Information processing system and information processing method

Abstract

Providing information processing systems and methods suitable for patient care. [Solution] The information processing system includes: an acquisition unit that acquires first data associating a clinical department with a risk set representing the risk of patients related to that clinical department; a processing unit that, when it acquires information indicating that a patient hospitalized in the first clinical department is receiving treatment from the second clinical department, performs an evaluation process for each of the one or more risks included in the second risk set, which is the risk set associated with the second clinical department in the first data; and an output processing unit that presents the results of the evaluation process for the second risk set to the nurses of the first clinical department.

Description

Technical Field

[0001] The present invention relates to an information processing system, an information processing method, and the like.

Background Art

[0002] Conventionally, a method for obtaining information about patients has been known. For example, Patent Document 1 discloses a method for automatically evaluating the risk of hospital-acquired conditions (HACs).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] To provide an information processing system, an information processing method, and the like suitable for the care of patients and the like.

Means for Solving the Problems

[0005] One aspect of the present disclosure relates to an information processing system including an acquisition unit that acquires first data in which a medical department is associated with a risk set representing the risk of a patient related to the medical department, a processing unit that performs an evaluation process for each of one or more risks included in a second risk set, which is the risk set associated with a second medical department, when information indicating that a patient admitted to a first medical department is receiving medical treatment in a second medical department is acquired, and an output processing unit that presents the result of the evaluation process of the second risk set to a nurse in the first medical department.

[0006] Another aspect of this disclosure relates to an information processing method in which an information processing system acquires first data that associates a clinical department with a risk set representing the risk of patients related to the clinical department, and when it acquires information indicating that a patient hospitalized in the first clinical department is receiving treatment from the second clinical department, it performs an evaluation process for each of the one or more risks included in the second risk set, which is the risk set associated with the second clinical department in the first data, and presents the results of the evaluation process for the second risk set to the nurse of the first clinical department. [Brief explanation of the drawing]

[0007] [Figure 1] This diagram explains the background of this disclosure. [Figure 2] This is a diagram illustrating an example of the configuration of an information processing system. [Figure 3] This is a diagram illustrating an example of a bed configuration. [Figure 4] This diagram illustrates an example configuration of a bedside terminal device. [Figure 5] This diagram illustrates an example of a server system configuration. [Figure 6] This diagram illustrates an example of a terminal device configuration. [Figure 7A] This diagram shows an example of data that correlates medical departments with risk sets. [Figure 7B] This diagram shows an example of a risk set. [Figure 8] This figure shows an example of data that correlates risk, sensors, and algorithms. [Figure 9] This figure shows an example of data for setting notification preferences for each nurse. [Figure 10] This is a sequence diagram explaining the procedures related to risk (sudden change in patient condition). [Figure 11] This figure shows an example of a screen displayed on the display unit of a bedside terminal device. [Figure 12] This figure shows an example of a screen displaying the results of a risk assessment. [Figure 13A]It is a diagram showing an example of data associating a medical department with an instruction set. [Figure 13B] It is a diagram showing an example of an instruction set. [Figure 14] It is a diagram showing an example of data associating an instruction, a sensor, and an algorithm. [Figure 15] It is a sequence diagram explaining the processing related to an instruction (patient behavior). [Figure 16] It is a diagram showing an example of a screen for displaying the content of an instruction from a doctor. [Figure 17] It is a diagram showing an example of a screen for displaying the judgment result of an instruction. [Figure 18] It is a sequence diagram explaining the processing related to a patient's intention (nurse call). [Figure 19] It is a diagram showing an example of a prompt for a generation AI. [Figure 20A] It is a diagram showing an example of a screen displayed on a terminal device at the time of a nurse call notification. [Figure 20B] It is a diagram showing an example of a screen displayed on a terminal device at the time of a response to a nurse call. [Figure 20C] It is a diagram showing an example of a screen for displaying countermeasures against a nurse call. [Figure 21] It is a sequence diagram when performing processing related to risk at the time of nurse registration. [Figure 22] It is a diagram showing an example of a screen for setting notification for each risk. [Figure 23] It is a sequence diagram when performing processing related to an instruction at the time of nurse registration. [Figure 24A] It is a diagram showing an example of a screen for displaying the degree of risk as a comprehensive score. [Figure 24B] It is a diagram showing an example of a screen for displaying the degree of risk as a comprehensive score. [Figure 24C] It is a diagram showing an example of a screen for displaying the degree of risk as a comprehensive score. [Figure 25] It is a diagram explaining the classification according to the evaluation results of multiple risks. [Figure 26A]This figure shows an example screen displaying the risk assessment results indicating the patient's current condition. [Figure 26B] This figure shows an example screen displaying the results of a risk assessment indicating a sudden change in a patient's condition. [Figure 26C] This figure shows other examples of display objects in an example screen that displays the results of a risk assessment indicating a sudden change in a patient's condition. [Figure 27A] This figure shows an example of data included in an electronic medical record. [Figure 27B] This figure shows an example of data included in an electronic medical record. [Modes for carrying out the invention]

[0008] This embodiment will be described below with reference to the drawings. In the drawings, the same or equivalent elements are denoted by the same reference numerals, and redundant descriptions are omitted. This embodiment described below is not intended to unduly limit the content described in the claims. Furthermore, not all of the configurations described in this embodiment are essential components of this disclosure.

[0009] 1. Background Figure 1 is a diagram illustrating the background of the invention relating to this disclosure, and illustrates the work environment of a nurse working in a hospital or the like. As shown in Figure 1, devices for sensing the patient's condition, such as an imaging device 700 and a detection device 810, are placed near the patient's bed 100. The nurse determines the patient's condition by acquiring sensing data obtained by these devices. For example, the nurse may view the sensing data displayed on the display unit 240 of the bedside terminal device 200, or she may view the sensing data using a terminal device 600 that she carries with her (details will be described later with reference to Figure 2). Also, if it is determined that the patient's condition has suddenly changed based on the sensing data, the nurse may receive a notification to that effect.

[0010] Patients may also use a nurse call system to summon a nurse. For example, the bedside terminal device 200 may have a function to execute a nurse call, or a separate nurse call device with buttons, a microphone, a speaker, etc., may be placed near the bed 100.

[0011] As shown in Figure 1, nurses receive instructions from physicians. These instructions specify actions that nurses must follow in order to provide appropriate treatment to patients, such as "imposing dietary restrictions on the patient" or "administering specific medications to the patient." The term "physician" here may include both physicians from the department to which the nurse belongs (hereinafter also referred to as the "affiliated department") and physicians from other departments, as shown in Figure 1. Physicians from other departments may be physicians from different departments within the same hospital as the nurse, or physicians from different hospitals.

[0012] Instructions from a physician may be communicated verbally or in writing by the physician or other nurses during handover or other means. Alternatively, data representing the physician's instructions may be registered in the electronic medical record, and nurses may access the electronic medical record using devices such as bedside terminals 200 or terminals 600 to obtain the physician's instructions.

[0013] Furthermore, as shown in Figure 1, nurses can provide more appropriate care by utilizing the tacit knowledge of highly skilled nurses or by delegating some of their tasks to staff from other professions such as occupational therapists and physical therapists.

[0014] It should be noted that methods for digitizing the tacit knowledge of experienced nurses are already known, and similar methods may be applied to the sharing of tacit knowledge in this embodiment. Digitization here may involve, for example, creating training data that represents the tacit knowledge of experienced nurses using digital data, or creating application software (hereinafter simply referred to as "application") that allows users to perform care according to the training data. By using such an application, it becomes possible for less experienced nurses to perform the same actions as experienced nurses.

[0015] As described above, nurses must comply with physicians' instructions and take appropriate action when changes in a patient's condition are observed based on sensing data, or when a nurse call is made. To achieve this, nurses are required to clearly understand the physician's instructions, know how to respond appropriately to sudden changes in a patient's condition, and appropriately understand the patient's wishes when a nurse call is made. However, some of this information is not easy for nurses to grasp, making it difficult for them to take appropriate action.

[0016] Information that nurses may find difficult to grasp includes instructions (patient behavior and changes in instructions), risks (sudden changes in patient condition), and patient wishes.

[0017] For example, if a patient has multiple illnesses, the doctor's instructions regarding the patient may include instructions related to the nurse's department and instructions related to other departments. For instance, if a patient being treated for diabetes develops a severe condition and is hospitalized in the ophthalmology department due to cataracts, the ophthalmology nurse will receive instructions regarding cataracts from the ophthalmologist in their department, as well as instructions regarding diabetes from the internist, and will monitor the patient's behavior to ensure compliance with both sets of instructions. In this example, the ophthalmology nurse needs to monitor the patient's behavior based on the instructions regarding dietary and fluid restrictions for diabetes. However, ophthalmology nurses are often unfamiliar with caring for diabetic patients, making it difficult for them to understand how to comply with the instructions (how to monitor the patient's behavior).

[0018] Furthermore, in the acute phase, when a patient's condition is likely to change rapidly, physicians' orders may change suddenly. In this case, regardless of the medical department, it is not easy for nurses to grasp the changes in physicians' orders (changes in orders). For example, nurses check the latest orders when handover takes place, but if the orders change after that handover, it is not easy for nurses to grasp the changes. It is assumed that the changed orders will be registered in the electronic medical record, but for nurses to grasp these orders, they would need to proactively check the details of the electronic medical record, and considering the workload, it is difficult to force nurses to do such checking.

[0019] Furthermore, patient risks (which may specifically include risks related to changes in condition, and which will also be referred to as sudden changes in patient condition hereafter) can be detected using sensing data, etc., as described above, and by having nurses take appropriate measures according to the risks, it becomes possible to suppress the worsening of the patient's condition. However, as can be seen from the examples of internal medicine and ophthalmology above, nurses may have difficulty understanding risks that they do not frequently encounter in their department, and therefore may also have difficulty understanding countermeasures. For example, if a patient with asthma is hospitalized in the ophthalmology department, the ophthalmology nurse can understand the patient's asthma using the electronic medical record, but even if they detect symptoms such as coughing, they may not immediately connect the detection result with asthma. Moreover, the difficulty in linking symptoms and risks (such as asthma) is particularly noticeable at night or when nurses who are not assigned to the patient are taking care of them. This is because nurses do not necessarily have the information necessary to interpret what the patient's behavior means.

[0020] Furthermore, when a patient uses the nurse call system, the patient's requests can be diverse, making it difficult to understand the patient's needs (or understand their wishes) solely from the call system. For example, even if a patient complains of a headache during a nurse call, the nurse may not be able to identify the underlying disease or determine the necessary response. This is particularly noticeable when the patient has multiple illnesses, as in the example above.

[0021] Therefore, the information processing system 10 according to this embodiment enables nurses to easily recognize information that is difficult to grasp. Specifically, the information processing system 10 automatically activates sensors and performs automatic monitoring to automatically monitor "patient behavior that is difficult to grasp" in relation to physician's instructions. The information processing system 10 also automatically evaluates "sudden changes in the patient's condition that are difficult to grasp" in relation to risks and notifies the nurse of the evaluation results. Furthermore, the information processing system 10 automatically interprets "patient intentions that are difficult to grasp" in relation to patient intentions (nurse calls) and notifies the nurse of the interpretation results.

[0022] The following will first describe the configuration examples of the information processing system 10 and the devices included in the information processing system 10. Then, the processing for making nurses aware of risks (sudden changes in patient condition), instructions (patient behavior), and patient wishes will be explained. Furthermore, specific examples of risk output methods (display methods), including risks that are relatively easy to grasp, will be described.

[0023] 2. Example System Configuration Figure 2 is a diagram illustrating an example configuration of the information processing system 10 according to this embodiment. The information processing system 10 includes a bed 100, a bedside terminal device 200, a server system 300, an electronic medical record server 400, a station terminal device 500, and a terminal device 600 (portable terminal device). The information processing system 10 may also include an imaging device 700, a detection device 810, a measuring device 820, an authentication card 830, etc. However, the configuration of the information processing system 10 is not limited to the example in Figure 2, and various modifications can be made, such as omitting some components or adding other components. For example, the information processing system 10 according to this embodiment corresponds to the server system 300 in Figure 2, and the other components may be external devices directly or indirectly connected to the information processing system 10 of this embodiment. Furthermore, the information processing system 10 according to this embodiment may be realized by distributed processing of two or more of the devices shown in Figure 2. Also, the bedside terminal device 200 may be omitted from the configuration shown in Figure 2. For example, some or all of the processing performed by the bedside terminal device 200 described below may be performed by other devices such as the terminal device 600. Furthermore, modifications such as omitting or adding components to the configuration are possible, as is also the case with Figure 3 and other diagrams described later.

[0024] The bed 100 is bedding used by a patient. For example, a mattress 170 is placed on the bed 100, and the patient lies on the mattress 170. An imaging device 700 and a detection device 810 are provided near the bed 100. The imaging device 700 is a camera fixed to, for example, the frame of the bed 100, but it may also be fixed to the wall of the patient's room. The detection device 810 is a sheet-like or plate-like device provided between, for example, the bed 100 and the mattress 170.

[0025] The imaging device 700 outputs captured images using sensors such as a CCD (Charge-Coupled Device) image sensor or a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor. The captured images here may be moving images or still images. The imaging device 700 is installed near the bed 100 and, for example, captures images of a patient in a hospital room.

[0026] The detection device 810 is a device that senses information related to the patient's sleep. The detection device 810 includes a pressure sensor (e.g., a pneumatic sensor) that outputs a pressure value. When the user lies down, the detection device 810 detects the user's body vibrations (body movement, vibration) via the mattress 170. Based on the detected body vibrations, the detection device 810 obtains information related to respiratory rate, heart rate, activity level, posture, wakefulness / sleep, and getting out of bed / staying in bed. The detection device 810 may output sensor data representing body vibrations, and other devices such as the server system 300 may perform processing to obtain information such as respiratory rate based on this sensor data. The following describes an example in which the detection device 810 outputs information such as respiratory rate.

