Healthcare management system using biometric sensors, workload assessment program for healthcare management system, and data structure for healthcare management system

Abstract

To provide a health management system that integrally supports both work of a personnel management and health and safety management administrators perform, and, when an excessive work burden is predicted, issues detailed warnings to the administrator and workers to prevent industrial accidents and the like in advance, and is excellent in convenience for both the administrator and worker.SOLUTION: A health management system is the health management system in which a sensor 1, a gateway terminal 2, a server 3 and an information visualization system 4 cooperate thanks to an application. The health management system is configured to, as to bio-data the sensor 1 acquires, convert a core temperature, a heart rate, and a rate of change in a moving average value for each stress index into an evaluation point on the basis of, an item evaluation table, and, when a warning score totaling an evaluation point of four kinds obtained by converting a difference rate obtained in comparison of a moving average value of a respiratory rate with a set reference value into an evaluation point on the basis of the item evaluation table exceeds a first threshold, issue a first warning, and when the score exceeds a second threshold, issue a second warning.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a technology relating to a health management system that uses a biosensor. [Background technology]

[0002] Construction sites, security guards, and other workplaces are often subject to harsh working conditions, such as extreme heat or cold outdoors or in enclosed spaces. In Japan, the aging of workers, caused by the declining birthrate and aging population, is also an issue that must be addressed. Workers vary in age, gender, physical strength, and other attributes, but compared to general office work, on-site work places a heavy burden on both the mind and body, and there is a risk that workers may be suddenly exposed to excessive workloads. In fact, there have been confirmed cases of problems that are difficult for the parties involved to predict, such as workers who were deemed fit to work in advance medical examinations and showed no abnormalities, developing unexpected heatstroke. On the other hand, for managers who employ many workers, the employer liability is a broad and important responsibility. Among the management tasks that managers perform on a daily basis while shouldering such responsibilities, the proper management of labor and occupational safety and health, which are directly related to the safety of workers' health and lives, has become an extremely important and urgent management issue. In recent years, work style reform has been discussed in Japan, and awareness of such management tasks has increased. Coupled with the implementation of the new international standard (ISO45001), the realization of a good working environment that meets more advanced occupational safety and health management standards than ever before is now required. Various research and development efforts have been carried out in the past that have attempted to contribute to the prevention of heat stroke, a serious industrial accident. For example, the heat stroke warning device in Patent Document 1 attempts to warn workers wearing protective clothing at nuclear power plants, etc. of the risk of heat stroke by monitoring their core body temperature and heart rate. Patent Document 2 also focuses on core body temperature and heart rate, but is a small, lightweight earplug-type heat stroke warning device that is easy to use in a variety of environments. In addition, although it has a purpose other than heat stroke prevention, various proposals have been made so far regarding devices for evaluating mental and physical stress, which is considered to be a disturbance in autonomic nervous function. The autonomic nervous function evaluation device in Patent Document 3 evaluates autonomic nervous function by extracting a heart rate cycle fluctuation component corresponding to respiratory fluctuation and a heart rate cycle fluctuation component corresponding to blood pressure fluctuation, and has been proposed as a device that is simpler and cheaper than conventional complicated and expensive products. In addition, a device in Patent Document 4 has also been proposed as a technology for evaluating mental stress by analyzing heart rate fluctuation, i.e., heart rate cycle fluctuation and respiratory cycle. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2009-108451 A [Patent Document 2] JP 2010-131209 A [Patent Document 3] Japanese Patent Application Publication No. 10-328148 [Patent Document 4] JP 2010-234000 A [Non-patent literature]

[0004] [Non-Patent Document 1] "The Science of Stress and the Autonomic Nervous System" (http: / / hclab.sakura.ne.jp / ) verified on February 1, 2021. [Non-Patent Document 2] “Study on mental stress assessment method using heart rate fluctuation” (Life Support vol. 22 No. 3 2010) Matsumoto Yoshiaki, Mori Nobuaki, Mitajiri Ryo, and Jiang Zhongwei (https: / / www.jstage.jst.go.jp / article / lifesupport / 22 / 3 / 22_105 / _pdf) Retrieved February 1, 2021. [Non-Patent Document 3] “Rapid Response System (RRS) - Japanese Society of Emergency Medicine” Shinsuke Fujiwara, Takamitsu Kodama, Kazuaki Ataka, Masashi Nakagawa, and Shigeki Fujitani (https: / / jsem.me / pdf / about_rrs_1301110.pdf) accessed on February 1, 2021. Summary of the Invention [Problem to be solved by the invention]

[0005] The objective of the present invention is to provide a health management system and a program to be used therewith, which effectively supports the proper and efficient execution of both labor management (employment management) and occupational safety and health management (health management) for both physical and mental aspects, carried out by managers, for workers engaged in harsh tasks at remote work sites where direct management is difficult. Regarding heat stroke prevention and stress assessment, various proposals have been made, such as those in the above-mentioned patent documents, but before a major health risk such as heat stroke appears, excessive stress accumulates continuously in both the mind and body, regardless of whether the worker is aware of it or not. It is necessary to detect minute signs of increased risk as early as possible while also taking into account individual differences between workers, and to prevent the risk from appearing, but it is difficult to think that such a problem has been solved by conventional technology. Furthermore, with the enforcement of a new international standard (ISO45001) that requires the establishment and continuous maintenance of an advanced occupational safety and health management system, there is a growing momentum to more appropriately manage the working conditions and both physical and mental health of workers than ever before. However, a health management system that comprehensively supports these management tasks and is easy to use and convenient for both managers and workers has not yet been realized. For these reasons, the problem that this invention aims to solve is to realize a health management system that can prevent risks from manifesting, by making it possible to assess workload more precisely than ever before and issue warnings when an increase in excessive workload that will adversely affect the health of workers is predicted, and that can provide integrated support to managers in both labor management and physical and mental occupational safety and health management tasks, thereby realizing a health management system that is easy to use for both labor and management. [Means for solving the problem]

