Inspection support system, inspection support device, program, and inspection support method

Abstract

We provide testing support systems that enable safe and efficient testing, taking into account the overall situation of the patient and paramedical staff. [Solution] The examination support system comprises: an acquisition unit that acquires at least one piece of information from patient examination information held in the facility's information system and person status information detected by a detection unit; a detection information generation unit that generates detection information on a specific theme using the information acquired by the acquisition unit; an evaluation information generation unit that evaluates a person's behavior or events using the detection information generated by the detection information generation unit and generates evaluation information; a business proposal information generation unit that generates business proposal information based on at least one piece of information from master information relating to a fixed organization, rule information relating to rules decided by the organization performing the examination work, skill information relating to the skills of the staff and detection information, and evaluation information; and a display unit that displays the business proposal information generated by the business proposal information generation unit.

Description

Technical Field

[0001] The present disclosure relates to an examination support system, an examination support apparatus, a program, and an examination support method. Background Art

[0002] In recent years, along with the increase in elderly patients, the number of examinations has increased and the degree of difficulty of examinations has become more sophisticated in examination operations. In response to this, the number of paramedical personnel such as nurses, radiological technologists, physiological examination technologists, and clinical laboratory technologists who perform and support examinations has been gradually decreasing in line with the trends of the working population. Furthermore, coupled with the improvement of the working environment such as work-style reform, it has become difficult to secure resources that can cope with the increase in patients and examinations and the trend of increasing examination difficulty.

[0003] These situations may lead to increased waiting times and poor examination results for patients, which can further lead to troubles caused by the burden of hospital visit experience, insufficient access to appropriate medical care, and indirect factors for reduced QOL. For paramedical personnel, there is also a concern that the increase in overtime hours, the reduction in possible examinations, and the lack of flexible working hours will reduce the time for guidance and learning for mid-career and young staff, leading to a decline in organizational strength in the medium to long term.

[0004] To address these problems, Patent Document 1 describes a medical care support apparatus that calculates the congestion status when a patient receives medical care in a medical care process using patient position information and displays it on a display screen. Patent Document 2 describes a radiography system that displays specific content of a recognition result on a display when the arrangement of a subject in a camera image does not match the arrangement of the subject in an imaging menu. Patent Document 3 describes a medical diagnosis support apparatus including a determination unit that determines whether or not the content of examination performed by an operator of an X-ray imaging apparatus is appropriate, and a display unit that notifies the operator of the determination result from the determination unit. Prior Art Documents Patent Documents

[0005] Patent Document 1 Patent No. 7116192 [Patent Document 2] Patent No. 7235856 [Patent Document 3] Japanese Patent Publication No. 2008-264231 [Overview of the project] [Problems that the invention aims to solve]

[0006] Conventional technologies can provide support for specific tests to specific patients and paramedical staff. However, there is a challenge in accurately understanding the overall situation of patients and paramedical staff before and during tests, and providing appropriate support information and work information to each patient and paramedical staff member.

[0007] Therefore, the purpose of this disclosure is to provide an examination support system, examination support device, program, and examination support method that can safely and efficiently conduct examinations while taking into account the overall situation of the patient and paramedical staff, in order to solve the above-mentioned problems. [Means for solving the problem]

[0008] The inspection support system related to this disclosure is: A testing support system that aggregates information related to testing operations, An acquisition unit that acquires at least one piece of information from patient examination information held in the facility's information system and person status information detected by a detection unit, A detection information generation unit generates detection information related to a specific theme using the information acquired by the acquisition unit, An evaluation information generation unit generates evaluation information by evaluating a person's actions or events using the detection information generated by the detection information generation unit, A business proposal information generation unit generates business proposal information based on master information about a fixed organization, rule information about rules decided by the organization performing the inspection work, skill information about the skills of individuals, and at least one of the aforementioned detection information, along with the aforementioned evaluation information. A display unit that displays the business proposal information generated by the business proposal information generation unit, It is equipped with.

[0009] The inspection support device related to this disclosure is: An inspection support device that aggregates information related to inspection work, An acquisition unit that acquires at least one piece of information from patient examination information held in the facility's information system and person status information detected by a detection unit, A detection information generation unit generates detection information related to a specific theme using the information acquired by the acquisition unit, An evaluation information generation unit generates evaluation information by evaluating a person's actions or events using the detection information generated by the detection information generation unit, The system includes a business proposal information generation unit that generates business proposal information based on at least one of the following pieces of information: master information relating to a fixed organization, rule information relating to rules established by the organization performing the inspection work, skill information relating to the skills of individuals, and the aforementioned detection information, along with the aforementioned evaluation information.

[0010] The program related to this disclosure is The computer included in the inspection support device that aggregates information related to inspection work, An acquisition unit that acquires at least one piece of information from patient examination information held in the facility's information system and person status information detected by a detection unit. A detection information generation unit generates detection information related to a specific theme using the information acquired by the acquisition unit. An evaluation information generation unit evaluates a person's actions or events using the detection information generated by the detection information generation unit and generates evaluation information. It is configured to run as a business proposal information generation unit that generates business proposal information based on at least one of the following pieces of information: master information about a fixed organization, rule information about the rules decided by the organization that performs inspection work, skill information about the skills of individuals, and the aforementioned detection information, along with the aforementioned evaluation information.