[0027] For example, the detection device 810 may analyze the periodicity of body movement and calculate the respiratory rate and heart rate from the peak frequency. The periodicity analysis may be performed using, for example, a Fourier transform. The respiratory rate is the number of breaths per unit time. The heart rate is the number of heartbeats per unit time. The unit time is, for example, one minute. The detection device 810 may also detect body vibrations per sampling unit time and output the number of detected body vibrations as the activity level. Furthermore, when the user gets out of bed, the detected pressure value decreases compared to when the user is in bed, so the detection device 810 may determine whether the user is in bed or out of bed based on the pressure value and its time-series changes. However, the method for determining whether the user is in bed or out of bed is not limited to this, and various modifications such as detecting vibrations can be implemented. The detection device 810 may also determine non-REM sleep and REM sleep, and determine the depth of sleep. The sleep-related determination may be performed based on the respiratory rate and heart rate, based on the amount of body movement (e.g., activity level), or using both.

[0028] The detection device 810 outputs biological information (such as respiration and heart rate) representing the patient's biological activity status, and sleep information (such as sleep / wake status, sleep depth, and whether the patient is out of bed or in bed) to the server system 300. The detection device 810 may output the sensing results to the server system 300 via the bedside terminal device 200, or it may output the sensing results to the server system 300 without going through the bedside terminal device 200.

[0029] A device equipped with a load sensor may be used as the bed 100. In this case, the bed 100 may output at least one of the patient's biological information and sleep information. The bed 100 may also output information representing changes in the patient's center of gravity and information representing their sleeping posture. Furthermore, the detection device 810 and the bed 100 with a load sensor may be used in combination.

[0030] The bedside terminal device 200 is a device that has a patient status notification function (notification function) and may be connected to the detection device 810 or to other devices included in the information processing system 10 via a network.

[0031] For example, the bedside terminal device 200 may include a display device and a connection device. The display device is, for example, a tablet-type display terminal that displays various information and accepts input for various operations. The connection device is a hub device for connecting the display device and the various devices. For example, the connection device may be connected to a detection device 810 to continuously acquire the patient's biometric information. The connection device may also receive biometric information from various measuring devices 820 (for example, a thermometer) or from a device worn by the patient (for example, a wristwatch-type wearable measuring device). The connection device may also perform authentication processing (for example, patient authentication or login processing for nurses, etc.) by reading an authentication card 830. The connection device may perform authentication processing using NFC (Near Field Communication), which is an example of short-range wireless communication, or it may perform authentication processing using other methods such as barcodes, infrared, or IC tags. For example, a user (nurse, doctor, care staff) can check the values ​​of biometric information and notification content by logging in, and can register that information in the electronic medical record as needed.

[0032] The network is connected to, for example, a server system 300, an electronic medical record server 400, a station terminal device 500, and a terminal device 600.

[0033] The server system 300 is a server that provides various services and may be connected to the LAN within the hospital or facility, or it may be located externally via the internet.

[0034] The server system 300 may consist of one server or may include multiple servers. For example, the server system 300 may include a database server and an application server. The database server stores various data such as biometric information and sleep information. The application server performs processing described later using Figures 10, 15, and 18. The multiple servers here may be physical servers or virtual servers. If virtual servers are used, they may be located on a single physical server or distributed across multiple physical servers. As described above, the specific configuration of the server system 300 in this embodiment can be modified in various ways.

[0035] The electronic medical record server 400 is a server that stores electronic medical record information about patients. The electronic medical record server 400 is typically a server connected to a network within the hospital or facility, but it may also be an external cloud server, for example. Electronic medical records can contain various types of information. Electronic medical records, as shown in Figures 27A and 27B, for example, include basic information, medical history, diagnostic information, medication information, test results, consultation information, treatment information, rehabilitation history, vaccination history, lifestyle information, family history, consent forms and signatures, insurance and medical information, medical team information, social background, mental health, emergency response plan, medical resource history, communication with family, infection control, risk assessment, physician's instructions, measurement results, response history, nurse call history, rehabilitation care history, transfer / referral letter history, therapeutic guidance history, patient complaint / chief complaint history, patient behavior information, blood glucose level information, diet therapy information, water intake information, electrocardiogram information, pulse information, information on visits to multiple departments, interdepartmental coordination information, behavioral video recordings, video file storage information, physician / nurse responses based on videos, respiratory rate information, arterial blood oxygen saturation information, excretion information, room temperature and humidity information, etc. However, Figures 27A and 27B are specific examples of information included in electronic medical records, and various modifications are possible, such as omitting some information or adding other information. Details of each piece of information are shown in Figures 27A and 27B, so a detailed explanation is omitted here.

[0036] The electronic medical record system of this embodiment may include other systems. Examples of other systems include nursing support systems, rehabilitation department systems, critical care department systems, and nutrition department systems.

[0037] The station terminal device 500 is a terminal device installed in the nurses' station or management room. By using the station terminal device 500, users such as nurses can check the status of the bedside terminal device 200 (i.e., the status of the patient corresponding to the bedside terminal device 200) from a location other than the patient's room.

[0038] The terminal device 600 is a portable device used by medical staff, such as doctors and nurses. The terminal device 600 for medical staff connects to a network (e.g., LAN) via a wireless connection. By using the terminal device 600, medical staff can check information from the bedside terminal device 200 from various locations within the hospital.

[0039] Furthermore, at least one of the station terminal device 500 and the terminal device 600 may perform a process to notify (inform) the nurse of the patient's condition. The notification process performed by the station terminal device 500 and / or the terminal device 600 may be initiated by the bedside terminal device 200 or by the server system 300.

[0040] Next, using Figures 3-6, we will explain an example of the configuration of each device included in the information processing system 10.

[0041] Figure 3 shows an example configuration of the bed 100. The bed 100 includes, for example, a processing unit 110, a storage unit 120, a communication unit 130, an operation unit 140, a drive unit 150, and a movable part 160. The bed 100 may also include load sensors and the like, which are not shown in Figure 3.

[0042] The processing unit 110 of this embodiment is composed of the following hardware. The hardware may include at least one of a circuit that processes digital signals and a circuit that processes analog signals. For example, the hardware may consist of one or more circuit devices or one or more circuit elements mounted on a circuit board. One or more circuit devices may be, for example, an IC (Integrated Circuit) or an FPGA (Field-Programmable Gate Array). One or more circuit elements may be, for example, a resistor or a capacitor.

[0043] Furthermore, the processing unit 110 may be implemented by the following processor. The bed 100 in this embodiment includes a memory for storing information and a processor that operates based on the information stored in the memory. The information is, for example, a program and various data. The memory may be a storage unit 120 or another type of memory. The processor includes hardware. Various types of processors can be used, such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a DSP (Digital Signal Processor). The memory may be a semiconductor memory such as SRAM (Static Random Access Memory), DRAM (Dynamic Random Access Memory), or flash memory, or a register, or a magnetic storage device such as a hard disk drive (HDD), or an optical storage device such as an optical disk drive. For example, the memory stores instructions that can be read by the computer, and the functions of the processing unit 110 are realized as processing when the processor executes these instructions. The instructions here may be instructions from an instruction set that constitutes a program, or instructions that instruct the hardware circuit of the processor to operate.

[0044] The storage unit 120 is the work area of ​​the processing unit 110 and stores various information. The storage unit 120 can be implemented using various types of memory, and the memory may be semiconductor memory such as SRAM, DRAM, ROM (Read Only Memory), or flash memory, or it may be a register, a magnetic storage device, or an optical storage device. The storage unit 120 may also store drive programs, etc., for driving the drive unit 150, which will be described later.

[0045] The communication unit 130 is an interface for communication over a network, and when the bed 100 performs wireless communication, it includes, for example, an antenna, an RF (radio frequency) circuit, and a baseband circuit. However, the bed 100 may also perform wired communication, in which case the communication unit 130 may include a communication interface such as a USB connector or an Ethernet connector, and a control circuit for said communication interface. The communication unit 130 may operate according to the control of the processing unit 110, or it may include a communication control processor different from the processing unit 110.

[0046] The operating unit 140 is an operating interface for controlling the position and angle of the movable part 160. For example, the operating unit 140 may be an operating panel connected by wire or wirelessly to a control box where the processing unit 110 is located. The operating panel may include, for example, buttons for setting the driving direction and amount of the movable part 160.

[0047] The drive unit 150 is an actuator or the like that drives the movable part 160 of the bed 100. The bed 100 in this embodiment may be, for example, a nursing care bed in which the angle of the bottom, which is the surface on which the mattress is placed, can be adjusted. For example, the bed 100 has a plurality of bottoms divided into a plurality of members as the movable part 160, and the drive unit 150 adjusts the angle of the bottom by changing the position and orientation of at least a part of the plurality of bottoms based on the control of the processing unit 110. For example, the drive unit 150 may control the back angle (control of the bottom on the head side) or the foot angle (control of the bottom on the foot side).

[0048] The bed 100 may also be a bed with adjustable leg height. In this case, the drive unit 150 is an actuator or the like that drives the leg, which is the movable part 160. In this way, the height of the entire bed 100 can be adjusted by driving the drive unit 150.

[0049] As shown in Figure 3, the bed 100 may be connected to the imaging device 700 via the communication unit 130. The processing unit 110 may perform image processing such as object recognition on the captured image captured by the imaging device 700.

[0050] Figure 4 shows an example configuration of a bedside terminal device 200. The bedside terminal device 200 includes, for example, a processing unit 210, a storage unit 220, a communication unit 230, a display unit 240, an operation unit 250, a notification unit 260, and an interface unit 270. The bedside terminal device 200 may also include a display device and a connection device as described above. For example, the display device includes the display unit 240 in Figure 4, and the connection device includes the processing unit 210, storage unit 220, communication unit 230, operation unit 250, notification unit 260, and interface unit 270 in Figure 4. However, the configuration shown in Figure 4 can be provided in either the display device or the connection device at will, and it is not prohibited for them to be distributed and provided in both the display device and the connection device.

[0051] The processing unit 210 is comprised of hardware including at least one of a circuit for processing digital signals and a circuit for processing analog signals. The processing unit 210 may also be implemented by a processor. Various types of processors can be used, such as a CPU, GPU, or DSP. The functions of the processing unit 210 are realized as processing when the processor executes instructions stored in the memory of the bedside terminal device 200.

[0052] The storage unit 220 is the work area of ​​the processing unit 210 and is implemented by various types of memory such as SRAM, DRAM, and ROM. For example, the storage unit 220 stores biological information acquired from the detection device 810 and the measuring device 820.

[0053] The communication unit 230 is an interface for communication over a network and includes, for example, an antenna, an RF circuit, and a baseband circuit. The communication unit 230 communicates with the server system 300, for example, over a network. The communication unit 230 may perform wireless communication or wired communication, and the specific communication method is not limited.

[0054] The display unit 240 is an interface for displaying various information, and may be a liquid crystal display, an organic EL display, or another type of display. The operation unit 250 is an interface for receiving user input. The operation unit 250 may be a button or the like provided on the display device or connection device. Alternatively, the display unit 240 and the operation unit 250 may be a touch panel configured as an integrated unit.

[0055] The notification unit 260 may include a light-emitting unit, a vibration unit, a sound output unit, etc. The light-emitting unit is, for example, an LED (light-emitting diode) and provides notification by emitting light. The vibration unit is, for example, a motor and provides notification by vibration. The sound output unit is, for example, a speaker and provides notification by sound.

[0056] The interface unit 270 is the interface for connecting the measuring device 820 and reading the authentication card 830. The interface unit 270 is a port that accepts connections of connectors such as USB (Universal Serial Bus). However, the communication unit 230 may also function as the interface unit 270, as is the case when the authentication card 830 is read using NFC.

[0057] Figure 5 is a block diagram showing a detailed configuration example of the server system 300. The server system 300 includes, for example, a processing unit 310, a storage unit 320, and a communication unit 330.

[0058] The processing unit 310 is comprised of hardware including at least one of a circuit for processing digital signals and a circuit for processing analog signals. The processing unit 310 may also be implemented by a processor. Various types of processors can be used, such as a CPU, GPU, or DSP. The functions of the processing unit 310 are realized as processing when the processor executes instructions stored in the memory of the server system 300.

[0059] The memory unit 320 is the work area of ​​the processing unit 310 and stores various information. The memory unit 220 is implemented by various types of memory.

[0060] The communication unit 330 is an interface for communication over a network, and when the server system 300 performs wireless communication, it includes, for example, an antenna, an RF circuit, and a baseband circuit. The specific communication method of the communication unit 330 can be implemented in various variations.

[0061] The electronic medical record server 400, like the server system 300, is a device that includes a processing unit, a storage unit, a communication unit, and the like. A detailed explanation of the configuration of the electronic medical record server 400 will be omitted.

[0062] Figure 6 is a block diagram showing a detailed configuration example of the terminal device 600. The terminal device 600 may include, for example, a processing unit 610, a storage unit 620, a communication unit 630, a display unit 640, and an operation unit 650.

[0063] The processing unit 610 is comprised of hardware including at least one of a circuit for processing digital signals and a circuit for processing analog signals. The processing unit 610 may also be implemented by a processor. Various types of processors can be used, such as a CPU, GPU, or DSP. The functions of the processing unit 610 are realized as processing when the processor executes instructions stored in the memory of the terminal device 600.

[0064] The memory unit 620 is the work area of ​​the processing unit 610 and is implemented by various types of memory such as SRAM, DRAM, and ROM.

