a person requiring support 1 approaching a location requiring support 3, and constructs a risk database 52. The support device 50 then determines the need for notification based on the constructed risk database 52, and if notification is necessary, it notifies the notification device 32. By providing such a support device 50, it is possible to achieve both accident prevention and the promotion of spontaneous actions by the person requiring support 1.

[0046] Figure 3 is a block diagram showing the hardware configuration of the support device 50 of this disclosure. The support device 50 is configured with, for example, a CPU (Central Processing Unit) 1001, RAM (Random Access Memory) 1002, ROM (Read Only Memory) 1003, I / F (Interface) 1004, bus 1005, etc. The CPU 1001, RAM 1002, ROM 1003, and I / F 1004 are connected, for example, via bus 1005. The support device 50 is realized when a predetermined control program (for example, the support method of this disclosure) stored in ROM 1003 is loaded into RAM 1002 and executed by CPU 1001. The exchange of signals and information between the support device 50 and various devices (detection device 10, notification device 32, server, etc.), external networks, etc., is performed in hardware terms through I / F 1004.

[0047] Returning to Figure 1, the support device 50 comprises a detection unit 51, a risk DB 52, a notification decision unit 53, a notification execution unit 54, the above-mentioned location DB 55, a target person DB 56, an input unit 57, a construction unit 58, and an output unit 59. At least some (or all) of the functional units constituting the support device 50 (detection unit 51, risk DB 52, notification decision unit 53, notification execution unit 54, location DB 55, target person DB 56, input unit 57, construction unit 58, and output unit 59) may be stored on a server (not shown; cloud, etc.) located remotely from the facility 30.

[0048] The detection unit 51 is a functional unit that detects when a person requiring support 1 approaches a location requiring support 3. This detection includes identifying which person requiring support 1 approached which location requiring support 3, and identifying the location 3 that was approached. As described above, the detection is performed when a detection signal from the detection device 10 (identification device 31 in the illustrated example) is input to the detection unit 51.

[0049] Risk DB52 is a database that stores the level of accident risk at location 3 for person 1 who has been detected by the detection device 10 as approaching location 3. Accident risk, as described above, is the risk that person 1 will cause an accident at location 3. For the sake of explanation, we will first refer to Risk DB52, and then explain Person DB56, which is used to construct Risk DB52.

[0050] Figure 4 is a schematic diagram of Risk DB52. Risk DB52 is a database that stores information relating the level of support needed (1), the location requiring support (3), and the level of accident risk. However, in Figure 4, the level of accident risk is stored as the notification method. For example, if the accident risk is high, a circle ("○") is recorded as a notification method using sound and light to further promote awareness. On the other hand, if the accident risk is moderate, a triangle ("△") is recorded as a notification method using only sound (or only light) to further promote awareness. For example, if the accident risk is low, there is no need for support, so an X ("×") is recorded as the notification method.

[0051] In the case of person A receiving support, in the toilet, which is designated as a location requiring support 3, the risk of accidents is high, so notification using sound and light will be provided. On the other hand, in the room (bedroom, etc.), which is also designated as a location requiring support 3, the risk of accidents is low, so no notification will be provided. In the case of person B receiving support, in the toilet, which is designated as a location requiring support level 3, the risk of accidents is moderate, so notification will be given using only sound. On the other hand, in the room (bedroom, etc.), which is also designated as a location requiring support level 3, the risk of accidents is moderate, so notification will be given using only sound.

[0052] In the case of person C receiving support, in the toilet, which is designated as a location requiring support (location 3), the risk of accidents is moderate, so notification is given using only sound. On the other hand, in the room (bedroom, etc.), which is also designated as a location requiring support (location 3), the risk of accidents is low, so no notification is given. In the case of person D receiving support, no notification will be given in the toilet, which is designated as a location requiring support 3, because the risk of an accident is low. On the other hand, in rooms (such as bedrooms), which are also designated as a location requiring support 3, notification will be given using only sound because the risk of an accident is moderate.

[0053] By using a risk database like DB52, it is possible to determine accident risks objectively and consistently. Furthermore, it is possible to objectively evaluate "what kind of person, in what kind of place, and what kind of action leads to danger," thereby reducing accidents in areas requiring assistance (area 3) where accident risk exists.