[0011] The inspection support method related to this disclosure is: A method for supporting inspections that aggregates information related to inspection work, An acquisition step that acquires at least one piece of information from patient examination information held in the hospital's information system and person status information detected by the detection unit, A detection information generation step that generates detection information related to a specific theme using the acquired information, An evaluation information generation step involves using the generated detection information to evaluate a person's behavior or event and generate evaluation information, The system includes a business proposal information generation step that generates business proposal information based on evaluation information and at least one of the following pieces of information: master information relating to a fixed organization, rule information relating to rules set by the organization performing inspection work, skill information relating to the skills of the staff, and detection information. [Effects of the Invention]

[0012] According to this disclosure, work proposal information is generated by comprehensively considering the behavior, events, and other circumstances of individuals, including patients and paramedical staff, within the facility, and then proposed to patients and paramedical staff, enabling them to perform their duties safely and efficiently. [Brief explanation of the drawing]

[0013] [Figure 1] This figure shows an example of a schematic configuration of the inspection support system according to this embodiment. [Figure 2] This is a block diagram showing an example of the configuration of the inspection support device according to this embodiment. [Figure 3]This is a diagram showing an example of the functional configuration of the detection information generation unit, evaluation information generation unit, business proposal information generation unit, and information configuration unit according to the present embodiment. [Figure 4] This is a diagram for explaining the hierarchical structure of detection information, evaluation information, and business proposal information according to the present embodiment. [Figure 5] This is a flowchart showing an example of the operation of the examination support apparatus according to the present embodiment. [Figure 6] This is a diagram showing an example of input and output data of the walking state determination learning model according to the present embodiment. [Figure 7] This is a diagram showing an example of input and output data of the examination time evaluation learning model according to the present embodiment. [Figure 8] This is a diagram showing an example of input and output data of the staff determination learning model according to the present embodiment. [Figure 9] This is a diagram showing an example of input and output data of the information configuration determination learning model according to the present embodiment. [Figure 10] This is a diagram showing an example of display information respectively displayed on the tablet of the medical staff and the tablet of the patient according to the present embodiment. [Modes for Carrying Out the Invention]

[0014] Hereinafter, the examination support system, examination support apparatus, program, and examination support method according to the preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

[0015] [Configuration Example of Examination Support System 100] Figure 1 shows an example of the schematic configuration of the inspection support system 100 according to this embodiment. The inspection support system 100 comprises an inspection support device 1, an information system 2, an optical camera 3, a human presence sensor 4, an RFID reader 5, and a display device 6. The inspection support device 1, the information system 2, the optical camera 3, the human presence sensor 4, the RFID reader 5, and the display device 6 are connected to each other via a network N. Examples of the network N include wireless LANs or wired LANs such as a hospital's core LAN, the Internet, as well as WANs, telephone communication networks, satellite communication networks, etc. LAN is an abbreviation for Local Area Network. WAN is an abbreviation for Wide Area Network. In addition, the optical camera 3, the human presence sensor 4, the RFID reader 5, the display device 6, etc. may be directly connected to the inspection support device 1 via wiring. Note that the placement and number of the optical camera 3, the human presence sensor 4, the RFID reader 5, the display device 6, etc. are not limited to those shown in Figure 1.

[0016] The examination support device 1 is composed of a computer such as a server. The examination support device 1 uses one or more collected pieces of information transmitted from the information system 2, optical camera 3, human presence sensor 4, RFID reader 5, etc., to generate suggestion information, including suggestions for the next actions that the paramedical staff or patient should take. The collected information includes, for example, patient information transmitted from the information system 2, camera information transmitted from the optical camera 3, sensor information transmitted from the human presence sensor 4, and entry / exit information transmitted from the RFID reader 5.

[0017] Information system 2 includes, for example, RIS20, electronic medical record 21, LIS22, and nursing support system 23. RIS is an abbreviation for Hospital Information System and is also called the radiology information system. LIS is an abbreviation for Laboratory Information Systems and is also called the clinical laboratory information system. Information system 2 holds personal information such as patient names and examination information including examination types. Information system 2 is linked to the examination support device 1 via network N and transmits examination information to the examination support device 1.

[0018] Optical cameras 3 are installed, for example, in the reception area, waiting room, first examination room, second examination room, staff traffic flow, and corridors between rooms. Optical cameras 3 photograph patients, paramedical staff, etc., who are waiting or moving around in each room, corridor, etc., and transmit camera information, which is an example of status information obtained from the photography, to the examination support device 1. The camera information may be video or still images.

[0019] The motion sensors 4 are installed, for example, in the passageway between the first examination room and the staff traffic flow, and in the passageway between the second examination room and the staff traffic flow. The motion sensors 4 are infrared sensors, ultrasonic sensors, or microwave sensors. The motion sensors 4 detect paramedical staff moving in the passageway, etc., and transmit sensor information, which is an example of status information corresponding to the detection result, to the examination support device 1.

[0020] The RFID reader 5 is installed, for example, at the entrance and exit of the first examination room and the entrance and exit of the second examination room. The RFID reader 5 detects the entry of paramedical staff and patients into each room and their exit from each room, and transmits entry / exit information, which is an example of status information corresponding to the detection result, to the examination support device 1.

[0021] The display device 6 includes a stationary display 60 installed in the waiting room, a tablet 61 held by a paramedical staff member, and a tablet 62 held by a patient. The display 60 is located in the waiting room. The paramedical staff member with the tablet 61 is staying in the first examination room. The patient with the tablet 62 is waiting in the waiting room. The display device 6 displays information transmitted from the examination support device 1, suggesting the next actions for the patient and paramedical staff member.

[0022] [Example configuration of inspection support device 1] Figure 2 is a block diagram showing an example of the configuration of the inspection support device 1 according to this embodiment. The inspection support device 1 comprises a processing unit 10, an operation unit 20, a display unit 30, a storage unit 40, and a communication unit 50. The processing unit 10, the operation unit 20, the display unit 30, the storage unit 40, and the communication unit 50 are connected to each other via wiring 70 such as a bus. In this embodiment, the inspection support device 1 is configured to include an operation unit 20 and a display unit 30, but it is not limited to this configuration, and it may be configured without an operation unit 20 and a display unit 30.