[0006] The present invention, which is a means for solving the problems, provides a health management system using a biological information sensor as described below, and a workload judgment program used therein. The health management system is a system in which sensors, gateway terminals, and servers work together through application software (hereinafter referred to as "apps"), and if necessary, an information visualization system is incorporated to improve the user interface and assist users in obtaining and understanding information. The sensor acquires biometric data related to the sensor wearer's core body temperature (C=Core·Temperature), heart rate (H=Heart·Rate), stress index (L=LF / HF), and respiratory rate (R=Respiratory·Rate), and transmits these and other data over time to the gateway terminal. The gateway terminal analyzes and processes the received biometric data as time-series data, converts it into a moving average value per predetermined unit time, and transmits it to the server together with the work data and other data input by the sensor wearer from the gateway terminal. The server receives the moving average value per predetermined unit time and other data related to these biometric data, and manages them by linking them to an identification code (ID) for each sensor wearer. Then, a web page is provided on the browser of the access terminal where the biometric data, work data, and other data for each sensor wearer can be inquired for system users who access the system via an information and communication network such as the Internet. In addition, the server is provided with an application having an item evaluation table that defines comparison values ​​for determining evaluation points related to deep body temperature (C), heart rate (H), stress index (L), and respiration rate (R). Then, for each of deep body temperature (C), heart rate (H), and stress index (L), the rate of change obtained by comparing the moving average value at the present time with that of a predetermined time ago is converted into an evaluation point based on the item evaluation table, and the difference obtained by comparing the moving average value at the predetermined time ago with the set reference value of the respiration rate (R) is converted into an evaluation point based on the item evaluation table. Next, a warning score is obtained by summing up the four evaluation points of deep body temperature (C), heart rate (H), stress index (L), and respiration rate (R), and when this exceeds a first threshold, first warning data is transmitted to the gateway terminal and the access terminal, and when the warning score exceeds a second threshold, second warning data is transmitted. That is, the sensor, gateway terminal, and server work together through the application to execute the workload determination function of this system. In addition, an emergency notification function is added to this workload judgment in order to predict risk increases in advance and prevent them from manifesting. When the movement of the moving average value of deep body temperature (C) and heart rate (H) is confirmed to be either increasing or decreasing in the same trend continuously, the evaluation points for each are added (or the "warning score" is added directly depending on the specifications. The same applies below). Furthermore, if the same trend continues for more than a specified number of times, second warning data is sent to the gateway terminal and the access terminal. In addition, posture data of the sensor wearer is obtained, and if this is continuously confirmed as information indicating a recumbent state and exceeds a specified time, the warning score is added. In addition, if the moving average value of the respiratory rate (R) exceeds a specified number of times, the evaluation point is also added. In addition to these emergency notifications, the system is equipped with a correction function based on the exercise intensity during work to adjust the workload judgment related to the issuance of a warning in order to issue a warning with a higher priority on safety or to suppress issuance of a warning that is considered excessive for the work content. That is, the exercise intensity is calculated from the maximum heart rate set according to the age of the sensor wearer and the actually measured heart rate (H), and when the rate of change of the moving average value of the exercise intensity per predetermined unit time is increasing for a predetermined number of consecutive times and the rate of change of the moving average value of the respiratory rate (R) is decreasing for the same predetermined number of consecutive times, a correction process is performed in which a predetermined coefficient exceeding 1 is multiplied by the evaluation point. In addition to the method using the maximum heart rate, there are other calculation methods for exercise intensity based on the reserve heart rate or oxygen intake. The following description will be given taking the exercise intensity based on the maximum heart rate as an example.

[0007] The workload judgment program for this system executes information processing according to the following process. The program includes a first step of obtaining time series data from four types of biometric data related to the sensor wearer's deep body temperature (C), heart rate (H), stress index (L), and respiration rate (R), and a second step of obtaining a moving average value per a predetermined unit time from the time series data. Next, for each of the deep body temperature (C), heart rate (H), and stress index (L), a rate of change obtained by comparing the moving average value between a predetermined time ago and the present time is converted into an evaluation score based on an item evaluation table, and in conjunction with these conversion processes, for the respiration rate (R), a difference obtained by comparing a set reference value with the moving average value between a predetermined time ago is converted into an evaluation score based on the item evaluation table. The program includes a fourth step of issuing a first warning when the warning score obtained by adding up the evaluation scores of the four items related to the deep body temperature (C), heart rate (H), stress index (L), and respiration rate (R) obtained in the first to third steps, which are executed in a synchronized and continuous manner, exceeds a first threshold value, and a second warning when it exceeds a second threshold value. Depending on the specifications, there is a fifth step in which, when the rate of change of the moving average value of the core body temperature (C) and the heart rate (H) increases or decreases continuously for a predetermined number of times, the evaluation points for each are added, and when the consecutive number of times exceeds a predetermined number set at a higher level, a second warning is issued. There is also a sixth step in which posture data is obtained from the sensor wearer, and this is continuously confirmed as information indicating a recumbent state, and when this exceeds a predetermined time, the warning points are added. Furthermore, the workload judgment program has an emergency notification function and includes a seventh step in which the evaluation points are added when the moving average value of the respiratory rate (R) exceeds a predetermined number of times. The fifth to seventh steps are executed in synchronization with the workload judgment of the third and fourth steps. Furthermore, the workload assessment program is equipped with an exercise intensity correction function that has an eighth step as a correction process that analyzes the exercise intensity during work from a maximum heart rate preset in accordance with the age of the sensor wearer and the heart rate (H) measured by the sensor, and multiplies the evaluation score by a predetermined coefficient exceeding 1 when the rate of change of the moving average of the exercise intensity per predetermined unit time shows an increasing trend for a predetermined number of consecutive times, and at the same time, the rate of change of the moving average of the respiratory rate (R) shows a decreasing trend for the same predetermined number of consecutive times. In addition, the data structure for the health management system that is analyzed and processed in the workload judgment of this system is used to continuously improve the accuracy of the data collection table and item evaluation table. Effect of the Invention