[0065] The communication unit 630 is an interface for communication over a network and includes, for example, an antenna, an RF circuit, and a baseband circuit. The communication unit 630 communicates with the server system 300, for example, over a network. For example, the communication unit 630 may communicate with the server system 300 over a LAN and also communicate with the bed 100 and the bedside terminal device 200 using short-range wireless communication such as Bluetooth®. Various variations are possible for the specific communication method of the communication unit 630.

[0066] The display unit 640 is an interface for displaying various information, and may be a liquid crystal display, an organic EL display, or another type of display. The operation unit 650 is an interface for receiving user input. The operation unit 650 may be a button or the like provided on the terminal device 600. Alternatively, the display unit 640 and the operation unit 650 may be a touch panel configured as an integrated unit.

[0067] Furthermore, the terminal device 600 may include configurations not shown in Figure 6. For example, the terminal device 600 may have various sensors such as motion sensors like acceleration sensors and gyro sensors, pressure sensors, and GPS (Global Positioning System) sensors. The terminal device 600 may also include a light-emitting unit, a vibration unit, a sound input unit, a sound output unit, etc. The light-emitting unit is, for example, an LED (light-emitting diode) and provides notification by emitting light. The vibration unit is, for example, a motor and provides notification by vibration. The sound input unit is, for example, a microphone. The sound output unit is, for example, a speaker and provides notification by sound.

[0068] The station terminal device 500, like the terminal device 600, is a device that includes a processing unit, a storage unit, a communication unit, a display unit, an operation unit, etc. The terminal device 600 may be, for example, a smartphone or a tablet terminal, and the station terminal device 500 may be, for example, a PC (Personal Computer). A detailed explanation of the configuration of the station terminal device 500 is omitted.

[0069] The information processing system 10 according to this embodiment may include an acquisition unit, a processing unit, and an output processing unit. The acquisition unit acquires first data that associates a clinical department with a risk set representing the risk of patients related to that clinical department. When the processing unit acquires information indicating that a patient hospitalized in the first clinical department is receiving treatment from the second clinical department, it performs evaluation processing on each of the one or more risks included in the second risk set, which is the risk set associated with the second clinical department in the first data. Note that the status of visits to clinical departments may be determined based on the electronic medical record in the electronic medical record server 400. For example, acquiring information indicating that a patient hospitalized in the first clinical department is receiving treatment from the second clinical department corresponds to acquiring the electronic medical record showing the visit status from the electronic medical record server 400. The output processing unit presents the results of the evaluation processing of the second risk set to the nurses of the first clinical department. For example, if the risk set associated with the first clinical department in the first data is designated as the first risk set, the output processing unit presents the evaluation results of the second risk set to the nurses of the first clinical department with priority over the evaluation results of the first risk set. Prioritizing presentation means, for example, that the evaluation results for the second risk set contain a large amount of information (the evaluation results for each risk are displayed individually), while the evaluation results for the first risk set contain less information (the evaluation results for multiple risks are displayed as an overall score). However, the manner of prioritizing presentation is not limited to this. For example, the evaluation results for the second risk set may be displayed as a pop-up on the screen of each device (bedside terminal device 200, terminal device 600, etc.), while the evaluation results for the first risk set may only be displayed as a list without being displayed as a pop-up.

[0070] The information processing system 10 may be, for example, a server system 300. In this case, the acquisition unit of the information processing system 10 may be a processing unit 310 (storage processing unit) that performs the reading process of the first data stored in the storage unit 320. The processing unit of the information processing system 10 may also be the processing unit 310 of the server system 300. The output processing unit of the information processing system 10 may also be a processing unit 310 (display processing unit) that performs the processing of displaying the evaluation result on the display unit 640 of the terminal device 600, etc. However, as described above, the information processing system 10 of this embodiment is not limited to the server system 300, but may also be a terminal device 600 or a bedside terminal device 200. That is, the acquisition unit, processing unit and output processing unit of the information processing system 10 may be the processing unit 610 of the terminal device 600 or the processing unit 210 of the bedside terminal device 200. The information processing system 10 may also be implemented by distributed processing of multiple devices, in which case the acquisition unit, processing unit and output processing unit of the information processing system 10 may be implemented by a combination of two or more of the processing units 310, 610 and 210. The following describes an example in which the acquisition unit, processing unit, and output processing unit of the information processing system 10 are realized by the processing unit 310 of the server system 300.

[0071] As described above using Figure 1, information that is difficult to grasp, such as risks (sudden changes in patient condition), instructions (patient behavior), and patient wishes, is likely to occur when nurses have to respond to matters related to departments other than their own. In the method of this embodiment, at least the patient's risk is managed in association with the department (first data), and the risk related to a department other than the nurse's department (first department) (second department) is evaluated, and the evaluation results are presented to the nurse. Therefore, since information on risks that are difficult for nurses to grasp can be presented, it becomes possible to appropriately support the nurse's response to such risks. Furthermore, as will be described later, the information processing system 10 of this embodiment may also process instructions (patient behavior) and patient wishes, and present the processing results to the nurse.

[0072] Patent Document 1 discloses a method for automatically evaluating the risk of in-hospital adverse events. However, Patent Document 1 does not take into account the medical department, so there is a possibility that an excessive number of risk notifications will be presented, and nurses may have to specifically organize the content of the risks, which can take time to understand. For example, for a nurse who is well aware of the risk of heart failure, receiving multiple notifications about heart failure risk assessment is not only meaningless but also likely to cause them to overlook other notifications. In contrast, the method of this embodiment presents information from a medical department different from the nurse's department, as described above, so it can present information that is highly relevant to the nurse. As will be described later using Figures 24A-24C, this embodiment does not prevent the presentation of risk information that is easy for nurses to grasp, but risks that are difficult to grasp are displayed preferentially (in a manner with a relatively higher degree of detail), making it easier for nurses to understand the content. Furthermore, the method of this embodiment can also present information on instructions (patient behavior) and patient wishes, which is not possible with conventional methods such as Patent Document 1.

[0073] Furthermore, some or all of the processing performed by the information processing system 10 in this embodiment may be implemented by a program. The processing performed by the information processing system 10 refers, in a narrow sense, to the processing performed by the server system 300, but may also include processing performed by other devices such as the terminal device 600 and the bedside terminal device 200.

[0074] The program according to this embodiment can be stored in a non-temporary information storage medium (information storage device), which is a medium readable by a computer. The information storage medium can be implemented as, for example, an optical disc, a memory card, an HDD, or a semiconductor memory. The semiconductor memory is, for example, a ROM. The processing unit 310, etc., performs various processing according to this embodiment based on the program stored in the information storage medium. That is, the information storage medium stores a program that causes the computer to function as the processing unit 310, etc. A computer is a device that includes an input device, a processing unit, a storage unit, and an output unit. Specifically, the program according to this embodiment is a program that causes the computer to execute each of the steps described later using Figures 10, 15, 18, etc.

[0075] Furthermore, the method of this embodiment can be applied to an information processing method that includes the following steps: The information processing method includes: the information processing system acquiring first data that associates a clinical department with a risk set representing the risk of patients related to that clinical department; when information is acquired indicating that a patient hospitalized in the first clinical department is receiving treatment from the second clinical department, performing an evaluation process for each of the one or more risks included in the second risk set, which is the risk set associated with the second clinical department in the first data; and presenting the results of the evaluation process of the second risk set to the nurse of the first clinical department.

[0076] 3. Processing Flow Next, we will explain the specific processing flow. In this embodiment, we assume that there are three types of information that are difficult for nurses to grasp: risk (sudden change in patient condition), instructions (patient behavior), and patient's wishes. Therefore, we will explain the processing related to these in order below. After that, we will also explain variations of the processing flow.

[0077] 3.1 Risks (sudden deterioration of patient condition) <Example of table data> First, let's explain the data used in risk processing. Figures 7A and 7B show examples of the first data set, which associates a clinical department with the risks expected in that department. In the following, we will explain examples where each data is tabular data such as a relational database, but the data format is not limited to this.

[0078] As shown in Figure 7A, the first data associates information identifying a medical department with a risk set related to that department. The medical departments here include various specialties such as internal medicine, surgery, pediatrics, gynecology, dermatology, and ophthalmology. Furthermore, subspecialized departments such as cardiology, respiratory medicine, and gastroenterology may be included within internal medicine. Similarly, subspecialized departments such as cardiac surgery and neurosurgery may be included within surgery. The medical departments are not limited to those exemplified above. Also, the number and types of medical departments set may differ from hospital to hospital.

[0079] A risk set is a collection of risks that may occur in a corresponding medical department. As shown in Figure 7B, a single risk set can contain multiple risks. For example, in Figure 7A, medical department A is internal medicine, and risk set A is a collection of risks that may occur in internal medicine. As shown in Figure 7B, risk set A includes, for example, diabetes risk, heart failure risk, cardiomyopathy risk, etc. These risks represent risks related to diseases such as diabetes, heart failure, and cardiomyopathy. For example, a high risk of diabetes means that there is a high probability of developing diabetes, or that existing diabetes is likely to become severe.

[0080] For example, the processing unit 310 of the server system 300 may identify diseases with a history of occurrence in a specific medical department based on the electronic medical record stored in the electronic medical record server 400, and perform processing to add the risk related to that disease to the risk set corresponding to that medical department. For example, the processing unit 310 acquires a risk set by associating and acquiring information from the information contained in the electronic medical record shown in Figures 27A and 27B with the history of onset of which diseases in which medical departments. In acquiring the risk set, risk assessment and patient behavior information shown in Figure 27B may also be used. The processing unit 310 stores the risk sets for each of the multiple medical departments, and information identifying the specific risks included in the risk sets (Figures 7A and 7B), as first data in the storage unit 320.

[0081] Alternatively, the processing unit 310 of the server system 300 may perform a process to identify a risk set corresponding to a medical department based on operational input from an expert in that medical department. Here, the expert is, for example, a physician. Alternatively, the processing unit 310 may perform a process to identify a risk set corresponding to a medical department based on data other than the electronic medical record. For example, the processing unit 310 may identify a risk set based on data obtained from a recording server (not shown in Figure 2) that stores data other than the electronic medical record. Here, the recording server may be a server that stores patient data not included in the electronic medical record, such as data obtained outside the hospital. Alternatively, the recording server may store open data that is widely available in the medical and nursing care fields (for example, publicly available academic journals).

[0082] The processing unit 310 may also perform a process to push notification of the identified risk set to the nurse's terminal device 600. For example, an application installed on the terminal device 600 may present the risk set requested by the processing unit 310 to the nurse and receive the nurse's judgment on whether each risk in the risk set is difficult to grasp. The processing unit 310 may store the risk set consisting of risks selected by the nurse as first data in the storage unit 320. Alternatively, as will be described later in Figure 21 (particularly in step S405), the nurse may perform an input operation at the timing of executing specific processing based on the risk set (first data).

[0083] Figure 8 shows an example of second data that associates sensors used in risk assessment processing with the specific processing content (algorithm) of the assessment process. Here, risk refers to individual risks such as diabetes risk, heart failure risk, and cardiomyopathy risk, as shown in Figure 7B. For example, if risk A is diabetes risk, sensor A is information that identifies the sensor used in the diabetes risk assessment process, and model A is information that identifies the algorithm for the diabetes risk assessment process.

[0084] The sensors and algorithms used here may also be data corresponding to the tacit knowledge of experts. For example, highly skilled doctors and nurses can determine the degree of risk based on sensing data representing a patient's biometric information. This biometric information can include various types of data such as heart rate (pulse rate), respiratory rate, blood pressure, and arterial oxygen saturation (SpO2). Experts can also estimate a patient's risk from various other pieces of information such as facial expressions, complexion, activity level, frequency of bowel movements, temperature, and humidity. In other words, experts implicitly know which of these diverse pieces of information are useful for risk assessment and how to use that information to accurately determine the risk.

[0085] Therefore, by identifying the sensors used by experts to acquire information and the processing algorithms for the sensing data output by those sensors, it becomes possible to digitize the tacit knowledge of experts. For example, the algorithm in this embodiment may be a trained model obtained by performing machine learning such as a neural network. For example, the processing unit 310 of the server system 300 creates a trained model by performing machine learning using data that associates sensing data acquired for a certain patient with the expert's judgment result regarding the patient's risk (for example, a score representing the degree of risk) as training data. The identification of sensing data, i.e., the identification of the sensors to be used, may be based on the expert's selection or may be done using machine learning. The processing unit 310 stores the data in which the identified sensors and the information of the trained model are associated with the risk as second data in the storage unit 320.

[0086] The input to the algorithm (trained model) is sensing data. The output of the algorithm, in a narrow sense, is information such as a score representing the degree of risk, but it may also include other information. For example, the algorithm may perform a process to determine the actions (countermeasures, actions) that a nurse should take in response to a risk, based on the sensing data. For example, in the machine learning described above, by including the actions that an expert has actually taken in response to a risk as training data, it is possible to generate a trained model that outputs the actions that a nurse should take in response to that risk.

[0087] Alternatively, the storage unit 320 of the server system 300 may pre-store table data (not shown) that associates recommended actions for nurses with each risk. For example, the storage unit 320 stores table data that associates risks with actions that nurses should take when the evaluation result of the risk (the score output by the algorithm) meets predetermined conditions. Note that multiple different actions may be associated with a single risk depending on the evaluation result (the numerical range of the risk). In this case, it is possible to identify the action that a nurse should take by comparing the table data with the output of the algorithm shown in Figure 8.

[0088] Furthermore, the learning process is not limited to that performed on the server system 300, but may be executed on other learning servers, etc. Also, the algorithm is not limited to a model using a neural network, but may be a model using other methods such as SVM (support-vector machine). In addition, the algorithm may be obtained by methods that do not use machine learning (e.g., regression analysis).