[0023] The processing unit 10 has a processor such as a CPU or GPU, which includes an arithmetic unit, a control unit, and registers, and controls the overall operation of the inspection support device 1. CPU is an abbreviation for Central Processing Unit. GPU is an abbreviation for Graphics Processing Unit. The processor may be configured as a single circuit or as a combination of multiple independent circuits.

[0024] The processing unit 10 includes an in-hospital system linkage unit 11, a sensor linkage unit 12, a detection information generation unit 13, an evaluation information generation unit 14, a business proposal information generation unit 15, and an information configuration unit 16. The processing unit 10 realizes the functions of the in-hospital system linkage unit 11, the sensor linkage unit 12, the detection information generation unit 13, the evaluation information generation unit 14, the business proposal information generation unit 15, and the information configuration unit 16 by executing a program 40a stored in the storage unit 40, for example.

[0025] The hospital system integration unit 11 is connected to the information system 2 via the network N and acquires patient information and other data transmitted from the information system 2. The sensor integration unit 12 is connected to the optical camera 3, the motion sensor 4, and the RFID reader 5 via the network N and acquires camera information transmitted from the optical camera 3, sensor information transmitted from the motion sensor 4, and entry / exit information transmitted from the RFID reader 5. The hospital system integration unit 11 and the sensor integration unit 12 are examples of acquisition units.

[0026] The detection information generation unit 13 uses at least one piece of information acquired by the in-hospital system linkage unit 11 and the sensor linkage unit 12 to determine a specific theme related to the behavior and events of patients and paramedical staff, and generates detection information for each theme. Examples of detection information related to patients include linking information that links patients detected by the detection unit with examination information, walking status information that shows the patient's walking state, movement speed information that shows the patient's movement speed, difficulty information that shows the number of difficult places the patient encountered during examinations, and usage information that shows whether or not the examination room was used. Examples of detection information related to paramedical staff include examination experience information that shows the number of times the paramedical staff have performed examinations, and activity speed information that shows the paramedical staff's examination preparation speed, examination execution speed, etc.

[0027] The evaluation information generation unit 14 uses real-time detection information, as well as at least one of past detection information and examination information, to evaluate certain behaviors and events of patients and paramedical staff, and generates evaluation information for each evaluation item. The evaluation information generation unit 14 can understand the progression of the patient's condition by comparing the current detection information with past detection information. Examples of evaluation information include information indicating the difficulty of performing an examination on a patient waiting in the examination room, and information indicating the trend of the time it takes for a patient to undergo a predetermined examination.

[0028] The business proposal information generation unit 15 generates business proposal information that leads to the actions of patients and paramedical staff during examination work, using at least one of the following information in addition to evaluation information: camera information, detection information, fixed master information, variable rule information, and paramedical staff skill information. Master information refers to information that is unique to a facility or organization and does not change, and includes information such as the number of examination rooms, the area of ​​each room, entrances and exits for patients and paramedical staff, the standard examination time set by the facility, and the temporal performance of the examination equipment. Rule information refers to operational rules set by hospital organizations such as the nursing department, radiology department, physiological testing department, and clinical laboratory department, and includes information such as patient waiting time, overtime hours, submission time for work communication forms, and priority of tasks. Skill information includes paramedical staff examination experience information generated by the detection information generation unit 13, years of service indicating the paramedical staff's experience level, and career history. This skill information may be updated sequentially. Examples of business proposal information include organizational staffing shifts, including suggestions for moving patients and paramedical staff to the laboratory; recommendations for the most suitable staff for specific patients; suggestions for changing thresholds within the acceptable limits of the rules; and suggestions for contacting specific patients. Here, a suggestion for changing thresholds within the acceptable limits of the rules would be, for example, extending the initial threshold from 2 minutes to 4 minutes, when the maximum waiting time is 5 minutes.

[0029] The information configuration unit 16 uses business proposal information, etc., to form a screen configuration that includes display information for suggesting the next actions to the patient and paramedical staff, and outputs it to the display device 6. The display information includes text, images, etc., for suggesting the next actions to the patient and paramedical staff. The patient and paramedical staff may be one person or multiple people of a specific group. The information configuration unit 16 can also change the display information shown on the display device 6.

[0030] The operation unit 20 includes, for example, at least one of a mouse, keyboard, switch, button, and sensor. The operation unit 20 may also be, for example, a touch panel integrated with the display. The operation unit 20 receives instructions in response to user input and outputs an operation signal based on the received instructions to the processing unit 10, which functions as a control unit.

[0031] The display unit 30 is, for example, a display device such as a liquid crystal display or an organic EL display. The display unit 30 can display a GUI for receiving various input operations from the user, display information to be displayed on the display device 6 described above, etc. EL is an abbreviation for Electro Luminescence. GUI is an abbreviation for Graphical User Interface.

[0032] The storage unit 40 includes, for example, at least one of the following storage modules: HDD, SSD, ROM, and RAM. HDD is an abbreviation for Hard Disk Drive. SSD is an abbreviation for Solid State Drive. ROM is an abbreviation for Read Only Memory. The storage unit 40 stores, for example, a program 40a for controlling the overall operation of the examination support device 1 and executing various functions of the processing unit 10, as well as various data. The storage unit 40 is also provided with an information aggregation unit 40b for aggregating and storing information related to examination work. The information aggregation unit 40b stores patient information from the information system 2, camera information from the optical camera 3, sensor information from the human presence sensor 4, entry and exit information from the RFID reader 5, and past detection information. The information aggregation unit 40b also stores master information related to fixed facilities and organizations, rule information related to operational rules established by the hospital organization, skill information related to the skills of paramedical staff, etc.