[0008] This system can monitor and manage the health condition of the worker W who wears the sensor in real time, and also performs a multifaceted analysis of excessive stress on both the mind and body to detect signs of mental and physical disorders, including heat stroke, which are difficult to predict, and prevent the manifestation of these risks as early as possible. In addition, by supporting occupational safety and health management that covers not only the physical aspect but also both the mental and physical aspects, the health management system will comply with new international standards, and labor management, including cumbersome administrative procedures such as reporting and confirmation by both labor and management, will be simplified and integrated, making it possible to continue carrying out both of these management tasks effectively and efficiently. Regarding the workload judgment for issuing a warning, in addition to using core body temperature (C) and heart rate (H), which are heavily influenced by physical factors, as the main evaluation items, stress index (L) and respiratory rate (R) are also used as biological data that mental factors also affect the physical aspect, and judgment is made based on a multifaceted evaluation item. Furthermore, emergency notifications and exercise intensity correction make it possible to make more detailed judgments than before. And, by linking the item evaluation table and the data accumulation table, we have created a highly scalable workload assessment function that can be adapted to each type of work with different environments and contents, and each worker with individual differences.In addition, based on the data structure for the health management system that is accumulated and associated on a large scale in this system, it is a health management system that can realize continuous improvement in the accuracy of the workload assessment function. [Brief description of the drawings]

[0009] [Figure 1] 1 is an overview of the overall configuration of the system. [Diagram 2] This is an example of data handled by the system. [Diagram 3] 1 shows a flow of biological data analysis processing. [Figure 4] 1 is an outline of a workload judgment program. [Diagram 5] 1 is an example of an item evaluation table. [Figure 6] 1 is an example of a data collection table. [Figure 7] 1 is an example of a home screen. [Figure 8] 13 is an example of a user management menu screen. [Figure 9] 13 is an example of a work place management menu screen. [Figure 10] 13 is an example of a work content management menu screen. [Figure 11] 13 is an example of a measurement parameter management menu screen. [Figure 12] 13 is an example of an attendance setting menu screen. [Figure 13] 13 is an example of a mail setting menu screen. [Figure 14]13 is an example of a work start screen of the gateway terminal. [Figure 15] 13 is an example of a work screen of a gateway terminal. [Figure 16] 13 is an example of a deep body temperature screen of a gateway terminal. [Figure 17] 13 is an example of a warning confirmation screen. [Figure 18] 13 is an example of a work viewer screen. [Figure 19] 13 is a portion of the graph display section of the sensor information screen. [Figure 20] 13 is a portion of the data display section of the sensor information screen. [Figure 21] 13 is an example of a work management screen. [Figure 22] This is an example of a labor management document (work schedule). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] The present invention will now be described with reference to the drawings. Figure 1 shows the basic configuration of this system, which constitutes an information network system. At both ends of the network, both an administrator M and a worker W are displayed as system users in the broad sense. Worker W at the bottom left is a sensor wearer who wears sensor 1 and operates a gateway terminal 2. Administrator M at the top left is a system user in the narrow sense who receives direct support from this system for management tasks, and operates an access terminal 5 in a location separate from worker W. A typical worker W is someone who is away from the direct control of a manager M and who works on-site in harsh working environments, such as outdoors in the blazing sun or freezing cold without air conditioning, or in closed spaces where both physical and mental stress is excessive. Workers W in construction work, security guard work, and even decommissioning work at nuclear power plants fall into this category regardless of their employment status, and there are a variety of worker W levels with a variety of duties. Managers M are people who are involved in the overall management of the business, including the labor of workers W and occupational safety and health, and include not only managers but also middle managers and those who assist them in their duties. Since both systems can have many people with various job duties and belonging to different hierarchical levels coexisting, the users of this system, Worker W and Manager M, are not in a one-to-one relationship between individual individuals, but rather in a many-to-many relationship in which multiple parties are interrelated.

[0011] In the health management system according to the present invention, the sensor 1, the gateway terminal 2, and the server 3 work together, and an information visualization system 4 is incorporated as a function improvement tool depending on the system specifications. In FIG. 1, the sensor 1 worn by the worker W is a tiny terminal for acquiring biometric data. The gateway terminal 2 is a portable terminal for information processing that functions in cooperation with the sensor 1. Both are operated by the worker W at the work site. The server 3 installed at a location away from the work site is responsible for data analysis and management, and has a database function. The information visualization system 4 that works with it is responsible for visual information processing, so-called visualization processing, to improve the user interface and to increase the usefulness of data. However, in many cases, the server 3 and the information visualization system 4 are functionally integrated. The access terminal 5 is an information terminal that the manager M uses for data communication with the server 3 in order to use the support service provided by this system. The manager M operates it at a management office or the like away from the work site to access this system. Of these five devices, the devices that make up this system are the sensor 1, the gateway terminal 2, the server 3, and the information visualization system 4. Each of these devices is a system in which hardware and application software for this system work together to process information such as biometric data acquired from the worker W. As will be mentioned below from time to time, in order to determine the optimal specifications for this system taking into account the performance of each device, it is possible to appropriately share the functions of the data analysis process between each device. Therefore, as variations of this system, it is anticipated that the system may be modified to have different functional divisions between each device, or, for example, to omit some of the visualization processing functions.

[0012] When starting work at a site, worker W wears sensor 1 near his chest, where biometric data such as body temperature and electrocardiogram can be acquired. Sensor 1 is designed to be ultra-compact and lightweight, so it does not interfere with work, but a fixing device for wearing it is used if necessary. After wearing sensor 1, worker W performs the initial operation of gateway terminal 2, and sensor 1 acquires multiple pieces of biometric data from worker W while working, and transmits them to gateway terminal 2 via wireless communication. At this time, activity data acquired by sensor 1, such as the number of steps, and environmental data such as temperature are also transmitted together with the biometric data. The data transmitted by the sensor 1 is received by the gateway terminal 2 at the work site. The gateway terminal 2 is a small, easily portable terminal equipped with functions for inputting, outputting, transmitting, processing, and managing data, and is equipped with an application for this system. The worker W can check the received data and communication status on the gateway terminal 2, and also uses it to input work data such as the start of work and report data such as physical condition. The various data including the biometric data acquired by the sensor 1 and transmitted to the gateway terminal 2, and the work data input by the worker W on the gateway terminal 2, are transmitted to the server 3 installed at a remote location from the work site via an information and communication network. At this time, the gateway terminal 2, for example, converts the biometric data into a one-minute moving average value, and adjusts the flow of data between the sensor 1 and the server 3.