[0089] Figure 9 shows an example of data that associates risks with the on / off status of notifications related to those risks. The table data shown in Figure 9 may be set for each nurse, for example, and Figure 9 illustrates the data for a nurse with Nurse ID 0001. Nurses have risks they are familiar with and risks they are unfamiliar with, depending on their work history and experience in each medical department. For example, among nurses belonging to the ophthalmology department, some may be completely unfamiliar with and unable to easily grasp risks related to medical departments other than ophthalmology, while others may be able to easily grasp risks related to internal medicine because they have experience working in internal medicine in the past. Also, even a nurse belonging to the ophthalmology department with no history of working in internal medicine may find the risk of diabetes easy to grasp if she has extensive experience caring for patients with both diabetes and cataracts.

[0090] Therefore, in this embodiment, individual settings may be made for each risk according to the degree to which the nurse understands it. Specifically, the memory unit 320 stores data for setting risk notifications for each nurse as shown in Figure 9. If the notification setting is off, the processing unit 310 does not have to perform an evaluation process and then send a notification, or it does not have to perform the evaluation process at all. In other words, turning the notification setting on or off may correspond to executing or skipping the evaluation process.

[0091] <Example Sequence> Figure 10 is a sequence diagram illustrating a process that presents information about risks that are difficult for nurses to grasp. For example, Figure 10 shows a process that is triggered when a new patient is admitted to the hospital. It is assumed that the table data described above has been obtained using Figures 7A-9 before the process in Figure 10 begins.

[0092] First, in step S101, the server system 300 receives input information about the patient. The input operation may be performed, for example, on a station terminal device 500, or on a terminal device of the department that handles administrative procedures related to hospitalization. Here, the information about the hospitalized patient includes information about the department in which the patient is hospitalized, and information about other departments in which the patient has a medical history.

[0093] In step S102, the processing unit 310 acquires risk sets other than the department in which the patient is hospitalized, based on the department associated with the hospitalized patient and the first data described above using Figures 7A and 7B. For example, based on the information received in step S101, the processing unit 310 identifies departments other than the department in which the hospitalized patient has a medical history, and identifies the risk sets associated with those departments based on the first data (Figure 7A).

[0094] As in the example above, in a case where a patient who was being treated in internal medicine for diabetes is hospitalized in ophthalmology due to developing cataracts, the processing unit 310 performs a process to obtain a risk set associated with internal medicine from the first data. This allows for the appropriate identification of risks that are difficult for ophthalmology nurses to grasp and that have a high probability of occurring in the patient. As can be seen from this example, the risk set obtained in step S102 may be limited to the risk sets of medical departments other than the nurse's department, and only include medical departments that the hospitalized patient has a history of visiting. In the example above, the processing unit 310 may exclude the risk set of a medical department that is neither ophthalmology nor internal medicine (for example, surgery) from the risk set obtained in step S102. In this way, information on risks with a low probability of occurring can be excluded from processing. In the example above, since the patient has no history of visiting surgery, risks related to surgery are not selected in step S102 because they have a low probability of occurring. However, the processing of this embodiment is not limited to this. For example, in step S102, the processing unit 310 may select both the risk set of medical departments other than the nurse's department where the inpatient has a history of visiting, and the risk set of medical departments where the inpatient has not a history of visiting. More precisely, the processing unit 310 may select the risk set for all medical departments other than the nurse's department. In this way, it becomes possible to select a risk set that broadly covers risks that would be difficult for nurses to grasp if they occur.

[0095] In step S103, the processing unit 310 identifies the sensor and algorithm to be used for the risk evaluation process for each risk in the risk set identified in step S102 by referring to the second data (Figure 8).

[0096] In step S104, the processing unit 310 may perform adjustment processing for risks that are difficult to grasp based on the notification settings for each nurse shown in Figure 9. The processing unit 310 reads the notification settings for nurses in the department where the patient is hospitalized from the storage unit 320 and adjusts the risks that are difficult to grasp based on those notification settings. For example, even if diabetes risk is selected as a risk that is difficult to grasp in step S102, if diabetes risk is turned off in the nurse's notification settings (meaning the nurse can grasp diabetes risk), the processing unit 310 removes diabetes risk from the risks that are difficult to grasp. Specifically, in the processing from step S105 onward, the processing excludes diabetes risk from the processing target. Note that the nurses here may be all nurses in the department where the patient is hospitalized. In this case, if the notification setting for diabetes risk is turned on for at least one nurse, diabetes risk will not be excluded and will be included in the processing target. Alternatively, the nurses here may be limited to some of the nurses in charge of the patient. Furthermore, if the assignments are on a shift system and change depending on the time of day, the processing unit 310 may perform adjustment processing for risks that are difficult to grasp for each time of day.

[0097] In step S105, the processing unit 310 performs the process of activating the sensor determined in step S103. Various devices such as the imaging device 700, detection device 810, and measuring device 820 can be used as the sensor here. This makes it possible to acquire sensing data necessary for assessing risks that are difficult for nurses to grasp.

[0098] In step S106, the sensor acquires sensing data. The sensor transmits the acquired sensing data to the server system 300. The transmission of sensing data may be done directly or via the bedside terminal device 200 or terminal device 600.

[0099] In step S107, the processing unit 310 takes the acquired sensing data as input and performs a risk assessment process according to the algorithm (trained model) identified in step S103. The assessment result here may be information such as a score representing the degree of risk, as described above, or it may be information including recommended actions for nurses.

[0100] In step S108, the processing unit 310 performs a process to display the risk assessment results on the terminal device 600 used by the nurse. For example, the processing unit 310 performs a process to display a screen on the display unit 640 of the terminal device 600 that includes the risk assessment results (presence or absence of risk, or score) and recommended actions for the nurse.

[0101] For example, the process shown in step S108 may be performed via the bedside terminal device 200. Specifically, the processing unit 310 causes the display unit 240 of the bedside terminal device 200 to display the link data for displaying the screen.

[0102] Figure 11 shows an example of a patient screen displayed on the display unit 240 of the bedside terminal device 200. As shown in Figure 11, the patient screen includes the patient's name, pictograms, link data, etc.

[0103] The pictograms here represent information about the patient's condition. For example, a patient with the designation AAAA uses walking as their mode of transportation, so the patient screen includes a pictogram representing walking. Similarly, the patient screen includes pictograms corresponding to the state where the patient is allowed to move freely within the room (free movement within the room), the state where they are allowed to consume beverages, and the state where eating and drinking at night is prohibited.

[0104] Furthermore, patients with a rating of AAAA are associated with factors requiring special attention, such as needing blood glucose monitoring, wearing hearing aids in both ears, going out or staying overnight elsewhere, being prohibited from consuming alcohol, and being prescribed medication to be taken before meals. The patient screen may include text that represents these details, for example. The patient screen may also include information such as the attending physician, assigned nurse, admission date, expected discharge date, and medical department.

[0105] In step S108 described above, the processing unit 310 performs a process to display link data for presenting risk information on the patient screen. The link data here is, for example, a QR code (registered trademark) as shown in the lower right of Figure 11, but other forms of data may be used. Also, as described above using Figure 1, if the interface unit 270 of the bedside terminal device 200 includes an NFC reader, the processing unit 310 may transmit image data or link data to the terminal device 600 via NFC.

[0106] As mentioned above, the patient screen can display various pieces of patient information, but the amount of information is overwhelming. Therefore, using QR codes or NFC to present risk information makes it possible to present risk assessment results to nurses in an easily understandable manner.

[0107] Figure 12 shows an example of a risk assessment results screen displayed on the display unit 640 of a terminal device 600 that reads a QR code. As shown in Figure 12, the risk assessment results screen may include information identifying the target patient, the date and time, specific assessment results, and information on actions recommended for nurses (countermeasures, actions taken).

[0108] In the example shown in Figure 12, blood glucose levels are acquired as sensing data, and the evaluation results screen includes specific numerical values ​​for blood glucose levels. Furthermore, the risk assessment based on blood glucose levels displays that the risk of impaired consciousness and the risk of falls are high. This allows for the presentation of information in a way that makes it easy for, for example, ophthalmic nurses unfamiliar with caring for diabetic patients to understand the risks associated with diabetes.

[0109] Furthermore, as shown in Figure 12, the evaluation results screen includes actions recommended to nurses, such as "administer glucose solution intravenously" and "remove obstacles around the bed and set the bed height to the lowest setting." This makes it possible to encourage appropriate actions regarding risks that are difficult for nurses to identify. As shown in Figure 12, the actions (measures) here may include both highly urgent and less urgent actions (those that should be implemented from a permanent perspective).

[0110] As shown in Figure 12, the evaluation results screen may also include a call button and a share button. The call button is, for example, a button for a nurse to start a call with a doctor. For example, if a doctor's instructions are needed to address a risk, the nurse can use this button to quickly seek the doctor's judgment. The doctor's instructions given verbally during the call may be included in the doctor's instructions, which are an example of information that is difficult for nurses to grasp. For example, the doctor's instructions given during the call may be stored in the electronic medical record as doctor's instruction information (Figure 27B). Registration in the electronic medical record may be performed automatically by the processing unit 310, etc., using speech recognition processing, or it may be performed manually by a doctor or nurse. For example, instructions given verbally by a doctor to a nurse may be information that is difficult for other nurses to grasp, but in this embodiment, it is possible to appropriately manage such instructions. The share button is a button for sharing risk-related information with other nurses. For example, when the share button is pressed, the terminal device 600 may perform a process to send the evaluation results screen to the terminal device 600 used by other nurses. This approach makes it possible to share information about high-risk patients among multiple nurses.

[0111] Furthermore, while the above example shows a nurse actively viewing the evaluation results screen by scanning a QR code, the method is not limited to this. For example, as shown in step S109 of Figure 10, if the evaluation results meet the given conditions, the processing unit 310 may push notification of the evaluation results to the nurse's terminal device 600. The screen displayed on the display unit 640 of the terminal device 600 is similar to, for example, Figure 12. In this way, it becomes possible to appropriately notify the nurse when urgent measures are needed. Note that the recipient of the notification in step S109 may be limited to the assigned nurse, or it may include all nurses in the department where the patient is hospitalized, or it may include doctors.

[0112] Furthermore, while the above example shows how to display the evaluation results screen shown in Figure 12 on the nurse's terminal device 600 using a QR code or NFC, the method is not limited to this. For example, as described above using Figure 2, the bedside terminal device 200 may include an interface unit 270 that performs user authentication using an authentication card 830 or the like. For example, a nurse performs a login operation to the bedside terminal device 200 using an authentication card 830 that stores information indicating that the user attribute is "nurse". When the processing unit 210 of the bedside terminal device 200 accepts the nurse's login operation based on the reading result from the interface unit 270, it may display an evaluation results screen including the content shown in Figure 12 on the display unit 240. For example, when a nurse logs in, the display unit 240 may display a staff screen containing more detailed information for nurses, and display the evaluation results screen as a pop-up screen on that staff screen. In this way, the processing for displaying the evaluation results screen and the device for displaying the evaluation results screen can be implemented in various variations. Also, if the terminal device 600 is capable of NFC communication, the terminal device 600 may also function as the authentication card 830. Furthermore, when a nurse logs into the bedside terminal device 200 using NFC communication, the bedside terminal device 200 may transmit information representing the evaluation results screen to the nurse's terminal device 600. In other words, the evaluation results screen may be displayed on the terminal device 600 triggered by the login operation to the bedside terminal device 200.

[0113] As described above, the acquisition unit of the information processing system 10 according to this embodiment (for example, the processing unit 310 of the server system 300) acquires second data (Figure 8) that associates risks included in the risk set, sensors used for risk evaluation processing, and an algorithm indicating the evaluation processing (evaluation algorithm). For each of the one or more risks included in the second risk set, the processing unit of the information processing system 10 (for example, the processing unit 310) acquires sensor data from the sensors identified by the second data (step S106), and executes processing according to the evaluation algorithm that uses the sensor data as input data as the evaluation processing for that risk (step S107). The evaluation algorithm is also information identified by the second data for each risk.

[0114] According to the method of this embodiment, risks that are difficult for nurses to grasp can be identified based on the first data, and an evaluation process related to those risks can be performed based on the second data. As a result, information on risks that are difficult for nurses to grasp can be presented, making it possible to encourage nurses to take appropriate measures. The process shown in S106-S109 of Figure 10 is repeatedly executed during the period in which the patient is hospitalized in the first clinical department.

[0115] 3.2 Instructions (patient behavior) <Example of table data> This section describes the data used in processing instructions (patient behavior). Figures 13A and 13B are examples of third data that associate a clinical department with the expected physician's instructions within that department, specifically the patient behavior that nurses should monitor. As shown in Figure 27B, the electronic medical record may also include physician's instruction information, and the attending physician's instructions represented by the physician's instruction information correspond to the physician's instructions in this embodiment. Therefore, it is not essential that the third data described below be provided as table data separate from the electronic medical record. In other words, the processing unit 310 may use the electronic medical record as the third data to perform the processing described later using Figure 15, in which case the table data described below may be omitted.

[0116] As shown in Figure 13A, the third data set associates information identifying the clinical department with a set of instructions related to that department. The clinical departments are identified in the same way as in the example described above using Figure 7A.

[0117] An instruction set is a collection of instructions that are highly likely to be given to nurses in a corresponding medical department. An instruction set can also be described as a collection of patient behaviors that nurses in a corresponding medical department need to monitor. As shown in Figure 13B, one instruction set contains multiple instructions. For example, in Figure 13A, medical department A is internal medicine, and instruction set A is a collection of patient behaviors that internal medicine nurses need to monitor. As shown in Figure 13B, instruction set A includes, for example, dietary restrictions, fluid restrictions, and blood glucose management.