[0033] The communication unit 50 includes, for example, communication modules such as a NIC and a transceiver. NIC is an abbreviation for Network Interface Card. The communication unit 50 communicates various information and data with the information system 2 and detection units such as the optical camera 3 via the network N.

[0034] [Example configuration of detection information generation unit 13, etc.] Figure 3 shows an example of the functional configuration of the detection information generation unit 13, evaluation information generation unit 14, business proposal information generation unit 15, and information configuration unit 16 according to this embodiment. First, the functional configuration of the detection information generation unit 13 will be described. The detection information generation unit 13 includes a patient / examination identification learning model 130, a walking state determination learning model 131, a movement speed determination learning model 132, an examination difficulty determination learning model 133, and an examination room utilization determination learning model 134. The detection information generation unit 13 also includes an examination experience determination learning model 136 and a movement speed determination learning model 137. Note that the types of learning models corresponding to each theme in the detection information generation unit 13 are not limited to those shown in Figure 3.

[0035] The patient / examination identification learning model 130 is a learning model trained to output linked information that links the person detected by the detection unit with examination information when at least one of the following is input: examination information, camera information, sensor information, and entry / exit information. For example, when the patient / examination identification learning model 130 receives patient information generated at the time of examination registration and camera information of the patient being registered, it outputs linked information that links the examination information with the patient's image. Also, when the patient / examination identification learning model 130 receives camera information showing the patient and the corridor in front of the examination room where the patient is waiting, it outputs linked information that links the examination information with the patient's image. Furthermore, when the patient / examination identification learning model 130 receives camera image showing the patient meeting with a doctor, nurse, etc. from a specific medical department, it outputs linked information that links the examination information with the patient's image. The detection information generation unit 13 outputs the linking information acquired from the patient / test identification learning model 130 to the evaluation information generation unit 14 and also stores it in the information aggregation unit 40b.

[0036] The gait state determination learning model 131 is a model trained to output gait state information for each patient when at least one of the following is input: camera information, sensor information, etc. When associating the patient's name, examination information, etc. with the gait state information, the association information output by the patient / examination identification learning model 130 described above may be used. The gait state information consists of scores such as A, B, C, and D. An "A rating" indicates that the patient is able to walk independently. A "B rating" indicates that the patient is able to walk independently, but their walking speed is unstable due to the use of a cane, etc. A "C rating" indicates that the patient is able to walk independently, but assistive devices such as handrails are needed. A "D rating" indicates that the patient is unable to walk independently and requires a wheelchair. Specifically, when camera information of a patient walking independently in the waiting room is input to the gait state determination learning model 131, it outputs an "A rating" as the gait state information for that patient. By acquiring the gait state information, it is possible to determine the difficulty level when welcoming the patient into the examination room. The detection information generation unit 13 outputs the walking state information acquired from the walking state determination learning model 131 to the evaluation information generation unit 14 and stores it in the information aggregation unit 40b. The walking state may be determined by a single learning model as described above, or it may be determined by combining the results of multiple learning models set up for different purposes. Examples of learning models for different purposes include a determination model for determining the presence or absence of canes, wheelchairs, intravenous drips, walking aids, etc., a determination model for determining walking obstruction factors such as the wearing of casts or the holding status of personal belongings, a determination model for determining the presence or absence of an accompanying person to the patient, and a determination model for determining the patient's walking stability.

[0037] The movement speed determination learning model 132 is a model trained to output patient movement speed information in waiting rooms, corridors, examination rooms, etc., when at least one of the following is input: camera information, sensor information, etc. When associating patient name, examination information, etc. with movement speed information, the association information output by the patient / examination identification learning model 130 described above may be used. The movement speed information consists of scores such as A, B, C, and D, where the movement speed increases from D to A. Specifically, when camera information of a patient moving at a fast speed in a waiting room is input to the movement speed determination learning model 132, it outputs "A judgment" as the patient's movement speed information in the waiting room. The detection information generation unit 13 outputs the movement speed information obtained from the movement speed determination learning model 132 to the evaluation information generation unit 14 and also stores it in the information aggregation unit 40b.

[0038] The examination difficulty determination learning model 133 is a model that has been trained to output difficulty information, which counts the number of difficult areas in a patient, when at least one of the following is input: real-time examination information, past examination information stored in the information aggregation unit 40b, etc. In other words, the examination difficulty determination learning model 133 determines examination difficulties for each patient, including areas where positioning during the examination is not possible or difficult, and precautions to take when administering intravenous fluids, etc. Examples of difficult inspection points include the following: • Orthopedic joint radiography, such as for the ankle, requires a specific shooting angle, and patients often cannot maintain this position due to pain or other reasons. • Inspection conditions that frequently result in retakes due to image defects (model judgment based on past performance data) • Examination conditions that involve actions such as getting onto an imaging table for wheelchair users • Examination conditions requiring a change of clothes due to wearing casts, IVs, etc. When linking patient names, examination information, etc., to information about difficult areas, the linked information output by the patient / examination identification learning model 130 described above may be used. Specifically, when the examination difficulty judgment learning model 133 receives examination information that includes a note indicating that positioning is difficult and caution is required when administering intravenous fluids, it outputs "2 locations" as information about difficult areas. The detection information generation unit 13 outputs the information about difficult areas obtained from the examination difficulty judgment learning model 133 to the evaluation information generation unit 14 and also stores it in the information aggregation unit 40b.

[0039] The laboratory occupancy determination learning model 134 is a model trained to output detection information indicating the presence or absence of use in each laboratory when at least one of camera information, sensor information, and entry / exit information is input. For example, when the laboratory occupancy determination learning model 134 receives camera information showing no person inside the laboratory, it outputs "no detection" as detection information. The detection information generation unit 13 outputs the detection information obtained from the laboratory occupancy determination learning model 134 to the evaluation information generation unit 14 and also stores it in the information aggregation unit 40b.