[0054] Returning to Figure 1, the support device 50 includes a construction unit 58 as described above, and the construction unit 58 is a functional unit that constructs the risk DB 52. The construction unit 58 constructs the risk DB 52 using first information and second information. The first information is information that is either a predetermined accident risk for each person receiving support 1, or information that influences the level of accident risk from past information of person receiving support 1 (person information described later). The first information is recorded, for example, in the person DB 56 described below. The second information is information that is either a predetermined accident risk for each location requiring support 3, or information that influences the level of accident risk (location information described above). The second information is recorded, for example, in the location DB 55 described above. Therefore, the construction unit 58 constructs the risk DB 52 using, for example, the location DB 55 and the person DB 56. By including the construction unit 58, it is possible to associate person receiving support 1, location requiring support 3, and the level of accident risk, and to create a risk DB 52 that serves as a criterion for notification decisions. The following describes DB56, the subject database in which the first information was recorded.

[0055] Figure 5 is a schematic diagram of the target person database 56. The target person database 56 is a database that stores information relating to the level of accident risk of a supported person 1, information from the past of supported person 1 that influences the level of accident risk, and a predetermined level of accident risk. Hereinafter, the information from the past of supported person 1 that influences the level of accident risk will be referred to as the target person information. For each supported person 1, the target person information and the accident risk predetermined by supporter 2, etc., are stored in the target person database 56. The meaning of the recorded accident risk is the same as the accident risk explained with reference to Figure 2 above.

[0056] The information about the individuals being supported includes indicators that may lead to accidents in the future. For example, the older the individual, the higher the likelihood of an accident occurring in the future, and therefore the higher the accident risk. From this perspective, the information stored for each supported individual may include, for example, age, level of care needed (or level of support needed), residency history (e.g., nursing home, care facility), vital signs (temperature, blood pressure, etc.), and although not shown in the diagram, it may also include activity records (including ADL (Activities of Daily Living)) such as meals (menu, amount eaten, etc.) and excretion (time, frequency, etc.), accident reports, near-miss accident records (number of incidents, details, etc.). In addition, although not shown in the diagram, images and videos acquired by imaging devices such as cameras and image sensors can also be stored. Furthermore, the location, speed, posture, and facial expressions of supported individual 1, obtained through analysis and estimation of these images and videos, can also be stored.

[0057] Target information is, for example, time-series information (past information) that can be obtained by comparing the usual state of each support recipient 1 with their recent state. Target information can be obtained, for example, by a caregiver or other supporter 2 having continuous contact with support recipient 1 every day while the support recipient 1 is using the facility 30 equipped with support location 3. Alternatively, for example, support recipient 1 can wear a device that acquires target information (for example, a vital signs sensor), thereby automatically acquiring the information without requiring any effort from supporter 2.

[0058] In the target person DB56, the accident risk assigned to each person receiving support may be determined, for example, by supporter 2, or by using a predetermined machine learning model. When using a machine learning model, the accident risk can be determined using a machine learning model in which target person information is the explanatory variable and accident risk is the dependent variable. Since the target person information is historical data and includes data over time, it is preferable to use a regression neural network that is excellent at learning time-series features. Furthermore, from the viewpoint of learning a vast amount of target person information over a long period of time, it is preferable to use a long-term short-term memory (LSTM) model that is capable of long-term computation.

[0059] Returning to Figure 1, the support device 50 includes an input unit 57. The input unit 57 is a functional unit that accepts input of target person information and appropriate accident risks. By including the input unit 57, target person information and accident risks can be input into the support device 50, and the target person database 56 can be constructed. Furthermore, by including the input unit 57, target person information can be input each time, for example, at predetermined intervals. This allows the target person database 56 to be modified each time input is made, and the target person database 56 can be "updated" in a sense. In addition, by recording target person information in the target person database 56, the supporter 2 can understand who they should pay particular attention to.

[0060] In the Risk DB52, the Construction Unit 58 labels each person (support recipient 1) as having a high or low accident risk. Therefore, the user, such as supporter 2, may intervene in the labeling of accident risk levels.

[0061] The support device 50 includes an output unit 59 and an input unit 57. The output unit 59 is a functional unit that outputs accident risks stored in the risk DB 52 to the user (supporter 2, etc.). Output is performed, for example, by displaying on the display device 60. This allows accident risks stored in the risk DB 52 to be proposed to the supporter 2, etc., and provides the supporter 2, etc., with an opportunity to correct the stored accident risks. The input unit 57 is a functional unit that receives accident risks stored in the risk DB 52 from the user.