[0118] For example, the processing unit 310 of the server system 300 may identify instructions that a doctor has given to a nurse in a particular medical department based on the electronic medical records stored in the electronic medical record server 400, and add those instructions to the instruction set corresponding to that medical department. The processing unit 310 stores the instruction sets for each of the multiple medical departments, and information identifying the specific instructions included in those instruction sets (Figures 13A and 13B), as third data in the storage unit 320. Also, similar to the example of the first data, the processing unit 310 of the server system 300 may identify the instruction set corresponding to a medical department based on expert input related to that medical department.

[0119] Figure 14 shows an example of a fourth set of data that associates sensors used in the patient behavior determination process with the specific processing content (algorithm) of the determination process. Here, instructions (patient behavior) refer to individual instructions such as dietary restrictions and water restrictions shown in Figure 13B. For example, if instruction a is dietary restrictions, sensor a is information that identifies the sensor used in the evaluation process of dietary restrictions, and model a is information that identifies the algorithm for the evaluation process of dietary restrictions.

[0120] For example, a method for determining the amount of food consumed and the amount of nutrients based on changes in images taken of the area where tableware is placed before and after a meal is described in U.S. Patent Application No. 18 / 120116, filed on March 10, 2023, titled "INFORMATION PROCESSING SYSTEM AND AN INFORMATION PROCESSING METHOD." This patent application is incorporated by reference in its entirety in this specification. It is also possible to determine the amount of water consumed by taking time-series images of beverage containers such as PET bottles. In this case, the imaging device 700 may be used as a sensor related to dietary and water restrictions. Alternatively, the amount of food and water intake may be determined by placing scales (weighing scales) in the area where food is served and the area where beverage containers are placed. In this case, the scales may be used as sensors related to dietary and water restrictions. Furthermore, the amount of food and water intake may be determined based on changes in weight using a weighing scale. The weighing scale may be a dedicated device or a load sensor provided on the bed 100.

[0121] Alternatively, the sensor mapping in the fourth data set may be configured such that a relatively small number of the above types of sensors are used for weight measurements performed on all patients, while a relatively large number of the above types of sensors are used when measuring patients who are on dietary or fluid restrictions. This would allow for higher accuracy in determining food and fluid intake when processing instructions.

[0122] Furthermore, detecting specific intake amounts is not essential when determining patient behavior related to dietary and fluid restrictions. For example, if an object estimated to be food (dish) is detected by object detection processing on the captured image at a time other than a regular mealtime, the algorithm may omit a specific determination of whether or not the food was consumed and instead estimate that the dietary restriction has been violated. Similarly, in the case of fluid restrictions, instead of specifically detecting changes in the amount of liquid in the beverage container, processing to detect human movement may be performed. For example, the algorithm may detect throat movement based on the captured image and estimate that the fluid restriction has been violated if swallowing is detected a predetermined number of times. Alternatively, the algorithm may estimate that the fluid restriction has been violated if a beverage container is detected from the captured image and swallowing is detected a predetermined number of times. Assuming that the number of swallows is proportional to the amount of fluid intake, the threshold for the number of swallows required to determine a fluid restriction violation can be determined according to the allowable amount of fluid intake. Note that the sensor for detecting swallowing is not limited to the imaging device 700, but may be other sensors such as a throat microphone attached to the throat.

[0123] Furthermore, the algorithm is the same as that for the second data, and may, for example, be data corresponding to the tacit knowledge of an expert. For example, the processing unit 310 or an external learning device may generate a trained model by performing machine learning using the expert's judgment results as training data.

[0124] In a narrow sense, the output of the algorithm (trained model) is the result of determining whether or not the nurse is following the doctor's instructions (whether or not the patient's behavior is in accordance with the doctor's instructions). Alternatively, the output of the algorithm may be information such as a score representing the degree to which the nurse is following the doctor's instructions.

[0125] <Example Sequence> Figure 15 is a sequence diagram illustrating a process that presents information related to instructions that are difficult for nurses to grasp. For example, Figure 15 is a process that is triggered when a new patient is admitted to the hospital, similar to the process shown in Figure 10. It is assumed that the table data described above has been obtained using Figures 13A-14 before the process in Figure 15 begins.

[0126] First, in step S201, the server system 300 receives input information about the patient. Also in step S201, the processing unit 310 receives information including instructions given by a doctor regarding the hospitalized patient.

[0127] In step S202, the processing unit 310 obtains a set of instructions for departments other than the department in which the patient is hospitalized, based on the departments associated with the hospitalized patient and the third data described above using Figures 13A and 13B. For example, based on the information received in step S201, the processing unit 310 identifies departments other than the department in which the hospitalized patient has a medical history, and identifies the set of instructions associated with those departments based on the third data (Figure 13A).

[0128] Furthermore, the processing unit 310 may identify instructions from the physician obtained in step S201 that are included in the set of instructions identified based on the third data. Since the set of instructions identified based on the third data is a collection of instructions that are difficult for nurses to grasp, if instructions actually given by the physician are included in this set of instructions, those instructions will be difficult for nurses to grasp.

[0129] In step S203, the processing unit 310 identifies the sensor and algorithm to be used for determining the instruction for each of the one or more instructions identified in step S202 (i.e., instructions that are difficult for the nurse to grasp) by referring to the fourth data (Figure 14).

[0130] In step S204, the processing unit 310 causes the contents of one or more identified instructions to be displayed on the terminal device 600.

[0131] Figure 16 shows an example of the instruction confirmation screen displayed on the display unit 640 of the terminal device 600 in step S204. The instruction confirmation screen may be displayed by reading the link data displayed on the display unit 240 of the bedside terminal device 200, as described above using Figure 11, or by receiving a push notification from the server system 300. Also, as described above in the explanation of the evaluation result screen shown in Figure 12, the display unit 240 of the bedside terminal device 200, or the display unit 640 of the terminal device 600, may display the instruction confirmation screen shown in Figure 16, triggered by the nurse performing a login operation to the bedside terminal device 200 using an authentication card 830 or the like.

[0132] As shown in Figure 16, the instruction confirmation screen contains information about the specific instructions from the physician. In the example in Figure 16, the instruction confirmation screen includes text representing instructions such as "Measure the patient's blood glucose level four times daily," "Adjust the insulin treatment dosage," and "Collaborate with a dietitian to create a meal plan and limit calorie intake." Similar to the evaluation results screen shown in Figure 12, call and share buttons may also be displayed. These buttons allow nurses to confirm the instructions with the physician and share the instructions with other nurses.

[0133] In step S205, the processing unit 310 performs the process of activating the sensor determined in step S203. This makes it possible to acquire sensing data necessary for processing instructions that are difficult for nurses to grasp.

[0134] In step S206, the sensor acquires sensing data. The sensor transmits the acquired sensing data to the server system 300.

[0135] In step S207, the processing unit 310 takes the acquired sensing data as input and performs a judgment process on the instruction by processing it according to the algorithm (trained model) identified in step S203. The result of this judgment process may be information on whether or not the instruction has been followed, as described above, or it may be information such as a score.

[0136] In step S208, the processing unit 310 executes a process to display the instruction judgment result on the terminal device 600 used by the nurse. This process may be executed when the judgment result satisfies a given condition (for example, when it is determined that the instruction was not followed). For example, the processing unit 310 pushes information indicating the judgment result to the terminal device 600. However, the judgment result screen may also be displayed by reading a QR code or the like displayed on the display unit 240 of the bedside terminal device 200.

[0137] Figure 17 shows an example of a judgment result screen for instructions displayed on the display unit 640 of the terminal device 600. Here, we consider an example where dietary restrictions and fluid restrictions are set as instructions that are difficult for nurses to grasp. Therefore, the judgment result screen includes information that the patient is acting in violation of dietary and fluid restrictions. Specifically, the judgment result screen may include text indicating that the patient may have eaten or drunk outside of the prescribed limits, a thumbnail of a video image (for example, an image captured by the imaging device 700) showing the eating or drinking outside of the prescribed limits, and a play button to play the video. For example, the processing unit 310 processes to display a video image of the most recent meal if the intake amount or weight change exceeds a prescribed value, or if it detects eating or drinking outside of the prescribed meal times. Whether or not the video image represents a meal may be determined using the results of object detection such as a plastic bottle or tableware, or other processing may be used. The fact that a call button and a share button may be displayed on the judgment result screen is the same as the evaluation result screen shown in Figure 12 and the instruction confirmation screen shown in Figure 16. Furthermore, similar to the risk-related processing, the target audience for displaying the instruction confirmation screen (Figure 16) and the judgment result screen (Figure 17) may be limited to the assigned nurse, include all nurses in the hospitalized department, or include doctors.

[0138] The processes shown in S206-S208 of Figure 15 are executed repeatedly during the period the patient is hospitalized in the first clinical department. Furthermore, the doctor's instructions may change during hospitalization. Therefore, the processes shown in steps S201-S205 may be executed again, triggered by a change in instructions.

[0139] As described above, the acquisition unit of the information processing system 10 acquires third data (Figures 13A and 13B) that associates a clinical department with a set of instructions representing instructions given by a doctor in that clinical department to a nurse. The acquisition unit also acquires fourth data (Figure 14) that associates an instruction with a sensor used for determining whether the instruction is being followed, and an algorithm (determination algorithm) that indicates the determination process based on the sensor's output. When a patient hospitalized in the first clinical department is receiving treatment from the second clinical department, the processing unit of the information processing system 10 acquires sensor data from the sensor identified by the fourth data for each of the one or more instructions included in the second instruction set, which is the set of instructions associated with the second clinical department in the third data (step S206), and executes a process according to the determination algorithm that uses the sensor data as input data as a determination process to determine whether the instruction is being followed (step S207). The determination algorithm is the information identified by the fourth data for each instruction. The output processing unit of the information processing system 10 then presents the results of the judgment process regarding the second instruction set to the nurse of the first clinical department (step S208). For example, if the instruction set associated with the first clinical department is designated as the first instruction set, the output processing unit prioritizes displaying the judgment results regarding the second instruction set over the judgment results regarding the first instruction set.

[0140] According to the method of this embodiment, instructions that are difficult for nurses to grasp can be identified based on the third data, and a judgment process regarding those instructions can be performed based on the fourth data. As a result, information regarding instructions that are difficult for nurses to grasp can be presented, making it possible to encourage nurses to take appropriate actions to comply with those instructions.

[0141] 3.3 Patient's wishes (nurse call) The processing unit of the information processing system 10 in this embodiment performs processing to estimate the cause of the call or the recommended response to the call for the nurse, based on the patient's electronic medical record information, the patient's speech content, and the nurse's speech content, when the patient makes a call to summon a nurse (nurse call). Specifically, the processing unit of the information processing system 10 may estimate the cause or response by inputting information (prompts described later) that specifies the cause or response content as processing conditions (input) and the cause or response content as output into a trained model. The trained model here may be, for example, a model corresponding to generative AI.

[0142] Patient intent (what the patient was thinking when they made a nurse call) can be difficult information for nurses to grasp. For example, even if a patient makes a nurse call due to pain or discomfort, various factors such as specific diseases, medication side effects, or mental stress may be the cause of the pain or discomfort, making it difficult for nurses to make an appropriate judgment. Also, depending on the patient's condition, such as in the case of dementia, it may be difficult for the patient to adequately verbalize their subjective symptoms. In this respect, the method of this embodiment allows for the estimation of the cause of a nurse call from the electronic medical record and the content of conversations between the patient and the nurse, thereby appropriately presenting information regarding the patient's intent to the nurse. As a result, it becomes possible to encourage nurses to respond appropriately to nurse calls. Furthermore, as described above, the method of this embodiment can also support nurses' responses by estimating the recommended response to a call.

[0143] Furthermore, if a patient admitted to the first clinical department is receiving treatment from the second clinical department, and the patient makes a call to summon a nurse from the first clinical department, the processing unit of the information processing system 10 may acquire information including the second risk set (risks that are difficult for nurses to grasp) as electronic medical record information, and estimate the cause or response based on the acquired electronic medical record information. In other words, information including the second risk set may be extracted from the electronic medical record server 400 as electronic medical record information used in the process of estimating the patient's intentions.

[0144] As mentioned above, the second risk set is a collection of risks that can be expected in medical departments other than the nurse's department. When such risks are the cause of a call, it is difficult for nurses to grasp these risks, making it difficult to identify the cause of the nurse call or to determine the appropriate response. For example, if a patient with asthma is admitted to the ophthalmology department, a nurse belonging to the ophthalmology department may not immediately connect a nurse call complaining of coughing with asthma. In this respect, by using electronic medical record information including the second risk set to estimate the cause or response, it becomes possible to present appropriate information that takes into account the nurse's department. In the above example, it becomes possible to draw attention to the link between coughing and asthma from the conversation during the nurse call. The specific processing flow will be explained below.

[0145] Figure 18 is a sequence diagram illustrating the process when a nurse call is made. The call system shown in Figure 18 is, for example, a device installed near the bed 100, which includes a microphone to acquire the patient's voice, a speaker to output the nurse's voice, and an operating unit to instruct the start of the nurse call. The call system may also be the bedside terminal device 200, or a different device from the bedside terminal device 200 may be used. For example, a terminal device such as a smartphone used by the patient may be used as the call system. Also, although Figure 18 shows an example where the nurse's response to the nurse call is performed by the terminal device 600, some or all of the response may be performed by the station terminal device 500.

[0146] First, in step S301, the call system accepts a nurse call operation from the patient. The process in step S301 may involve accepting a press of a dedicated button, or it may involve accepting a touch operation of the nurse call button displayed on the display unit 240 of the bedside terminal device 200. Alternatively, as in Figure 11, a QR code may be displayed on the display unit 240 of the bedside terminal device 200, and when the patient's terminal device reads the QR code, a nurse call button may be displayed on the terminal device's display unit.