[0040] The location information determination learning model 135 is a model that has been trained to output location information of paramedical staff necessary for using a specific examination room when at least one of the following is input: camera information, sensor information, and entry / exit information. For example, when camera information for each examination room is input, the location information determination learning model 135 outputs the location of the paramedical staff necessary for using each examination room, the location of the nurses who need to assist them, etc. as location information. The location information is used, for example, to identify paramedical staff when generating work proposal information. The detection information generation unit 13 outputs the location information obtained from the location information determination learning model 135 to the evaluation information generation unit 14 and also stores it in the information aggregation unit 40b.

[0041] The examination experience judgment learning model 136 is a model that, upon input of at least one of camera information, sensor information, and skill information, outputs examination experience information, such as the number of times paramedical staff have performed examinations, categorized by medical department. The examination experience information consists of scores such as A, B, C, and D, with scores increasing from D to A indicating a higher number of examinations performed by the paramedical staff. The detection information generation unit 13 outputs the examination experience information obtained from the examination experience judgment learning model 136 to the evaluation information generation unit 14 and also stores it in the information aggregation unit 40b. The examination experience information stored in the information aggregation unit 40b may be accumulated and updated sequentially, or it may be temporarily stored as separate data and updated when an update is triggered.

[0042] The behavioral speed determination learning model 137 is a model trained to output behavioral speed information, including the examination preparation speed and examination execution speed of paramedical staff, when at least one of the following is input: camera information, sensor information, entry / exit information, and skill information. The behavioral speed information consists of scores such as A, B, C, and D, with scores increasing from D to A indicating faster examination preparation and examination execution speeds of the paramedical staff. In addition to simple behavioral speed determination, the behavioral speed information may also be determined from camera information based on the execution status of specific procedures. Specifically, a learning model can be prepared that defines the procedures for completing a single examination. In this case, by determining the procedures used to perform a single examination, the learning model can individually output determination information such as whether specific procedures were omitted or the time taken for each procedure. The detection information generation unit 13 outputs the behavioral speed information acquired from the behavioral speed determination learning model 137 to the evaluation information generation unit 14 and also stores it in the information aggregation unit 40b.

[0043] Next, the functional configuration of the evaluation information generation unit 14 will be described. As shown in Figure 3, the evaluation information generation unit 14 has a difficulty level evaluation learning model 140 and an examination time evaluation learning model 141. Note that the types of learning models in the evaluation information generation unit 14 are not limited to those shown in Figure 3.

[0044] The difficulty level evaluation learning model 140 is a model that has been trained to output difficulty level information for patients waiting in waiting rooms, etc., when it receives linked information from the patient / examination identification learning model 130, difficulty information from the examination difficulty area determination learning model 133, etc. For example, the difficulty level information consists of A, B, and C scores, with scores decreasing from C to A indicating lower difficulty during the examination. Specifically, when the difficulty level evaluation learning model 140 receives difficulty information indicating that there are "two" difficult areas in an identified patient, it evaluates the difficulty level during the examination as higher than usual and outputs a "C rating" as difficulty level information. The evaluation information generation unit 14 outputs the difficulty level information obtained from the difficulty level evaluation learning model 140 to the business proposal information generation unit 15 and also stores it in the information aggregation unit 40b.

[0045] The examination time evaluation learning model 141 is a model trained to output examination time evaluation information that shows the trend of the expected time required for a patient's examination when at least one of the following is input: gait state information from the gait state judgment learning model 131, difficulty information from the examination difficulty point judgment learning model 133, past information, etc. The evaluation information consists of scores such as good trend, maintenance trend, and worsening trend. Specifically, when the examination time evaluation learning model 141 receives gait state information with a "B rating" and information on "2 difficult points," it evaluates that the examination will take longer than usual and outputs "worsening trend" as the examination time evaluation information. The evaluation information generation unit 14 outputs the examination time evaluation information obtained from the examination time evaluation learning model 141 to the business proposal information generation unit 15 and also stores it in the information aggregation unit 40b.

[0046] Next, the functional configuration of the business proposal information generation unit 15 will be described. As shown in Figure 3, the business proposal information generation unit 15 has a laboratory judgment learning model 150 and a staff judgment learning model 151. Note that the types of learning models in the business proposal information generation unit 15 are not limited to those shown in Figure 3.

[0047] The examination room determination learning model 150 is a model trained to output recommended examination room information when at least one of the following is input: difficulty evaluation information from the difficulty evaluation learning model 140 or examination time evaluation information from the examination time evaluation learning model 141. In addition, the input data for the examination room determination learning model 150 may include waiting area information, such as camera information from an image taken of the waiting room. The waiting area information is used to identify which examination room is closest to the patient by identifying the patient's current location, or to identify another waiting room that is close and available for patients in wheelchairs, etc., depending on how crowded the waiting room is. For example, when the examination room determination learning model 150 receives difficulty evaluation information of "C rating" and examination room usage information of "no detection" for an identified patient, it determines that the difficulty of acceptance and during the examination is high, and therefore the examination room closest to the patient in the waiting room is optimal, and outputs "Examination Room 2" as the recommended examination room information.

[0048] The staff determination learning model 151 is a model trained to output recommended staff information when at least one of the following is input: movement speed information from the movement speed determination learning model 132, examination time evaluation information from the examination time evaluation learning model 141, or past paramedical skill information. Specifically, when the staff determination learning model 151 receives movement speed information of "A rating" and examination time evaluation information of "worsening trend" for an identified patient, it determines that the examination will take a long time and outputs "Ms. Y," who has high experience, as the recommended staff information.