[0013] The server 3 receives a large amount of data transmitted by the multiple gateway terminals 2 from multiple work sites. The data is analyzed according to its type and managed in a database as a timely usable data group. Due to the specifications of this system, the server 3 mainly performs multifaceted analysis of biodata in stages and in part of the process simultaneously. At that time, analysis processing branched into multiple processes is also performed for both the physical and mental workload that the worker W unconsciously accepts continuously. Then, when necessary, a staged warning is issued to inform the manager M and the worker W of a high-risk excessive load state. Also, depending on the specifications of the system, visualization processing is performed when providing information in an information visualization system 4 that is functionally integrated with the server 3 and incorporated into the app. The various data relating to work and health that have been analyzed and accumulated in the server 3 are provided to the manager M and the workers W according to preset processing conditions or in response to individual requests from the gateway terminal 2 and the access terminal 5. The manager M connects to the Internet using the access terminal 5 and accesses the system from a login screen on the web. Therefore, the manager M can access the system even outside the management office as long as he is in an environment where he can connect to the Internet. The manager M can check the status of the workers W who are working in remote locations in real time, and the workers W can also know their own health conditions and overwork conditions.

[0014] The components of this system, that is, the sensor 1, the gateway terminal 2, the server 3, and the information visualization system 4, as well as the access terminal 5 used by the administrator M to connect to this system, will be described individually. The first component is the sensor 1. This acquires the biological data that is the main subject of analysis in this system, that is, biological data related to deep body temperature (C), heart rate (H), stress index (L) and respiratory rate (R) from the worker W as sensing data, and transmits each data to the gateway terminal 2 in chronological order. The sensor 1 is a device called a biological information sensor, and in order to reliably acquire the biological data, the electrodes of the sensor 1 are attached in close contact with the chest of the worker W, for example, by using a fixing device developed specifically for this system. The sensor is becoming smaller and more functional day by day, and products that can continuously acquire an electrocardiogram are provided in order to visualize the activity of the heart as an electrocardiogram. The multi-functional and highly functional sensor 1 can directly acquire biological data such as body surface temperature in addition to the electrocardiogram from the worker W, and can further analyze and process these primary biological data to obtain secondary biological data such as deep body temperature (C) and heart rate variability (also called "heart rate fluctuation"; hereafter, all of this will be referred to as "heart rate variability"). Moreover, the same sensor 1 can simultaneously acquire activity data such as the number of steps taken by the worker W and changes in posture, and environmental data such as the ambient temperature, relative humidity, and altitude of the work site. As an example of an ultra-compact, highly functional sensor 1, a small, lightweight, power-saving wearable biosensor provided by AffordSense Inc. (hereinafter referred to as "Company A") exists as a terminal that can be used in this system. In the current development process, the applicant has used a sensor 1 that has been optimized for this system after improvements by Company A, and has conducted demonstration tests at a work site. Its size is 6 cm x 3 cm x 1 cm in "total length x width x thickness", weighs about 15 g, and has a rounded shape that significantly reduces the burden on the worker W. It can operate continuously for a long time using a lithium-ion battery or the like, and can transmit data continuously via Bluetooth (registered trademark), which is one of the short-range wireless communication standards for digital devices. The sensor 1 calculates the heart rate (H), heartbeat interval, and heartbeat variability through electrocardiogram analysis, and further calculates a stress index (L) by analyzing the heartbeat variability. The sensor 1 also estimates the deep body temperature (C) from the body surface temperature and heat flow. Details of the stress index (L) will be explained later.

[0015] However, conventional biometric sensors have limitations in their information processing capabilities, and the amount of calculation processing and data processing that can be performed within them is also limited. For this reason, it is assumed that data is sent to devices subsequent to the gateway terminal 2, which has high processing capabilities, and that much of the complex analysis is left to the gateway terminal 2 and the server 3. However, due to recent improvements in microprocessors (MCUs), the sensor 1, which has become high-performance by being equipped with the necessary storage device and calculation device, is now in a situation where it can advance the analysis processing that would otherwise be postponed to a later process and perform it itself. This allows the system specifications to be optimized by dividing the functions among the devices in consideration of the processing capabilities of each device, etc., and the necessary analysis processing is performed within the sensor 1, and only the minimum amount of data is transmitted to the gateway terminal 2, reducing the burden on both the sending and receiving sides. With such improvements in the capabilities of the sensor 1 as a background, it is also possible to further advance part of the analysis processing of the server 3 and have the gateway terminal 2 share part of the functions, but for convenience and from the viewpoint of consistency of explanation, the following system will be described. That is, the worker W operates the gateway terminal 2, the connection between the sensor 1 and the gateway terminal 2 is established, and the connection between the gateway terminal 2 and the server 3 is also established, and data communication becomes possible between the sensor 1, the gateway terminal 2, and the server 3 that constitute this system. The biometric data acquired by the sensor 1 and transmitted to the gateway terminal 2 is analyzed by the application of the gateway terminal 2 and transmitted to the server 3. As an example of this analysis process, the biometric data continuously transmitted from the sensor 1 is received and temporarily stored in the storage area of ​​the gateway terminal 2, converted into time series data of one-minute moving average value, and transmitted to the server 3 after arranging the communication data and reducing the total amount of data. This means that the analysis process is optimally shared between the sensor 1 and the server 3, the flow of data is made efficient, and the load on the server 3 where data is accumulated on a large scale is reduced. Regarding the data accumulation in the server 3, the administrator M can periodically download the necessary data to the access terminal 5, and the load can be further reduced by deleting the data in order starting from the oldest data. The system will be described as one in which the server 3 estimates the workload of the worker W using a workload judgment function installed in the application based on the biometric data transmitted from a large number of gateway terminals 2, and issues a warning to the gateway terminal 2 and the access terminal 5 as necessary.