[0147] In steps S302 and S303, the call system notifies the server system 300 and the nurse's terminal device 600 that a nurse call has been made. Alternatively, the nurse call notification may first be sent to the server system 300, which then forwards the notification to the designated nurse's terminal device 600. The device that receives and forwards the nurse call may also be a station terminal device 500.

[0148] The process in step S303 enables the patient and nurse to initiate a conversation using the call system and the terminal device 600. Specifically, in step S305, the call system transmits the patient's voice to the terminal device 600, and the speaker of the terminal device 600 outputs the patient's voice. In step S307, the terminal device 600 transmits the nurse's voice to the call system, and the speaker of the call system outputs the nurse's voice. The processes in steps S305 and S307 are repeatedly executed, allowing the patient and nurse to communicate.

[0149] In this process, the call system transmits the patient's voice to the server system 300 (step S304). The terminal device 600 also transmits the nurse's voice to the server system 300 (step S306). This allows the server system 300 to acquire the content of the patient's and nurse's speech. Alternatively, either the call system or the terminal device 600 may collect both the patient's and nurse's voices, and that device may transmit both to the server system 300.

[0150] In step S308, the server system 300 identifies the patient who made the nurse call based on the notification content in step S302. For example, the call system is pre-associated with patients, and the server system 300 identifies the patient based on which call system made the notification.

[0151] In step S309, the server system 300 obtains information contained in the electronic medical record of the identified patient (hereinafter also referred to as electronic medical record information) from the electronic medical record server 400. As described above, the electronic medical record information may also include information from the second risk set. The process for obtaining the second risk set is the same as in the example described above using Figure 10.

[0152] In step S310, the server system 300 may acquire sensing data from sensors (imaging device 700, detection device 810, measuring device 820, etc.) located near the patient. For example, sensing data may be used as input for estimating the cause of a call and the content of the response, in which case the processing shown in step S310 is executed.

[0153] In step S311, the processing unit 310 of the server system 300 performs a process to estimate the cause of the nurse call and / or the response based on the electronic medical record information, patient voice, and nurse voice. As described above, in this process, the processing unit 310 may use the second risk set, the sensing data acquired in step S310, or other data.

[0154] Specifically, the processing unit 310 may estimate the factors and corresponding actions using artificial intelligence (AI). In this case, the processing in step S311 may be the process of creating a prompt that represents an instruction for the AI.

[0155] Figure 19 is an explanatory diagram of a prompt. For example, a prompt includes the task to be performed by the generating AI, the conditions for performing that task, and the output content. In this example, the task column sets up a scenario where a patient makes a nurse call, and then sets up a task to output the patient's intention for making the nurse call and effective countermeasures (responses). In the prompt shown in Figure 19, the task instructs that patient speech (patient voice), nurse speech (nurse voice), electronic medical record (electronic medical record information), and sensor output (sensing data) be used for processing. The output content column indicates that the data that the generating AI should output is the patient's intention (factor) and countermeasures (responses).

[0156] In the conditions column, patient utterances, nurse utterances, electronic medical records, and sensor outputs are specifically set. For example, the processing unit 310 adds the data obtained in step S304, which has been converted into text by speech recognition processing, to the conditions column as "patient utterances". The processing unit 310 adds the data obtained in step S306, which has been converted into text by speech recognition processing, to the conditions column as "nurse utterances". The processing unit 310 adds the electronic medical record information, including the second risk set, obtained in step S309, to the conditions column as "electronic medical records". The processing unit 310 adds the sensing data obtained in S310 to the conditions column as "sensor output". In this way, the processing unit 310 can automatically create prompts for the generation AI based on the data obtained regarding nurse calls. Note that the input to the generation AI may be the audio file itself, in which case the text conversion by speech recognition processing for at least one of the patient utterances and nurse utterances may be omitted.

[0157] In step S312, the processing unit 310 estimates the factors and corresponding actions by inputting the generated prompt into the generating AI.

[0158] In step S313, the server system 300 transmits the estimation result obtained in step S312 to the terminal device 600. In step S314, the display unit 640 of the terminal device 600 displays the received estimation result.

[0159] For example, the server system 300 and terminal device 600 execute the processes in steps S308-S314 in real time while the patient and nurse are conversing according to the processes shown in steps S305 and S307. In this way, the nurse can view the estimated results of the cause and response while conversing with the patient, enabling them to take an appropriate response to the nurse call.

[0160] Figures 20A and 20C show examples of screens displayed on the display unit 640 of the terminal device 600. Figure 20A shows an example of a screen displayed on the display unit 640 in step S303, for example. When a nurse call is notified, the display unit 640 of the terminal device 600 may display information identifying the patient who made the nurse call, as well as some basic information about the patient, as shown in Figure 20A. Figure 20A shows an example where the display unit 640 displays the patient's age, medical history, and sensing data. The display unit 640 may also prompt the nurse to decide whether or not to accept the nurse call by displaying two buttons, "Reject" and "Accept". If the "Accept" button is selected, a call between the patient and the nurse is initiated using the call system and the terminal device 600.

[0161] Figure 20B is an example of a screen displayed on the display unit 640 while a patient and a nurse are on a call, before the estimation processing by the generating AI. As shown in Figure 20B, the display unit 640 may display data that has been transcribed from the patient's voice and data that has been transcribed from the nurse's voice. The operation unit 650 of the terminal device 600 may also accept correction operations from the nurse on the text data. In the example in Figure 20B, a button labeled "Confirm Action" is displayed, and when this button is selected, the estimation result of the generating AI is displayed. For example, the processing unit 310 of the server system 300 may perform the processing from step S308 onwards when the terminal device 600 selects the "Confirm Action" button. Alternatively, the processing unit 310 may perform the processing from step S308 onwards in the background and send the estimation result to the terminal device 600 when the "Confirm Action" button is selected. Alternatively, the "Confirm Action" button may be omitted, and the processing from step S308 onwards may be performed automatically, and the estimation result may be displayed on the terminal device 600 when it is obtained.

[0162] Figure 20C shows an example of a screen displaying the estimation results by the generating AI. For example, when the generating AI displays actions, the display unit 640 may display one or more corresponding actions. In the example of Figure 20C, the display unit 640 presents two actions: checking the level of consciousness and administering antihypertensive drugs according to the doctor's instructions. In addition, the estimation process of the processing unit 310 may determine an index representing the likelihood of the estimation result (for example, similarity to past cases), and the actions may be displayed in descending order of the index value. In the example of Figure 20C, the display unit 640 displays action 1 with a similarity of 98 and action 2 with a similarity of 95 in that order.

[0163] Similar to the evaluation results screen shown in Figure 12, a call button and a share button may also be displayed. The display unit 640 may also switch the presence or absence of the call button depending on the content of the action, such as not displaying the call button when the action can be performed independently by the nurse (for example, checking the level of consciousness) and displaying the call button when the action requires a doctor's instruction (for example, administering antihypertensive drugs). For example, the storage unit 320 may store a table associating actions with whether or not a doctor's instruction is required, and the processing unit 310 may transmit information regarding the presence or absence of the call button associating it with the action when transmitting the estimation result to the terminal device 600. Alternatively, the processing unit 310 may obtain information representing the doctor's instruction (doctor's instruction information in Figure 27B) from the electronic medical record and determine whether or not the estimated action matches the doctor's instruction. The processing unit 310 may then perform a process to associate the call button with the action and display it if the action and the doctor's instruction do not match (if an action different from the doctor's instruction is recommended). In this way, nurses can seek the doctor's judgment on whether or not to implement measures that do not align with the doctor's instructions. Furthermore, any new instructions from the doctor given during the call may be registered in the electronic medical record as doctor's instruction information, as described above (Figure 27B).

[0164] 3.4 Variations We will now describe some variations of the process described above.

[0165] <Adjustments for each nurse> Similar to the example of adjusting for risks that are difficult to grasp for each nurse (step S104 in Figure 10), the processing unit 310 may adjust instructions (patient behaviors) that are difficult for nurses to grasp for each nurse. For example, the memory unit 320 may store notification settings for each nurse, associating instructions with the on / off status of notifications related to those instructions, as in Figure 9. The processing unit 310 then performs a process to adjust the instructions that are difficult to grasp based on these notification settings, for example, after step S203 in Figure 15. The processing unit 310 may then perform the processes from step S204 onward for instructions that were not excluded in the adjustment process.

[0166] <Registration timing> In the sequence diagrams shown in Figures 10 and 15, the process was triggered by the patient's hospitalization (steps S101 and S201). However, the method of this embodiment is not limited to this, and the process may be triggered by other triggers. Other triggers may include, for example, the registration of a nurse.

[0167] Figure 21 is another sequence diagram illustrating the risk assessment process. First, in step S401, the terminal device 600 accepts the registration of a nurse. In step S401, information such as the nurse ID to identify the nurse, the nurse's name, and the department to which the nurse belongs is entered. Note that the process in step S401 may be performed using the station terminal device 500 or using the terminal device of the administrative staff.

[0168] In step S402, the terminal device 600 registers the registration details in the server system.

[0169] In step S403, the processing unit 310 of the server system 300 acquires the risk set of medical departments other than the registered nurse's department, based on the first data. In step S404, the server system 300 transmits the acquired risk set to the terminal device 600.

[0170] In step S405, the display unit 640 of the terminal device 600 displays the acquired risk set and accepts approval input for each of the multiple risks included in the risk set, indicating whether or not to issue a notification.

[0171] Figure 22 shows an example of the screen displayed by the display unit 640 in step S405. The display unit 640 may display a list of risks included in the risk set, as well as an operation interface (e.g., a slide button) for inputting whether to turn notifications on or off for each risk. Based on this display, the nurse performs an operation to decide whether to turn notifications on or off for each risk.

[0172] In step S406, the terminal device 600 transmits the input content to the server system 300. The information transmitted here is, for example, information relating a risk to whether or not to turn on notifications for that risk.

[0173] In step S407, the processing unit 310 of the server system 300 determines the risk set corresponding to the target nurse based on the input content. Specifically, the processing unit 310 updates the information representing the target nurse's notification settings (specifically, the table described above using Figure 9) based on the information obtained in step S406. Then, if the patient under the target nurse's care has a history of visiting a medical department other than their own, the processing unit 310 identifies risks that are included in the risk set of that medical department and for which the nurse has turned on notifications as risks that are difficult for the nurse to grasp.

[0174] In step S408, the processing unit 310 determines the sensors and algorithms (trained models) necessary for evaluation processing for each risk that is difficult for nurses to grasp, based on the second data.

[0175] The processing unit 310 then performs risk assessment and display processing during the period in which the target nurse is on duty. Specifically, the processing unit 310 activates the necessary sensors (step S409), the sensors transmit sensing data to the server system 300 (step S410), the processing unit 310 performs evaluation processing from the sensing data and the trained model (step S411), and the processing unit 310 displays the results of the evaluation processing on the display unit 640 of the terminal device 600 (step S412). Furthermore, if the results of the evaluation processing meet predetermined conditions, the processing unit 310 may send a push notification to the terminal device 600 (step S413). The processing in steps S409-S413 is the same as steps S105-S109 in Figure 10, so a detailed explanation is omitted.

[0176] In the method of this embodiment, various methods are possible for combining the risk evaluation process triggered by patient registration as described above using Figure 10 and the risk evaluation process triggered by nurse registration as shown in Figure 21. For example, in the method of this embodiment, these two processes may be used in combination, with the process shown in Figure 10 being executed when a new patient is registered, and the process shown in Figure 21 being executed when a new nurse is registered. In this way, when a new patient or nurse is added, it becomes possible to perform a risk evaluation tailored to that patient and nurse and to present the results appropriately. For example, in the process shown in Figure 10, the processing unit 310 selects the risk, determines the sensor and algorithm tailored to the new patient, and in the process shown in Figure 21, it performs customization of the risk, etc., for each nurse. For example, if a new nurse is hired when inpatients are already registered, it becomes possible to apply the customization for each nurse to the base data (general-purpose data used for the entire medical department) created by the process shown in Figure 10 using the process shown in Figure 21. Furthermore, if nurses are already registered, data regarding notification settings for each nurse (see Figure 9) is created through the process shown in steps S405-S407 of Figure 21. Therefore, when registering a new patient, the customization shown in step S104 of Figure 10 can be automatically executed based on the notification setting data created for each existing nurse.

[0177] Furthermore, the steps for conducting specific risk assessments (steps S105-S109 in Figure 10 and steps S409-S413 in Figure 21) can be standardized as described above. For example, when a new patient is registered, the processing unit 310 executes the process shown in steps S101-S104 in Figure 10 as the patient registration phase. Also, when a new nurse is registered, the processing unit 310 executes the process shown in steps S401-S408 in Figure 21 as the nurse registration phase. The processing unit 310 may also periodically execute the process shown in steps S105-S109 in Figure 10, or the process shown in steps S409-S413 in Figure 21, as the actual risk assessment processing phase. Periodically executing here means, for example, executing the process at regular intervals during the period in which the target patient is hospitalized, but the timing of the execution of the process may be adjusted depending on the type of risk.