[0049] Next, the functional configuration of the information configuration unit 16 will be described. As shown in Figure 3, the information configuration unit 16 has an information configuration determination learning model 160. Note that the type of learning model in the information configuration unit 16 is not limited to the type shown in Figure 3.

[0050] The information configuration determination learning model 160 generates display information for the display device 6 to suggest the next actions for the patient and paramedical staff, based on recommended laboratory information and recommended staff information. The information configuration determination learning model 160 is a model that has been trained to output display information when it receives recommended laboratory information from the laboratory determination learning model 150 and recommended staff information from the staff determination learning model 151. The display information includes, for example, at least one suggestion from among the following: a suggestion for the next action that paramedical staff should take, a suggestion for instructions to paramedical staff judged by the head of the organization, and a suggestion for the next action that the patient should take. In addition, the information configuration determination learning model 160 can generate display information corresponding to the response information when it receives response information to the display information from the output destination display device 6. The business suggestion information generated by the business suggestion information generation unit 15 described above may simply be shared all at once, or it may be saved together by the information aggregation unit 40b and then displayed in a list. In such cases, the information configuration unit 16 can be omitted from the configuration of the examination support device 1.

[0051] The learning models described above may be generated by algorithms that implement machine learning, such as neural networks, support vector machines, and decision trees. Neural networks include, for example, convolutional neural networks (CNNs), recurrent neural networks (RNNs), and autoencoders. CNN is an abbreviation for Convolutional Neural Network. RNN is an abbreviation for Recurrent Neural Network. The learning method may be supervised learning or unsupervised learning. For example, a learning model may be created using camera images, sensor information, and scores corresponding to each theme of detection information and each evaluation item of evaluation information as training data. In this embodiment, an example has been described in which a learning model is provided in each of the detection information generation unit 13, evaluation information generation unit 14, business proposal information generation unit 15, and information structuring unit 16, but the embodiment is not limited to this. For example, each learning model may be stored in the storage unit 40 of the inspection support device 1, in an external device connected via a network, or across multiple such devices.

[0052] [Hierarchical structure of detection information, evaluation information, and business proposal information] Figure 4 is a diagram illustrating the hierarchical structure of detection information, evaluation information, and business proposal information according to this embodiment. In the lowest layer, the first layer, detection information for multiple themes is generated, fragmenting the behavior and events of patients and paramedical staff, using seamless information such as camera information and sensor information, as well as inspection information from the information system 2. Based on the detection information, by linking with the information system 2, it is also possible to identify patients detected by camera information, etc., and identify the type of examination performed on the identified patients. Examples of detection information related to patients include gait status information, movement speed information, and difficulty level information. Examples of detection information related to paramedical staff include examination experience information and movement speed information.

[0053] In the second layer, patient detection information generated in the first layer, such as gait status information, movement speed information, and difficulty level information, is combined to generate patient group evaluation information, which is an evaluation of certain patient behaviors and events. In the second layer, paramedical detection information generated in the first layer, such as examination experience information and movement speed information, is combined to generate paramedical group evaluation information, which is an evaluation of certain paramedical behaviors and events. In this case, by combining the paramedical individual skill information stored in the information aggregation unit 40b with the above-mentioned detection information, it is also possible to generate evaluation information for the paramedical group that takes past experience into consideration.

[0054] Furthermore, in the second layer, comprehensive evaluation information about patients and paramedical staff can be obtained by combining evaluation information about patient groups, evaluation information about paramedical staff, and examination information. In addition, in the second layer, evaluation information about organizational capabilities can be generated by combining master information and individual skill information stored in the information aggregation unit 40b. Moreover, evaluation information about external resources can also be generated in the second layer. Note that in the second layer, specific evaluation information may be stored and updated as a database in the information aggregation unit 40b, making it available for secondary use.

[0055] In the third layer, business proposal information to be presented to patients and paramedical staff is generated by combining one of the evaluation information generated in the second layer, such as evaluation information on patient groups, evaluation information on paramedical groups, evaluation information on matching, and evaluation information on organizational capabilities.

[0056] [Example of operation of inspection support device 1] Figure 5 is a flowchart showing an example of the operation of the examination support device 1 according to this embodiment. The paramedical staff's tablet 61 and the patient's tablet 62 either have an application that can execute the examination support functions according to this embodiment implemented, or are configured to execute the examination support functions according to this embodiment on a web browser. Furthermore, it is assumed that the patient and the paramedical staff have completed procedures such as user registration in advance in order to use the examination support functions according to this embodiment. In the following section, we will describe the case in which "Mr. X" is identified as the patient.

[0057] The hospital system integration unit 11 acquires examination information for each patient from each device in the information system 2. The sensor integration unit 12 acquires camera information from the optical camera 3, sensor information from the human presence sensor 4, and entry / exit information from the RFID reader 5 (step S1). The examination information, camera information, sensor information, and entry / exit information are stored in the information aggregation unit 40b of the storage unit 40.

[0058] Next, the detection information generation unit 13 generates detection information for each theme using at least one of the inspection information, camera information, and sensor information (step S2). Figure 6 shows an example of the input and output data of the walking state determination learning model 131 according to this embodiment. The detection information generation unit 13 inputs the camera information and sensor information to the walking state determination learning model 131 and obtains walking state information, which is the output of the walking state determination learning model 131. For example, the camera information includes an image of a patient walking with a cane. In this case, the walking state determination learning model 131 determines that the patient's walking speed is unstable and outputs "B judgment" as the patient's walking state information.

[0059] Similarly, the detection information generation unit 13 inputs camera information and other data to other learning models and obtains predetermined detection information. For example, the detection information generation unit 13 obtains linking information, such as "Mr. X" from patient / examination identification learning model 130, with the examination type information "general photography" and the person in the camera image. The detection information generation unit 13 also obtains "2 locations" as difficult examination location information output from the difficult examination location determination learning model 133. Furthermore, the detection information generation unit 13 obtains "A rating" as movement speed information output from the difficult examination location determination learning model 133.