[0016] The second component of this system is the gateway terminal 2. The worker W wearing the sensor 1 performs an initial operation on the gateway terminal 2, pairs the sensor 1 with the gateway terminal 2, and establishes a data communication path from the sensor 1 to the server 3. The gateway terminal 2 receives and analyzes the biometric data transmitted from the sensor 1 over time, calculates a moving average value per predetermined unit time for each piece of biometric data, and transmits it to the server 3 together with the work data input by the worker W from the gateway terminal 2. The worker W carries the gateway terminal 2 together with the sensor 1, but unlike the sensor 1, it is not directly attached to the body, and it is sufficient that it is managed within a range where it can communicate with the sensor 1. The gateway terminal 2 has a built-in GPS, so it can grasp the current location and movement route of the worker W. A dedicated terminal for this system may be used, but in consideration of cost reduction and portability, small information and communication terminals commonly used in society such as smartphones and tablets can be used, and a dedicated app that runs on an OS such as iOS or Android (registered trademark) is built in. The operation and behavior of the app and its functions will be explained later as a specific method of using this system. Worker W inputs work data from the gateway terminal 2 in order to report the work engagement status, such as the start time and end time, to manager M. The input work data is sent to the server 3, linked with the biometric data, and managed by the server 3. Worker W can check the communication status with the sensor 1 and the received biometric data on the gateway terminal 2, and also uses it to check warnings issued by the server 3. Worker W can also use the gateway terminal 2 for various business communications with manager M.

[0017] The third component is a server 3, which is often connected to a plurality of gateway terminals 2 and access terminals 5 via an information and communication network such as the Internet. This server 3 has a database function, and receives biometric data that has been converted into time-series data of moving average values ​​per predetermined unit time by the gateway terminal 2, work data input by the gateway terminal 2, and the like. The large amount of received data is analyzed by the server 3 and managed for each worker W, i.e., each sensor wearer, and a web page is provided where individual biometric data, work data, and the like can be appropriately referenced. Then, according to preset processing conditions or in response to a request from the manager M, etc., the server 3 extracts necessary information from the database in which data has been accumulated, and transmits it as data that can be displayed on the access terminal 5 or as data that can be printed by a printer, etc. In addition, this system estimates the mental and physical workload state of the worker W in real time using time-series data obtained by analyzing the biodata and the workload judgment function of the server 3, and transmits warning data related to caution advice regarding excessive workload (YA=Yellow-Alert) and danger notice (RA=Red-Alert) to the worker W and the manager M. The types of data processed by the apps built into each device and an overview of the analysis process will be explained later as examples of data and the flow of data processing. This server 3 can be installed in an individual company, or can be located in a remote location different from the location of the company, or can be installed jointly for the purpose of centralized management of multiple work sites of multiple companies. The users of this system, such as the administrator M and the worker W, and the gateway terminal 2 and the access terminal 5 connected via the server 3, may be users or terminals of a single company, etc., or may be users belonging to different organizations, or may be terminals managed individually by each organization. This system can become a health management system that can be shared by multiple companies, etc. beyond the framework of a single organization, and the server 3 can also be installed as a data center that collectively manages large amounts of data, and the health information management service related to this system can be provided on the cloud.

[0018] When providing health information management services using this system, the information visualization system 4 can be incorporated as a functionality improvement tool to improve the user interface and support the manager M in quickly and accurately acquiring and understanding information. As an example, it is possible to use an application that has already been put into practical use as a business intelligence (BI) tool, with some modification. As a result, the information visualization system 4 functions together with the server 3 to easily visualize the results of analysis and processing of the worker W's biological data on the web space provided by this system in response to a request from the manager M. As a specific example, when many workers W are working at multiple sites, the map displayed on the access terminal 5 displays individual worker tabs based on the GPS data from the gateway terminal 2, allowing the manager M to immediately grasp the positions of the workers W and the work sites. When the work location tab displayed in a list is clicked, a surrounding map including the site is immediately displayed on the screen. Furthermore, when the sensor information tab displayed in a list is clicked, the biometric data for each worker W is displayed in an easy-to-read form as a graph or a list.

[0019] The access terminal 5 is a terminal used by the manager M to access the system to obtain information and monitor and manage the status of the worker W in real time. Any general-purpose personal computer or tablet that is already widely used in society can be used, and there is no limitation on the model as long as it meets the required specifications. The manager M connects the access terminal 5 to the Internet and accesses the system by entering an ID and password on a login screen on the web. Therefore, it is possible to access the system from outside the management office if the tablet or smartphone is in an appropriate communication environment. Typically, such terminals are equipped with input / output devices, storage devices, arithmetic units, and transmission / reception devices in advance, or these can be added or connected later. Therefore, in the access terminal 5, the biometric data and working data transmitted from the gateway terminal 2 and the server 3 are displayed on the monitor in an easy-to-read manner, for example, as graphs, and the data can be downloaded and saved. In addition, information processing suited to the characteristics of the data is executed, such as outputting a warning transmitted after workload judgment visually (images, etc.) and audibly (voice, etc.) so that the manager M, etc. can recognize it reliably. When the manager M needs such data in the management of labor and occupational safety and health, he can output it from a printer as a work schedule, record book, etc., and view and store it.