[0178] Furthermore, when the process shown in Figure 21 (especially steps S405-S407) or the process shown in Figure 10, step S104 is executed, it is possible that even with the same risk, the notification on / off status may differ depending on the nurse. For example, as mentioned above, for ophthalmic nurses, diabetes risk is generally difficult to grasp, so it is standard practice to have the notification on, but some nurses may have diabetes risk notifications turned off. In this case, the processing unit 310 may switch whether or not to display the results of the risk assessment process depending on the nurse's work status. For example, if a nurse who has diabetes risk notifications turned on is on duty, the processing unit 310 will display the diabetes risk assessment results on the nurse's terminal device 600. On the other hand, if a nurse who has diabetes risk notifications turned off is on duty, the processing unit 310 will not display the diabetes risk assessment results on the nurse's terminal device 600. If the risk assessment results are not displayed, the processing unit 310 may omit the risk assessment process itself, or it may perform the risk assessment process in the background while omitting the output of the assessment results. The condition "The nurse has started work" in steps S409-S413 of Figure 21 takes the above into consideration, indicating that the processing reflects the notification settings of the nurse on duty, based on the nurse's work status.

[0179] Similarly, the instruction (patient behavior) determination process shown in Figure 15 may be triggered by nurse registration. Figure 23 is a sequence diagram illustrating the patient behavior determination process.

[0180] First, in step S501, the terminal device 600 accepts the registration of a nurse. In step S502, the terminal device 600 registers the registration details in the server system.

[0181] In step S503, the processing unit 310 of the server system 300 acquires instruction sets for medical departments other than the registered nurse's department, based on the third data. In step S504, the server system 300 transmits the acquired instruction sets to the terminal device 600.

[0182] In step S505, the display unit 640 of the terminal device 600 displays the acquired instruction set on the display unit and accepts approval input for each of the multiple instructions included in the instruction set, indicating whether or not to issue a notification. The specific processing is the same as in step S405 in Figure 21.

[0183] In step S506, the terminal device 600 transmits the input content to the server system 300. The information transmitted here is, for example, information relating an instruction to whether the notification for that instruction is turned on or off.

[0184] In step S507, the processing unit 310 of the server system 300 determines the instruction set corresponding to the target nurse based on the input content. Specifically, the processing unit 310 updates the information representing the notification settings for the target nurse's instructions based on the information obtained in step S506. The processing unit 310 then identifies instructions that are difficult for the nurse to grasp if the patient under the target nurse's care has a history of visiting a medical department other than their own, are included in the instruction set of that medical department, have actually been instructed by a physician, and the nurse has turned on the notification setting for those instructions.

[0185] In step S508, the processing unit 310 determines the sensors and algorithms (trained models) necessary for judgment processing for each instruction that is difficult for the nurse to grasp, based on the fourth data.

[0186] The processing unit 310 then performs instruction determination processing and display processing during the period in which the target nurse is on duty (steps S509-S512). The processing content of steps S509-S512 is the same as steps S205-S208 in Figure 15, so a detailed explanation is omitted.

[0187] Regarding the method of combining the instruction determination process triggered by patient registration as described above (using Figure 15) with the instruction determination process triggered by nurse registration as shown in Figure 23, a detailed explanation is omitted as it is the same as the risk assessment process combination method described above.

[0188] <The entity responsible for processing> In the above, the server system 300 is assumed to perform the risk assessment process and the patient behavior determination process, but this is not limited to this. For example, algorithms for performing the assessment process and determination process may be provided as application software. For example, the terminal device 600 may download and install application software corresponding to the algorithm shown in Figure 8 or Figure 13 from the server system 300. In this case, the process in step S107 of Figure 10 or the process in step S207 of Figure 15 may be executed by the terminal device 600.

[0189] As mentioned above, the button for activating the nurse call may be displayed on the display unit 240 of the bedside terminal device 200. In this case, the patient's requests may be classified in advance, and the display unit 240 may display multiple different buttons according to the classification and the request. For example, requests may be classified as "support for daily living," "medical requests," "emergencies," etc. Support for daily living includes requests for water and food, assistance with toileting, assistance with changing positions, operation of the television and air conditioner, answering visitors, handing over remote controls and telephones, and adjusting the bed position, and corresponds to tasks that can be performed by caregivers. Medical requests include complaints of pain or discomfort, requests regarding medication, and feeling unwell, and correspond to tasks that nurses perform. Emergencies include falls, injuries, and acute health problems, and correspond to tasks that require immediate attention.

[0190] In this case, the nurse call notification is sent to the server system 300, and the processing unit 310 performs a process to determine the recipient based on the content of the request. In the example above, the processing unit 310 notifies the caregiver's terminal device of calls where "support for daily living" is selected, and notifies the nurse's terminal device 600 of "medical requests" and "emergencies." The processing unit 310 may also select the terminal device of the administrative staff who purchase and manage supplies as the recipient. The consultation content (button) and the notification recipient may be uniquely associated, or the processing unit 310 may perform a process to estimate the notification recipient each time using a trained model or the like.

[0191] The process described above, using Figure 18, is executed, for example, when a nurse call notification is sent to the nurse's terminal device 600, and not when it is sent to any other recipient. This makes it possible to appropriately perform the process of estimating the patient's wishes in situations where it is most necessary.

[0192] Furthermore, the above explanation assumes that nurse calls made by patients are notified to nurses, and that nurses respond to such nurse calls, but this is not the only assumption. For example, if a patient requests water again via nurse call despite having drunk water five minutes earlier, the processing unit 310 may determine that the priority of the response is low based on the time of the previous response. In this case, the processing unit 310 may notify the nurse's terminal device 600 that the priority is low, or it may omit the notification altogether. Nurses may also be able to pre-set whether or not to send notifications according to priority. Whether or not water has been drunk may be detected using captured images, scales, etc., as described above regarding patient behavior.

[0193] Furthermore, the server system 300 of this embodiment may output data for administrators who manage nurses. For example, when a nurse call is notified to a nurse, the processing unit 310 may determine whether the nurse responded to the nurse call and obtain the total number of notifications, the number of responses, the number of non-responses, etc., for each nurse. The processing unit 310 transmits information representing the response status of each nurse to the administrator's terminal device. In this way, it becomes possible for administrators to be aware of the nurses' work status. For example, if an administrator finds that a nurse is excessively responding to low-priority nurse calls, they may instruct the nurse to make choices considering priority. Alternatively, the administrator may instruct nurses who have a high rate of not responding to high-priority nurse calls to be more proactive. Alternatively, if the processing unit 310 determines that a nurse is excessively responding to low-priority nurse calls, it may execute an automatic nurse call assignment process so that the nurse is not notified of low-priority nurse calls. Note that the automatic nurse call assignment process is not limited to being executed by the server system 300, but may be executed by other devices such as the station terminal device 500.

[0194] <Combinations> Furthermore, while the risk assessment process, patient behavior determination process, and call factor estimation process have been described above, only one of these processes may be executed, or two or more may be combined. For example, the information processing system 10 may perform both the risk assessment process and the patient behavior determination process for a given patient and transmit the processing results to the corresponding nurse's terminal device 600. For example, each device shown in Figure 2 (server system 300, terminal device 600, etc.) may perform the processes shown in Figure 10 and Figure 15 in parallel. In this case, some processes overlap, such as the processes in step S101 in Figure 10 and step S201 in Figure 15, and can therefore be standardized. Also, the first table (Figure 7A) and the third table (Figure 13A) are not limited to being different tables; a single table data set that associates the clinical department, the risk set related to that clinical department, and the instruction set related to that clinical department may be used. In addition, various variations are possible in the combination of the risk assessment process, patient behavior determination process, and call factor estimation process.

[0195] 4. Specific Examples of Risk Disclosure The above describes a process for presenting risks related to medical departments other than the nurse's department as information that is difficult for the nurse to grasp. However, the information processing system 10 of this embodiment may perform an evaluation process on risks that are relatively easy for the nurse to grasp and present the results of the evaluation process to the nurse. Alternatively, the risks subject to the evaluation process may be risks that have a relatively low need to be individually grasped, or they may include risks that are generally likely to occur over a long period of time (such as pressure ulcer risk). In other words, a generally likely risk here is a common risk that can occur in any medical department. For example, in the risk sets for each medical department stored in the memory unit 320 (Figures 7A and 7B), such risks may be included in the risk sets of all medical departments. The processing unit 310 may then determine that risks that are not included in the risk set of the nurse's department but are included in the risk sets of other medical departments are risks that are difficult for the nurse to grasp. In this way, generally likely risks are included in the risk set of the nurse's department and are therefore judged to be risks that are easy for the nurse to grasp. Alternatively, the memory unit 320 may store commonly occurring risks as a common risk set that cannot be associated with a specific medical department. The processing unit 310 may then determine that the risks included in this common risk set are risks that are easy for nurses to understand.

[0196] 4.1 Example of processing for the first risk set The processing unit of the information processing system 10 may perform evaluation processing for each of the multiple risks included in the first risk set, which is a risk set associated with the first clinical department in the first data (Figures 7A and 7B). In this way, it becomes possible to perform evaluation processing for risks that are highly likely to occur in hospitalized patients.

[0197] However, since the first risk set represents risks that are highly likely to occur in the nurse's department, the assessment results for the first risk set may be of little use depending on the nurse's skill level and experience. Furthermore, even if the risk assessment results are useful to the nurse, displaying detailed information for both the first and second risk sets can increase the burden of selecting and filtering information.

[0198] Therefore, the processing unit of the information processing system 10 may calculate a single overall score based on the results of the evaluation process for each of the multiple risks included in the first risk set. The output processing unit of the information processing system 10 then performs a process to present the overall score to the nurses of the first clinical department.

[0199] For example, in step S102 of the server system 300, the processing unit 310 acquires a first risk set of the medical department (affiliated department) to which the patient will be hospitalized based on the first data. Then, in step S103 of the same figure, the processing unit 310 identifies the necessary sensors and trained models for each of the risks included in the first risk set based on the second data. In step S105, the processing unit 310 activates the sensors corresponding to the first risk set. In step S107, the processing unit 310 performs a risk evaluation process for each of the risks included in the first risk set based on the sensing data from the corresponding sensors and the trained models. The evaluation result is, for example, a score representing the degree of risk.

[0200] The processing unit 310 then displays information regarding the second risk set (steps S108, S109) along with a screen related to the first risk set on the display unit 640 of the terminal device 600.

[0201] Figure 24A shows an example of a screen related to the first risk set displayed on the display unit 640. As shown in Figure 24A, the display unit 640 displays, for example, a single integrated score such as "Level 4" for the risk, along with attribute information such as the patient's name, age, and gender. This reduces the amount of information related to easily understandable risks, thereby reducing the burden on nurses in selecting and filtering information.

[0202] For example, in the example shown in Figure 24A, the patient's risks include fall risk and pressure ulcer risk. However, the display unit 640 does not display a score for each risk, but rather displays an overall score. For example, the processing unit 310 may calculate a score as an evaluation result for both fall risk and pressure ulcer risk, and then calculate the overall score by taking the average of these scores (including a weighted average).

[0203] However, as shown in Figure 24A, the processing unit 310 may request recommended measures to reduce risk, and the display unit 640 may display those measures. In this case, it may not be easy to determine specific measures based solely on the overall score. For example, even if the same overall score is obtained, the preferred measures may differ depending on whether the fall risk score is high and the pressure ulcer risk score is low, whether the fall risk score is low and the pressure ulcer risk score is high, or whether both scores are similar.

[0204] Therefore, the processing unit of the information processing system 10 may perform clustering processing to determine which of several classifications the evaluation result of the patient's first risk set falls into, based on the evaluation results for each of the multiple risks included in the first risk set. For each of the multiple classifications, the processing unit determines the recommended response for the nurses of the first clinical department, based on the data associated with the recommended response for nurses and the results of the clustering processing.

[0205] Figure 25 illustrates an example of clustering. The horizontal axis of Figure 25 represents the assessment results for pressure ulcer risk, and the vertical axis represents the assessment results for fall risk. In the example in Figure 25, both pressure ulcer risk and fall risk are classified into three levels, and the coordinate plane is divided into nine regions according to the combination of these levels. Each region is associated with an overall score (level) and recommended countermeasures. The association between regions, overall scores, and countermeasures can be determined using various methods, such as machine learning using past history and expert judgments.

[0206] In this way, the evaluation results for the first risk set can be concisely displayed as an overall score, while also enabling the determination and presentation of specific countermeasures based on the combination of evaluation results for multiple risks included in that first risk set.

[0207] The countermeasures presented here may include recommendations for sensor installation, recommendations for sensor settings, and expert countermeasures from doctors, nurses, etc. (such as changing the patient's position). For example, in the screen shown in Figure 24A, the display unit 640 presents the introduction of a bed exit sensor that detects changes in the patient's position and standing up as a countermeasure.

[0208] In this case, the proposed measures may be associated with a priority level indicating the degree of priority for their implementation. For example, priority may be set from a time perspective, such as by when the measures should be implemented. For example, the processing unit 310 may classify measures into those requiring immediate attention, those requiring attention on the same day, and those requiring attention in 1 to 3 days. The processing unit 310 then sets the priority of measures requiring immediate attention as "high," the priority of measures requiring attention on the same day as "medium," and the priority of measures requiring attention in 1 to 3 days as "low." In this way, it becomes possible to encourage nurses to implement high-priority measures.

[0209] The processing unit 310 may also output a score as an overall score, corresponding to the urgency (priority) of the countermeasures. Figure 24B is an example screen showing a score representing the degree of risk requiring urgent response as the overall score. The specific display content is the same as in Figure 24A, but as described above, the processing unit 310 performs a score calculation process that results in a higher overall score when the priority of the countermeasures is high.

[0210] Furthermore, while Figures 24A and 24B show examples where the overall score for one patient is displayed, the screens displayed on the display unit 640 are not limited to this. Figure 24C shows another example of a screen displayed on the display unit 640. As shown in Figure 24C, the display unit 640 may display a list of overall scores for multiple patients. For example, the display unit 640 may display a screen sorted by the emergency response risk score of multiple patients assigned to a nurse, from highest to lowest. The display unit 640 may also display a numerical value representing the priority of response for each patient. In this way, it becomes possible to appropriately support nurses in deciding the order in which to respond to multiple patients they are assigned.