[0060] Next, the evaluation information generation unit 14 generates evaluation information that evaluates certain behaviors and events of the patient and paramedical staff using the detection information, etc. (step S3). If past detection information corresponding to each piece of detection information is stored in the information aggregation unit 40b, the evaluation information generation unit 14 may generate examination time evaluation information using the past detection information. Figure 7 shows an example of input and output data of the examination time evaluation learning model 141 according to this embodiment. The evaluation information generation unit 14 inputs "B judgment" as walking state information and "2 locations" as information on difficult examination points to the examination time evaluation learning model 141. The evaluation information generation unit 14 also inputs "B judgment" as past walking state information and "1 location" as information on difficult examination points to the examination time evaluation learning model 141. Comparing the patient's current difficulty level with past difficulty level, the number of difficult examination points has increased by one. In this case, the examination time evaluation learning model 141 determines that the difficulty level of the identified patient's examination has increased compared to before, and outputs "worsening trend" as examination time evaluation information. Figure 7 shows areas indicating a worsening trend, highlighted in color.

[0061] Next, the work suggestion information generation unit 15 generates work suggestion information that leads to the next actions of the patient and paramedical staff using at least one of the detection information and evaluation information (step S4). Figure 8 shows an example of input and output data of the staff judgment learning model 151 according to this embodiment. The work suggestion information generation unit 15 inputs "deterioration trend" as examination time evaluation information and "A judgment" as movement speed information to the staff judgment learning model 151. Furthermore, in order to improve the accuracy of the recommended staff judgment, the work suggestion information generation unit 15 may also use movement speed information related to the patient, examination experience information related to paramedical staff, location information of paramedical staff necessary to use a specific examination room, etc., as input data. In this case, it is estimated that the identified patient has no problem with walking speed, but requires some effort to get them into the examination room due to their walking condition and difficulty in the examination area. Therefore, the staff judgment learning model 151 outputs "Mr. / Ms. Y" as recommended staff information, who has high experience, is on duty, and is in a specific location.

[0062] Similarly, the business proposal information generation unit 15 inputs "C rating" as difficulty evaluation information, "no detection" as examination room usage information, and waiting area information to the examination room judgment learning model 150. In this case, it is estimated that the identified patient requires an examination room that is close to the waiting room and equipped with facilities to accommodate the patient, as the difficulty level of the examination is high. Therefore, the examination room judgment learning model 150 outputs "Examination Room 2" as recommended examination room information, which is close to the waiting room and equipped with facilities to accommodate the patient.

[0063] Next, the information configuration unit 16 generates display information that suggests the following action to at least one of the patient and the paramedical staff using the business suggestion information, etc. (Step S5). Figure 9 shows an example of the input and output data of the information configuration judgment learning model 160 according to this embodiment. The information configuration unit 16 inputs "Mr. Y" as recommended staff information and "Examination Room 2" as recommended examination room information to the information configuration judgment learning model 160. The information configuration unit 16 also inputs the linked information acquired by the patient / examination identification learning model 130, such as "Mr. X" and "general radiography," to the information configuration judgment learning model 160. The information configuration judgment learning model 160 outputs display information that includes text, images, etc., suggesting the following action to the patient, etc., as display information.

[0064] Figure 10 shows an example of the display information shown on the paramedical staff's tablet 61 and the patient's tablet 62 according to this embodiment. The information configuration unit 16 displays the information 61a "The second examination room is available" on the screen of paramedical staff member Y's tablet 61. Next, the information configuration unit 16 displays the information 62a "Please wait in front of the second examination room" on the screen of the patient's tablet 62. Next, the information configuration unit 16 sequentially displays the information 61b "There are a total of 7 patients in the waiting room" and "This is the patient who has been waiting the longest" on the screen of the paramedical staff member's tablet 61. At this time, the information configuration unit 16 also displays the image 61c of patient X, which has been identified in association with the text "This is the patient who has been waiting the longest". Paramedical staff member Y looks at the image of patient X to confirm whether patient X is in the waiting room. If paramedical staff member Y finds patient X, they touch the image 61c displayed on the screen of tablet 61. When the information configuration unit 16 receives a touch operation on patient X's image 61c by paramedical Y, it uses the information configuration determination learning model 160 to acquire display information corresponding to the input data. Subsequently, the information configuration unit 16 displays the message "You will be called shortly" 62b on the screen of the patient's tablet 62.

[0065] According to this embodiment, detection information is generated by determining a specific theme using examination information registered moment by moment at hospital facilities, real-time camera information of patients and paramedical staff, etc., and evaluation information is generated by evaluating certain actions and events of patients, etc., using this detection information, etc. Furthermore, in this embodiment, work suggestion information is generated that proposes the next actions for patients and paramedical staff using the evaluation information, etc., and display information based on the work suggestion information is displayed on a display device 6 such as tablets 61, 62. As a result, patients and paramedical staff can perform examinations efficiently and safely by acting according to the content of the displayed information. This makes it possible to realize examination operations that allow for the prediction of optimal and high-quality examination results.

[0066] Furthermore, according to this embodiment, in addition to fixed, permanent master information and skill information, by using real-time updated paramedical skill information and constantly changing patient condition information, it is possible to propose the next actions to patients and paramedical staff while considering the real-time fluctuating situation within the hospital facility. Also, according to this embodiment, since work proposal information is generated using past examination information, it is possible to perform optimal evaluations for evaluation items while considering the patient's progress, thus enabling the presentation of more efficient and safer work proposals.