[0020] The main data handled by this system is shown in Figure 2. The biometric data of worker W shown on the left is the core data for analysis processing. As information on body temperature, in addition to the surface temperature of the skin where the sensor 1 comes into direct contact, deep body temperature (C), which is less affected by the outside air temperature, is considered, and this system places emphasis on the latter. Deep body temperature (C) fluctuates similarly to values ​​measured at the eardrum and rectum, and can be estimated by making corrections to these values, but it can also be estimated by performing a conversion process based on the surface body temperature and heat flow measured by the sensor 1 so as not to interfere with work. The sensor 1 acquires an electrocardiogram, which indicates the activity state of the heart, and based on this, analyzes the heart rate (H) and the respiration rate (R) related to oxygen intake. As will be described later, the electrocardiogram analysis process can obtain the heart rate (H) and the heartbeat interval, as well as information on heart rate variability. Furthermore, by analyzing the heart rate variability, the sensor 1 can obtain a stress index (L), which is an index of autonomic nerve balance and is related to the load on both the physical and mental aspects of the worker W. These analysis processes can be performed by the sensor 1, but it is also possible to assign the functions to other devices. The respiratory rate per unit time (R) is understood as the reciprocal of the breathing interval, and can be an index to determine whether the oxygen intake of the worker W is normal. Just as excessive mental and physical stress manifests as hyperventilation, this respiratory rate (R) is also susceptible to both mental and physical stress, like the stress index (L) and heart rate (H). The wearable biosensor manufactured by Company A directly measures the body surface temperature and electrocardiogram of the worker W, and through analysis processing using the installed real-time built-in signal processing algorithm, it is possible to instantly obtain these biodata required for this system. The biodata acquired by the sensor 1 is continuously transmitted to the gateway terminal 2, where it is analyzed and used in the subsequent information processing process.

[0021] In addition to the above-mentioned biological data, there are activity data and environmental data acquired by the sensor 1 and transmitted to the gateway terminal 2, and other device data such as the communication status and battery level of the sensor 1. The activity data is measured by an acceleration sensor built into the sensor 1, and includes the number of steps indicating the movement status of the worker W, and information related to the posture including the state of standing or lying down. The environmental data includes the environmental temperature, relative humidity, air pressure, and altitude of the work site and the vicinity of the sensor 1. The worker W is constantly subject to the effects of extreme heat, etc., both physically and mentally, and this also affects the workload. The values ​​of temperature, etc. measured by the sensor 1 are values ​​inside the work clothes, but in harsh environments where protective clothing is mandatory, the values ​​are higher than under other conditions. The heat index (WBGT value), which indicates the risk of heatstroke, is also environmental data, and can be analyzed from other acquired environmental data, and can also be acquired via the Internet by electronic information provision services and email distribution services. Furthermore, in addition to the data acquired by the sensor 1, there is also work data and report data input by the worker W from the gateway terminal 2.

[0022] Figure 3 shows an overview of the data flow as an analysis process flow, focusing on the main biological data. With regard to electrocardiogram and body temperature, which are the basis of data analysis in this system, multiple analysis processes are performed simultaneously and in parallel while the sensor 1 and gateway terminal 2 work together. Necessary information processing is also performed for each of the other biological data, activity data, environmental data, work data, and report data according to the set analysis procedures, but detailed explanations of these are omitted to avoid complicating the following explanation. First, the flow of data processing in sensor 1 will be described. Biometric data is acquired from the body of worker W by sensor 1 paired with gateway terminal 2. The apps installed on both devices work together to automatically and continuously acquire data, which is the initial point of data flow in this system. The biological data acquired by sensor 1 is one-off data measured at regular intervals such as seconds, but it is acquired and recorded over time and used as time-series data. Among biometric data, the most important information is electrocardiogram. Sensor 1 detects and amplifies the weak electric current generated by the cardiac activity of worker W, specifically the contraction activity of the heart muscle, from the body surface. Graphing the obtained electrocardiogram data over time gives an electrocardiogram, which is the basic data for analyzing heart rate (H) and heart rate variability. Thereafter, the process branches into an analysis procedure for the heart rate (H) shown in the upper part of FIG. 3 and an analysis procedure for the heart rate variability shown below it, and these procedures are synchronized and analyzed simultaneously in parallel.

[0023] The analysis of the heart rate branched upward from the heart rate interval in FIG. 3 will be described. In time series data of electrocardiogram, various peaks and valleys appear on the electrocardiogram graph. They are named P wave, Q wave, R wave, S wave, T wave, U wave, etc., as they indicate the waveform components linked to the activity of the heart. For example, R wave is seen in the middle of QRS wave, which indicates ventricular excitation, and is considered to be an electrical signal generated when the heart pumps blood to the whole body due to the rapid contraction activity of the ventricles, and draws the sharpest peak in the electrocardiogram. In general, the time difference between the occurrence of an R wave and the occurrence of the R wave immediately after it is calculated to be the heart beat interval (RRI = RR Interval), and the heart rate per unit time (H) can be obtained from the heart beat interval.

[0024] Next, an analysis of heart rate variability, which branches downward from the heart rate interval in FIG. 3, will be described. The heart rate interval (RRI) is not always constant, but has a periodic variation called heart rate variation. Examples of such graphs are shown in Non-Patent Document 1 and Fig. 8 of Non-Patent Document 2. The heart rate interval (RRI) fluctuates up and down finely over time, and the heart rate variation is observed even in a calm and active state, while abnormalities occur in the heart rate variation when the balance of the autonomic nerves is lost due to excessive load on both the body and mind. According to Non-Patent Document 1, the relationship with the autonomic nerves has been pointed out as a factor of heart rate variation, and there are two factors: the contribution of a fluctuation component having the same period as respiration and the contribution of a fluctuation component having the same period as blood pressure variation. When the autonomic nerves are functioning normally, both the fluctuation synchronized with respiration and the fluctuation synchronized with blood pressure appear in the heart rate variation, and when there is an abnormality in the autonomic nerve function, the fluctuation synchronized with respiration is lost. By analyzing this heart rate variation, it is possible to grasp the tendency of both the body and mind that the worker W unconsciously accepts, and this system also focuses on the effective use of this heart rate variation.

[0025] Furthermore, the system analyzes high frequency components (HF components) and low frequency components (LF components) that contribute to the composition of heart rate fluctuations based on data on heart rate fluctuations. As described in Non-Patent Document 1, the system analyzes power spectrum density (PSD) to extract a periodic structure from time series data of heart rate fluctuations. The periodic structure of heart rate fluctuations can be grasped by the obtained power spectrum density (PSD), and two fluctuation components that are considered to be indicators of autonomic nerve function balance can be obtained by analyzing the power spectrum density (PSD) using an autoregressive model or the like. That is, the high frequency component (HF component) and the low frequency component (LF component), the high frequency component (HF component) corresponding to respiratory fluctuations appears in heart rate fluctuations only when the parasympathetic nerves are tense and dominant (relaxed). On the other hand, the low frequency component (LF component) corresponding to blood pressure fluctuations appears in heart rate fluctuations both when the sympathetic nerves are tense (stressed) and when the parasympathetic nerves are tense (relaxed). For this reason, it has been proposed to use the ratio (LF / HF) of high frequency components (HF components) to low frequency components (LF components) as the stress index (L).