[0062] The construction unit 58 outputs the accident risks stored in the risk DB 52 to users such as supporter 2 (users of the support system 100; the same applies in this disclosure) via the output unit. At the same time, the construction unit 58 stores the accident risks entered by the users via the input unit 57 in the risk DB 52. This allows for the assignment of accident risks that are more in line with reality. Furthermore, it is possible to associate the level of accident risk with notification, and to objectively grasp the criteria for whether or not to notify. This makes it possible to clearly show what kind of person doing what kind of action leads to danger. Therefore, the risk DB 52 can be accountable to, for example, the person being supported 1, supporter 2, etc. And by fulfilling this accountability, supporter 2 can understand when and why they are notified, and use the support system 100 correctly.

[0063] In other words, the risk database 52 is updated based on user input through the input unit 57. This allows for notifications based on accident risks that are more accurately reflected in reality.

[0064] The notification determination unit 53 is a functional unit that determines the necessity of notification by the notification device 32 for a person requiring assistance 1 who is approaching the location requiring assistance 3, based on the level of accident risk stored in the risk DB 52. By providing the notification determination unit 53, the necessity of notification can be determined from the level of accident risk, and if it is determined to be necessary, notification can be sent by the notification device 32.

[0065] The notification execution unit 54 is a functional unit that executes notification via the notification device 32 when the notification judgment unit 53 determines that notification is necessary. In this way, instead of uniformly notifying when a person requiring support 1 approaches a location requiring support 3, the presence or absence of notification is varied according to the level of accident risk, thereby suppressing unnecessary notifications to the person requiring support 1 and promoting spontaneous action by the person requiring support 1.

[0066] In another embodiment, the notification execution unit 54 notifies in different ways depending on the level of accident risk. In this way, the supporter 2, who is located far from the location requiring support 3, can understand the level of accident risk from the notification method of the notification device 32 (the notification method shown in Figure 4 above). For example, if the accident risk is high, the supporter 2 can rush to the location requiring support 3, while if they are notified but the accident risk is not considered to be very high, they can prioritize the supporter 2's convenience before heading to the location requiring support 3, allowing for flexible responses.

[0067] For example, if the determined accident risk is a numerical value, a predetermined first threshold that indicates a high accident risk and a predetermined second threshold that indicates a moderate accident risk are used. If the numerical value of the determined accident risk is greater than or equal to the first threshold, there is a high probability of an accident occurring, or a possibility of a serious accident occurring. In this case, for example, the notification device 32 uses both sound and light to notify the supporter 2 in a way that is particularly easy for them to notice and not overlook.

[0068] On the other hand, if the determined accident risk value is less than the first threshold but greater than or equal to the second threshold, there is a possibility of an accident occurring, but the probability is not very high, or there is a possibility of a minor accident (including near misses). In this case, for example, the notification device 32 will notify using only sound (or light only). Furthermore, if the determined accident risk value is less than the second threshold, the probability of an accident occurring, including minor accidents, is considered low. In this case, for example, the notification device 32 will not notify.

[0069] Figure 6 is a flowchart showing the method of supporting this disclosure. The method of supporting this disclosure can be implemented, for example, by the support system 100 shown in Figure 1 above. Therefore, the explanation of Figure 6 will be given with reference to Figure 1 as appropriate. Furthermore, the matters described in relation to the support system 100 above can also be applied to the method of supporting this disclosure shown in Figure 6.

[0070] The support method described herein includes steps S1 to S3. First, the detection unit 51 detects that a person to be supported 1 has approached a location requiring support 3 where an accident is suspected (step S1, detection step). This detection also includes identifying which person to be supported 1 has approached the location requiring support 3 (i.e., who approached where). Next, the notification decision unit 53 determines the need for notification regarding the person to be supported 1 who has approached the location requiring support 3, based on the level of accident risk stored in the risk DB 52 (step S2, notification decision step). The risk DB 52 is a database that stores the level of accident risk, which is the risk that a person to be supported 1, who was detected to have approached the location requiring support 3 in step S1, will have an accident at the location requiring support 3.

[0071] If the notification execution unit 54 determines in step S3 that notification is necessary (Yes in step S3), it uses the notification device 32 to send a notification to the surrounding area (step S4, notification execution step). On the other hand, if it determines in step S3 that notification is unnecessary (No in step S3), the entire flow ends.

[0072] According to the support system 100 and the support method disclosed herein, notification is given when a person requiring support 1 needs assistance at the location requiring support 3, and no notification is given when a person requiring support 1 does not need assistance at the location requiring support 3. This promotes spontaneous action by the person requiring support 1. In addition, accidents involving the person requiring support 1 can be prevented.