[0067] Although preferred embodiments of this disclosure have been described in detail above with reference to the attached drawings, the technical scope of this disclosure is not limited to these examples. Furthermore, various modifications and improvements will naturally fall within the technical scope of this disclosure, within the scope of the technical ideas described in the claims for those skilled in the art. In this embodiment, examples have been described in which display information is shown on a stationary display 60, a paramedical staff member's tablet 61, and a patient's tablet 62 as the display device 6, but for example, the display information may be limited to paramedical staff members, etc.

[0068] Furthermore, in the embodiments described above, detection information was generated using both inspection information and state information such as camera information, but the system is not limited to this. For example, detection information may be generated using only state information such as camera information. In this case, the camera information can be used to recognize and identify individuals such as patients in a waiting room, and the business proposal information generated by the above process can be presented to the identified patients.

[0069] Furthermore, in the embodiments described above, the input information for each judgment process is merely an example, and other items may be included in other judgments. For example, by using linked information such as patient information obtained by the patient / examination identification learning model 130 for the recommended staff judgment, the nurse in charge of that patient may be selected preferentially. Similarly, in the information configuration judgment, by recognizing a patient with impaired vision based on the identified linked information, it becomes possible to switch the notification method from a screen to an alarm or a call from a nurse. Moreover, these processes may be performed sequentially in conjunction with each other, or multiple judgments may be performed in parallel. [Explanation of Symbols]

[0070] 2. Information Systems 3. Optical camera (detection unit) 4. Motion sensor (detection unit) 5 RFID reader (detection unit) 6 Display device (display section) 10. Processing Unit (Computer) 11. In-hospital System Integration Department (Acquisition Department) 12. Sensor integration unit (acquisition unit) 13 Detection Information Generation Unit 14. Evaluation Information Generation Unit 15 Business proposal information generation department 16 Information Configuration Section 40b Information Aggregation Department 60 Display (Display Unit) 61, 62 Tablet (Display Unit) 100 Inspection Support Systems

Claims

1. A testing support system that aggregates information related to testing operations, An acquisition unit that acquires at least one piece of information from patient examination information held in the facility's information system and person status information detected by a detection unit, A detection information generation unit generates detection information related to a specific theme using the information acquired by the acquisition unit, An evaluation information generation unit generates evaluation information by evaluating a person's actions or events using the detection information generated by the detection information generation unit, A business proposal information generation unit generates business proposal information based on master information about a fixed organization, rule information about rules decided by the organization performing the inspection work, skill information about the skills of individuals, and at least one of the aforementioned detection information, along with the aforementioned evaluation information. A display unit that displays the business proposal information generated by the business proposal information generation unit, An inspection support system equipped with the following features.

2. The aforementioned business proposal information includes at least one of the following: a suggestion for the next action that paramedical staff should take, a suggestion for instructions to paramedical staff that the head of the organization should decide, and a suggestion for the next action that the patient should take. The inspection support system according to claim 1.

3. The aforementioned detection information includes at least one of the following: patient identification, patient examination type identification, patient walking status, patient movement speed, patient's difficulty in the examination area, patient's use of the examination room, paramedical staff's examination experience, and paramedical staff's movement speed. The inspection support system according to claim 1.

4. The aforementioned evaluation information includes at least one of the following: a difficulty rating indicating the difficulty level when a patient undergoes a prescribed examination, and an examination time rating indicating the estimated time required for the patient to undergo the examination. The inspection support system according to claim 1.

5. The system includes an information aggregation unit that stores at least one of the following: the inspection information, the status information, the detection information, the evaluation information, the master information, the rule information, and the skill information. At least one of the detection information generation unit, the evaluation information generation unit, and the business proposal information generation unit generates the business proposal information using the information stored in the information aggregation unit. The inspection support system according to claim 1.

6. The display unit is equipped with an information configuration unit that forms a screen configuration for displaying display information based on the business proposal information on the display unit. The information configuration unit is capable of changing the display information displayed on the display unit. The inspection support system according to claim 1.

7. An inspection support device that aggregates information related to inspection work, An acquisition unit that acquires at least one piece of information from patient examination information held in the facility's information system and person status information detected by a detection unit, A detection information generation unit generates detection information related to a specific theme using the information acquired by the acquisition unit, An evaluation information generation unit generates evaluation information by evaluating a person's actions or events using the detection information generated by the detection information generation unit, A business proposal information generation unit generates business proposal information based on master information about a fixed organization, rule information about rules decided by the organization performing the inspection work, skill information about the skills of individuals, and at least one of the aforementioned detection information, along with the aforementioned evaluation information. An inspection support device equipped with the following features.

8. The computer included in the inspection support device that aggregates information related to inspection work, An acquisition unit that acquires at least one piece of information from patient examination information held in the facility's information system and person status information detected by a detection unit. A detection information generation unit generates detection information related to a specific theme using the information acquired by the acquisition unit. An evaluation information generation unit evaluates a person's actions or events using the detection information generated by the detection information generation unit and generates evaluation information. A business proposal information generation unit generates business proposal information based on master information relating to a fixed organization, rule information relating to rules established by the organization performing the inspection work, skill information relating to the skills of individuals, and at least one of the aforementioned detection information, along with the aforementioned evaluation information. A program to be executed as such.

9. A method for supporting inspections that aggregates information related to inspection work, An acquisition step that acquires at least one piece of information from patient examination information held in the hospital's information system and person status information detected by the detection unit, A detection information generation step that generates detection information related to a specific theme using the acquired information, An evaluation information generation step involves using the generated detection information to evaluate a person's behavior or event and generate evaluation information, A business proposal information generation step that generates business proposal information based on at least one of the following pieces of information: master information about a fixed organization, rule information about rules decided by the organization that performs inspection work, skill information about the skills of individuals, and the aforementioned detection information, along with the aforementioned evaluation information. A testing support method having the following characteristics.

Images