[0026] The analysis of body temperature shown in the lower part of FIG. 3 will now be described. The sensor 1 measures the body surface temperature and heat flow of the worker W. The core body temperature (C) is analyzed as an estimated value after conversion to the body surface temperature using an arithmetic formula set according to the specifications of the sensor 1. This analysis process has already been put to practical use in the bioinformation sensor manufactured by Company A. In addition, regarding obtaining deep body temperature, it is possible to change the specifications of sensor 1 and obtain the value by making corrections to the values ​​measured directly from parts of the worker W's body, as long as it does not interfere with the work. Data acquisition using the bio-information sensor mentioned above has already been put to practical use for the respiratory rate (R). The respiratory rate (R) of the worker W can be estimated by obtaining high frequency components (HF components) corresponding to respiratory fluctuations from power spectral density (PSD) analysis to grasp the periodic structure of heart rate fluctuations. Then, the data relating to core body temperature (C), heart rate (H), respiratory rate (R) and stress index (L) obtained by acquiring and analyzing the biometric data using the sensor 1 are transmitted over time from the sensor 1 to the gateway terminal 2 via wireless communication.

[0027] The gateway terminal 2, which has received the biometric data transmitted from the sensor 1, converts the normal time series data into time series data of moving average values ​​before transmitting the data to the server 3. This is intended to rectify the biometric data flowing from the sensor 1 to the server 3, and to avoid the influence of singular values ​​and stabilize warning determination by averaging each piece of data based on the analysis of electrocardiogram, which tends to fluctuate relatively widely. In addition, it is also intended to reduce the load on the server 3, which accumulates data on a large scale.

[0028] The four types of biometric data received by the server 3, namely, core body temperature (C), heart rate (H), respiratory rate (R) and stress index (L), are used as basic data for monitoring the health condition of the worker W and are also used to determine whether or not to issue a warning regarding excessive workload. Based on FIG. 4, a workload determination program, which is a data processing procedure installed in the server 3 and determines whether or not a warning should be issued, that is, an overview of information processing including a warning score algorithm will be described. The app of this system is equipped with an item evaluation table as shown in Figure 5, in which comparison values ​​are set to determine the evaluation points for core body temperature (C), heart rate (H), stress index (L), and respiratory rate (R). In the workload judgment of this system, the server 3 determines the evaluation score for each of the four evaluation items, i.e., core body temperature (C), heart rate (H), respiratory rate (R), and stress index (L), based on the four types of biological data that are time-series converted into moving average values ​​and the item evaluation table shown in Fig. 5. Furthermore, the server 3 calculates a warning score by adding them up, and judges whether or not to issue a warning by comparing the warning score with two thresholds that are the criteria for determining whether or not to issue a warning. This procedure is the warning score algorithm shown in the center of Fig. 4.

[0029] With regard to this warning score algorithm, in the course of carrying out demonstration tests at actual work sites such as security guards, we have examined the multifaceted nature of the biometric data used as the evaluation items in the item evaluation table, and the change rate that serves as the standard for adding evaluation points for each evaluation item, that is, the setting of the comparison value in Figure 5. We have noticed that depending on these settings, a bias occurs in the issuance of warnings, and there are many inappropriate and overly sensitive warnings that deviate from the level of workload that is estimated to be actually being accepted based on interviews with the worker W. In consideration of such problems, four types of biometric data, i.e., deep body temperature (C), heart rate (H), respiratory rate (R) and stress index (L), were selected as evaluation items, and judgment was made based on the time series data of moving average values, rather than on the single data. In addition, for the three evaluation items of deep body temperature (C), heart rate (H) and stress index (L), fixed values ​​as a comparison standard were not set, and the biometric data should be evaluated relative to each other within a work time axis that is periodically repeated on a daily basis, for example. Therefore, for each of deep body temperature (C), heart rate (H) and stress index (L), the change rate obtained by comparing the moving average value at a specified time before and the current time is converted into an evaluation score based on the item evaluation table. In addition, for respiratory rate (R), the difference obtained by comparing the set reference value with the moving average value at a specified time before is converted into an evaluation score based on the item evaluation table.

[0030] In the app of this system, the item evaluation table referenced by the warning score algorithm shows an example of specific comparison criteria and evaluation points for each evaluation item in Figure 5. For deep body temperature (C), the evaluation point is determined based on the rate of change in the one-minute moving average value between 15 minutes ago and the present time. For stress index (L), the evaluation point is determined based on the rate of change in the one-minute moving average value between 15 minutes ago and the present time, just like for deep body temperature (C), and for heart rate (H), the evaluation point is determined based on the rate of change in the one-minute moving average value between three minutes ago and the present time. On the other hand, for respiratory rate (R), a preset number of times is used as the reference value, and the evaluation point is determined by comparing this with the one-minute moving average value three minutes ago. In the item evaluation table, the core body temperature (C) is weighted highly on a 5-point scale, the minimum value (15%) of the added score standard for the heart rate (H) is set lower than that (30%) of the stress index (L), and the relative evaluation of the stress index (L) follows the heart rate (H). The composition of the item evaluation table, which continues to the absolute evaluation of the respiratory rate (R), is intended to supplement the evaluation by the core body temperature (C) and the heart rate (H) with the two evaluation items of the stress index (L) and the respiratory rate (R). In other words, in order to evaluate the workload in detail, it is possible to realize a multifaceted and realistic evaluation by placing more importance on the fluctuations of the core body temperature (C) and the heart rate (H), which are directly linked to the physical effects, while also using the stress index (L) and the respiratory rate (R), which are likely to reflect mental factors as both mental and physical loads. In this case, the stress index (L) is relatively unstable and has a large fluctuation range, so the comparison value range (50%) when deciding the added points is also set to a larger value than that of the heart rate (H) (35%). The respiratory rate (R) has a wide range of 9 to 20 breaths per minute, which is generally accepted as a normal value, and shows a relatively stable fluctuation compared to other evaluation items, so it is an absolute evaluation compared with a value set as a reference value in advance. The reference value of the respiratory rate (R) in Figure 5 is set to 15 breaths per minute with reference to the value of the scoring system in the Rapid Response System (RSS) in Non-Patent Document 3, etc., but in this system, the comparison standard, comparison value, and threshold value set in the item evaluation table, including this, can be changed as necessary. In this way, the first half of the warning score algorithm shown in the center of Figure 4 functions to obtain the four evaluation scores that serve as basic data for monitoring the health condition of worker W and are also used to determine whether or not a warning should be issued.