[0073] Furthermore, if there are support recipients 1 with different accident risks, supporter 2 can determine which support recipient 1 requires particular attention at any given time. This reduces the burden on supporter 2 while preventing accidents. Also, during times when the number of supporter 2 is small (for example, at night, early morning, or on holidays), it is difficult for supporter 2 to keep track of the actions of all support recipients 1. Therefore, by using the support system 100 and the support method disclosed herein to notify support recipients 1 whenever support is needed, accidents involving support recipients 1 can be prevented, and the workload of supporter 2 can be reduced.

[0074] Facilities 30, such as nursing homes, hospitals, residences for the elderly and people with disabilities, construction sites, and work sites, generally contain individuals 1 who are considered to be at high risk of accidents. Therefore, the application of the support device 50 and the support method disclosed herein to these facilities 30 is an area where the effects of this disclosure are particularly expected. [Explanation of Symbols]

[0075] 1. Persons eligible for support 10 Detection device 100 Support Systems 11. Identification device 2 supporters 21. Mobile communication terminal (notification device) 3 Locations requiring support 30 facilities 31 Identification device 32. Notification device 50 Support equipment 51 Detection unit 52. Risk DB (Risk Database) 53 News Judgment Department 54 Hochi Executive Department 55. Location DB (Location Database; Database) 56. Target Person Database (Target Person Database; Database) 57 Input section 60 Display device 61 Input device S1 Step (Detection Step) S2 Step (Notification Decision Step) S3 Step (Notification Execution Step)

Claims

1. A detection device that detects when a person requiring assistance approaches a location where an accident is a concern, A notification device that provides notification to the area surrounding the location where the notification device is installed, Equipped with support devices, The aforementioned support device is A risk database storing the level of accident risk, which is the risk of an accident occurring at a location requiring assistance, as detected by the detection device when the person requiring assistance approaches the location requiring assistance, A notification determination unit determines the necessity of notification by the notification device for a person requiring assistance who approaches the aforementioned location requiring assistance, based on the level of accident risk stored in the risk database. If the notification determination unit determines that notification is necessary, the notification execution unit executes the notification using the notification device, Equipped with A support system characterized by the following features.

2. A support system according to claim 1, The detection device is a device that performs at least one of the following: transmitting or receiving radio waves, or imaging. A support system characterized by the following features.

3. A support system according to claim 1, The notification device is a device that emits at least one of sound or light toward the surrounding area of ​​the notification device. A support system characterized by the following features.

4. A support system according to claim 1, The notification device is to be placed at the location requiring assistance. A support system characterized by the following features.

5. A support system according to claim 1, The aforementioned risk database is a database that stores information relating the person to be supported, the location requiring support, and the level of accident risk. The aforementioned support device is The first piece of information is, for each of the aforementioned supported persons, at least one of the predetermined accident risk, or information from the past information of the aforementioned supported persons that influences the level of the accident risk, For each of the aforementioned locations requiring assistance, the second piece of information is at least one of the predetermined accident risk or information that influences the level of the accident risk, The system includes a construction unit that uses the aforementioned risk database to construct it. A support system characterized by the following features.

6. A support system according to claim 5, The aforementioned support device is An output unit that outputs the accident risks stored in the risk database to the user, The input unit stores the accident risk in the aforementioned risk database and accepts the user's input of the accident risk, Equipped with, The aforementioned construction unit is The accident risk stored in the risk database is output to the user through the output unit, The accident risk entered by the user through the input unit is stored in the risk database. A support system characterized by the following features.

7. The support system according to claim 6, The risk database is updated based on input from the user through the input unit. A support system characterized by the following features.

8. A support system according to claim 1, The notification device is placed in a location where the accident risk is determined to be high, in a database that stores at least one of the following for each location requiring assistance: information that affects the level of accident risk, or the level of accident risk itself. A support system characterized by the following features.

9. A detection step that detects when a person requiring assistance approaches a location where an accident is a concern, A notification determination step is performed to determine the necessity of notification based on the level of accident risk stored in a risk database that stores the level of accident risk, which is the risk that the person to be supported, who was detected to have approached the location requiring support in the detection step, will have an accident at the location requiring support. The notification execution step includes, if it is determined in the notification determination step that notification is necessary, a notification execution step that executes notification to the surrounding area. A support method characterized by the following features.

Images