[0211] As described above, the evaluation results for the first risk set are displayed using an overall score. In contrast, if the second risk set includes multiple risks, the output processing unit of the information processing system 10 may individually present the evaluation results for each of the multiple risks included in the second risk set to the nurses of the first clinical department. In this way, by presenting detailed information for risks that are difficult to grasp and displaying simplified information for risks that are easy to grasp, it becomes possible to provide nurses with appropriate information.

[0212] In this case, the output processing unit may associate a display object for making a call with a doctor with the evaluation results for each of the multiple risks included in the second risk set. The display object for making a call is, for example, the call button shown in Figure 12. In this way, it becomes easier for nurses to consult with doctors about risks that may be difficult to handle, thus promoting more appropriate responses. On the other hand, the output processing unit does not have to associate a display object for making a call with a doctor with the overall score. In this way, the amount of information on relatively less important risks can be reduced, and a user-friendly interface can be realized.

[0213] 4.2 Long-term risks and sudden change risks The multiple risks subject to evaluation include risks that indicate the patient's condition (current state, or more narrowly, activity level), such as pressure ulcers and falls, as well as risks that indicate a sudden change in the patient's condition, such as sepsis and respiratory distress. The information processing system 10 of this embodiment may distinguish between risks that indicate the patient's condition and risks that indicate a sudden change in the patient's condition, and perform display processing according to the characteristics of the risks.

[0214] For example, if the first risk set includes multiple risks that represent the patient's current state due to their activity level and risks that represent a sudden change in the patient's condition due to their illness, the output processing unit of the information processing system 10 processes the risks representing the current state to present an overall score and processes the risks representing a sudden change to present the results of the evaluation process individually. In this way, it is possible to suppress excessive notifications and make it easier for nurses to respond compared to providing uniform notifications for risks with different characteristics.

[0215] Figure 26A is an example of a display screen showing the current status and risk, and Figure 26B is an example of a display screen showing a sudden change and risk. The screens shown in Figures 26A and 26B are displayed, for example, on the display unit 640 of the terminal device 600.

[0216] In the example in Figure 26A, the overall score displayed is "RISK SCORE: 3". Additionally, a button requesting a change in settings may be displayed as a recommended action. These settings could, for example, be related to the on / off status of a sensor that senses the patient, or to the judgment threshold used in the judgment based on the output data of the sensor. Alternatively, these settings could be related to the possibility of installing devices (including sensors) that are not already installed in the patient's room. For example, if the setting of an already installed device is presented as a countermeasure, the setting can be changed via an operation object (the "Apply" button in Figure 26A) displayed on the nurse's terminal device 600. In the example in Figure 26A, if a load sensor is already installed and unused, the display unit 640 displays a screen containing text suggesting the use of the load sensor to detect the patient getting up, and an operation object to initiate its use. The load sensor here could be a sensor installed on the bed 100, or it could be another sensor. The sensor used for detecting the patient getting up could also be another sensor, such as an imaging device 700.

[0217] On the other hand, if a proposal for the installation of a device that has not yet been installed is presented as a countermeasure, the terminal device 600 sends a notification to an administrator terminal authorized to decide on the installation, based on the selection of an operation object displayed on the terminal device 600 (the "Request" button in Figure 26A). The display unit of the administrator terminal (not shown) displays a screen for installation approval. This screen displays a list of patients who already have the target device installed, and patients who have received a request for installation of the device, in association with their overall score. For example, by sorting patients in order of overall score, it becomes possible for the administrator to understand the relative position of patients who have made installation requests. For example, if the majority of patients with an overall score similar to that of the patient who made the installation request have the target device installed, the administrator is more likely to decide to approve the installation request, as it indicates a high need for the device. On the other hand, if the majority of patients with an overall score similar to that of the patient who made the installation request have not yet installed the target device, the administrator is more likely to decide to reject the installation request, as it indicates that it is premature to install the device.

[0218] Thus, when making a decision while considering the circumstances of other patients, it is difficult to compare scores even if scores are displayed for each of the multiple risks. However, as described above, using an overall score makes it easy to compare patients.

[0219] On the other hand, as shown in Figure 26B, when targeting risks indicating sudden deterioration, the scores are displayed individually for each risk. In Figure 26B, the score for heart failure risk is "4.6" and the score for end-of-life risk is "5". End-of-life risk refers to the risk that the patient may die within a specified period, and the risk that care will be needed for that purpose. Although Figure 26B shows an example where information on multiple risks is included on one screen, the display unit 640 may also display different screens for each risk by switching between tabs.

[0220] As shown in Figure 26B, the display unit 640 may display data that forms the basis for calculating the score near the score. In the example in Figure 26B, data such as SpO2, blood pressure, and pulse rate are displayed. For example, the storage unit 320 of the server system 300 may store a table showing parameters (e.g., vital signs) related to the risk of sudden deterioration.

[0221] The memory unit 320 may also maintain a table showing the relationships between each nurse and the attending physician, on-call physician, etc. The processing unit 310 may display a screen on the display unit 640 that shows call buttons, such as a telephone button, corresponding to the risk of a sudden change in condition. When a call button is selected, the terminal device 600 executes the process of making a phone call to the attending physician or on-call physician associated with the nurse. At this time, the terminal device 600 may make the call while the doctor and nurse are sharing the screen display of the score associated with the risk and the vital information used as the basis for its calculation. In this way, it becomes possible to smoothly receive instructions from the doctor regarding risks of sudden changes in condition that require immediate attention.

[0222] The processing unit 310 may also suggest candidate responses based on the degree of similarity to past cases. Figure 26C is a part of the display screen showing the risk of sudden changes shown in Figure 26B, and is another example of the display portion related to end-of-life care risk. As shown in Figure 26C, the processing unit 310 may also display a screen on the display unit 640 that associates the degree of similarity of responses with the call button. In this way, if the past response was a call to a doctor, it is possible to receive appropriate instructions by calling immediately, and if the past response was not a call, it is less likely that a call will be made, thus suppressing unnecessary calls.

[0223] Furthermore, whether or not to notify of risks indicating a sudden change in condition may be customizable for each nurse or patient. For example, notification customization can be done using a screen similar to the one described above, as shown in Figure 22.

[0224] As described above, in the method of this embodiment, the risks included in the first risk set may be classified according to their characteristics. The output processing unit of the information processing system 10 presents an overall score for one classification and displays individual evaluation results for the other classification. In this case, it is possible to reduce the amount of information compared to the example where individual evaluation results are displayed for all of the first risk set, while still providing detailed information for information of high importance within the first risk set.

[0225] 4.3 Variations Figure 25 illustrates a simple clustering based on two risks, but it is not limited to this. For example, there may be three or more risks that are included in the calculation of the overall score. Furthermore, clustering is not limited to being performed solely based on the risk assessment results. For example, the processing unit 310 may perform clustering based on patient attribute information, personality information representing the patient's character, etc., in addition to the risk assessment results.

[0226] Attribute information includes the patient's age, sex, weight, height, and medical history. Personality information may be obtained, for example, by estimating the degree of calmness based on the number of nursing courses per unit time, or by using the results of questionnaires or psychological tests. Alternatively, the processing unit 310 may obtain personality information by performing facial expression analysis using the imaging device 700. Specifically, the processing unit 310 may obtain personality information based on the facial expressions when the patient is awake, from the patterns and degrees of change of each emotion (joy, sadness, anger, fear, surprise, disgust). In addition, personality information may be determined from heart rate variability (HRV), electrodermal activity (EDA), sleep patterns, activity levels, respiratory patterns, voice analysis (voice tone, pitch, speech pattern), electroencephalography (EEG), blood pressure (variability), exercise habits, etc.

[0227] Furthermore, clustering may be performed using other methods such as SVM or regression analysis. The parameters used for clustering may be set by input from nurses or other personnel, or they may be in a format that can be updated using machine learning or similar methods.

[0228] Furthermore, regarding which risks indicate a sudden change in condition, fixed values ​​entered by nurses may be used, or updates may be possible using machine learning.

[0229] Although this embodiment has been described in detail above, it will be readily apparent to those skilled in the art that many modifications are possible without substantially departing from the novel aspects and effects of this embodiment. Therefore, all such modifications are included within the scope of this disclosure. For example, any term that appears at least once in the specification or drawings together with a broader or synonymous term may be replaced with that different term anywhere in the specification or drawings. Furthermore, all combinations of this embodiment and its modifications are also included within the scope of this disclosure. In addition, the configuration and operation of the information processing system, server system, bed, bedside terminal device, terminal device, etc., are not limited to those described in this embodiment, and various modifications are possible. [Explanation of symbols]

[0230] 10... Information processing system, 100... Bed, 110... Processing unit, 120... Memory unit, 130... Communication unit, 140... Operation unit, 150... Drive unit, 160... Movable unit, 170... Mattress, 200... Bedside terminal device, 210... Processing unit, 220... Memory unit, 230... Communication unit, 240... Display unit, 250... Operation unit, 260... Notification unit, 270... Interface Chair unit, 300... Server system, 310... Processing unit, 320... Storage unit, 330... Communication unit, 400... Electronic medical record server, 500... Station terminal device, 600... Terminal device, 610... Processing unit, 620... Storage unit, 630... Communication unit, 640... Display unit, 650... Operation unit, 700... Imaging device, 810... Detection device, 820... Measurement device, 830... Authentication card

Claims

1. An acquisition unit acquires first data that associates a medical department with a risk set representing the risk of patients related to the said medical department, When information is obtained indicating that a patient admitted to the first clinical department is receiving treatment from the second clinical department, the processing unit performs an evaluation process for each of the one or more risks included in the second risk set, which is the risk set associated with the second clinical department from the first data. An output processing unit that presents the results of the evaluation process for the second risk set to the nurses of the first clinical department, An information processing system that includes this.

2. In the information processing system described in claim 1, The acquisition unit is, Second data is obtained that associates the risks included in the risk set, the sensors used in the evaluation process of the risks, and the evaluation algorithm that represents the evaluation process. The aforementioned processing unit, For each of the one or more risks included in the second risk set, An information processing system that uses sensor data from the sensor identified by the second data as input data, and performs processing according to the evaluation algorithm identified by the second data as the evaluation process.

3. In the information processing system described in claim 2, The acquisition unit is, A third set of data that associates the aforementioned medical department with a set of instructions representing the instructions given by the physicians of the aforementioned medical departments to the nurses, A fourth set of data is obtained that associates the aforementioned instruction with a sensor used for determining whether the instruction is being complied with and a determination algorithm that indicates the determination process. The aforementioned processing unit, If the patient admitted to the first clinical department is receiving treatment from the second clinical department, then for each of the one or more instructions included in the second instruction set, which is the instruction set associated with the second clinical department among the third data, The determination process is performed using the sensor data from the sensor identified by the fourth data as input data, and according to the determination algorithm identified by the fourth data. The output processing unit, An information processing system that presents the results of the judgment process regarding the second instruction set to the nurse of the first clinical department.

4. In the information processing system according to any one of claims 1 to 3, When the patient makes a call to summon the nurse, The aforementioned processing unit, An information processing system that performs a process of estimating the cause of the call or the response content by inputting information into a trained model, specifying the patient's electronic medical record information, the patient's spoken content, and the nurse's spoken content as conditions, and specifying the cause of the call or the response content recommended to the nurse in response to the call as output.

5. In the information processing system described in claim 4, If the patient admitted to the first clinical department is receiving treatment from the second clinical department, and the patient makes the call to summon the nurse in the first clinical department, The aforementioned processing unit, An information processing system that acquires information including the second risk set as the electronic medical record information, and performs processing to estimate the factors or the corresponding content based on the acquired electronic medical record information.

6. In the information processing system according to any one of claims 1 to 3, The aforementioned processing unit, The evaluation process is performed for each of the multiple risks included in the first risk set, which is the risk set associated with the first clinical department in the first data. Based on the results of the evaluation process for each of the aforementioned risks, a single overall score is calculated. The output processing unit, An information processing system that performs the process of presenting the overall score to the nurse of the first clinical department.

7. In the information processing system described in claim 6, The output processing unit, An information processing system that, when the second risk set includes multiple risks, individually presents the results of the evaluation process for each of the multiple risks included in the second risk set to the nurse of the first clinical department.

8. In the information processing system described in claim 7, The output processing unit, For each of the multiple risks included in the second risk set, the results of the evaluation process are displayed in association with a display object for making a call with a doctor. An information processing system that does not associate the aforementioned overall score with a display object for making a phone call with the physician.

9. In the information processing system described in claim 6, The aforementioned processing unit, Based on the results of the evaluation process for each of the multiple risks included in the first risk set, a clustering process is performed to determine which of the multiple classifications the evaluation result of the patient's first risk set falls into. An information processing system that determines the recommended response for the nurse in the first clinical department based on data in which the recommended response for the nurse is associated with each of the aforementioned multiple classifications, and the results of the clustering process.

10. In the information processing system described in claim 6, The plurality of risks included in the first risk set include risks that represent the current state due to the patient's activity capacity and risks that represent sudden changes due to the patient's illness. The output processing unit, The process of presenting the overall score for the risk representing the current state is performed. An information processing system that performs a process to individually present the results of the evaluation process for the risks representing the aforementioned sudden changes.

11. Information processing system, First data is obtained that associates a medical department with a risk set representing the risk to patients related to that medical department. When information is obtained indicating that a patient admitted to the first clinical department is receiving treatment from the second clinical department, an evaluation process is performed for each of the one or more risks included in the second risk set, which is the risk set associated with the second clinical department from the first data. The results of the evaluation process for the second risk set are presented to the nurses of the first clinical department. An information processing method that performs processing.

Images