[0031] Next, as a function of its latter half, the warning score algorithm calculates a warning score by adding up four evaluation points: core body temperature (C), heart rate (H), stress index (L), and respiratory rate (R). When the warning score, which varies from moment to moment, exceeds a first threshold, first warning data is sent to the gateway terminal 2 and the access terminal 5, and when it exceeds a second threshold, second warning data is sent. In the example of the item evaluation table in Figure 5, the maximum score for each evaluation item is 5 points for deep body temperature (C), which has a large weighting, and the remaining items, heart rate (H), stress index (L), and respiratory rate (R), each have a weighting of 3 points. Therefore, the warning score when all items are full marks is 14 points. Regarding the addition of the evaluation points based on the item evaluation table in FIG. 5, when the rate of change of the deep body temperature (C) reaches the comparison value of 0.5%, which is the standard for adding the evaluation points, 1 point is added. Furthermore, at 0.8%, the evaluation point is 2 points, at 1.0%, 3 points, at 1.3%, 4 points, and at a rate of change of 1.5% or more, the evaluation point is 5 points. Similarly, when the rate of change of the heart rate (H) reaches the comparison value of 15%, the evaluation point is 1 point, at 25%, 2 points, and at 35%, 3 points. The stress index (L) is 1 point at 30%, 2 points at 40%, and 3 points at 50%. The respiratory rate (R) is 1 point when it reaches the comparison value of 17.6 times, which is the standard value of 15 times plus 2.6 times, 2 points at 18.7 times, and 3 points at 20 times. When the warning score, calculated by tallying up these four evaluation points, reaches the first threshold of 9 points, a caution advisory (YA) is issued, and when it reaches the second threshold of 11 points, a risk announcement (RA) is issued.

[0032] In order to make more detailed judgments and prevent the manifestation of risks, an emergency notification function may be added to the workload judgment system in Fig. 4 to supplement the warning score algorithm. When the rate of change of the moving average value of deep body temperature (C) and heart rate (H) is confirmed to have the same tendency of either increase or decrease a predetermined number of times in succession, regardless of whether the worker W is aware of it or not, a unidirectional and rapid increase in risk is presumed, and a point is added to each evaluation point. When this is significant and reaches a predefined high consecutive number of times, an emergency notification is executed to send a danger notice (RA), which is a second warning, to the gateway terminal 2 and the access terminal 5 regardless of the warning score. As a specific example, when the same tendency of either increase or decrease in deep body temperature (C) or heart rate (H) occurs two consecutive times, two points are added to the evaluation point, three points are added for three consecutive times, and four points are added for four consecutive times, and when these occur five consecutive times, a danger notice (RA) is immediately issued. The function of monitoring the number of consecutive times is equivalent to the function of monitoring the continuous time, and the same applies to the exercise intensity algorithm described below and later. Furthermore, one of the activity data acquired by the sensor 1 is posture data indicating whether the worker W is in a standing or lying state, etc. This posture data suggests that the worker W is continuously lying down, and if this suggests that the fallen state will continue for more than the specified time of 5 minutes, for example, an emergency notification is issued in which 5 points are added to the warning score. In addition, with regard to the respiratory rate (R), which often shows a stable tendency, when the 1-minute moving average value exceeds a specified number of times, for example 25 times, it is assumed that there is an accumulation of stress on both the mind and body, and therefore an emergency notification is issued in which 3 points are added to the evaluation score.

[0033] In addition, the comparison values ​​for each evaluation item that are the conditions for adding evaluation points in this item evaluation table are variable and are determined based on a data collection table such as that shown in Figure 6, taking into account the safety margin that should be ensured. However, if necessary, the settings of these comparison values ​​can be changed as appropriate. FIG. 6, which shows an example of a data accumulation table, is a graph showing the fluctuation of the deep body temperature (C) of multiple workers W under severe environmental conditions where the heat index (WBGT value) is 28°C or higher, which increases the risk of developing heat stroke. The periodic time axis from the top right to the top left, going clockwise, is a periodic time axis from the start to the end of work, and lines showing the transition of the moving average value of the deep body temperature (C) of multiple workers W are plotted on top of each other. Even under the same conditions, the fluctuation of the deep body temperature (C) varies due to differences in the work content and individual differences, but this allows us to grasp the overall tendency regarding the transition of the deep body temperature (C) and obtain useful information when considering the safety margin in the workload judgment based on the item evaluation table. In this system, for each of the evaluation items of deep body temperature (C), heart rate (H), stress index (L), and respiration rate (R), a data accumulation table corresponding to multiple different work environments, such as a heat index (WBGT value) below or above 25°C, or below or above 28°C as shown in FIG. 6, is prepared. Also, multiple data accumulation tables are prepared corresponding to work contents with different exercise intensities, such as 0.8 (80%) and 1.2 (120%). Furthermore, a data accumulation table for each individual may be prepared that links and accumulates and manages the fluctuation status of each individual's biological data, warning data, and report data sent to the manager M via the server 3 when the worker W becomes aware of an abnormality in physical condition or when a warning is issued. Depending on the specifications of the app, multiple data accumulation tables are incorporated to perform a multifaceted analysis of the correlation between the fluctuation status of the four types of biological data that are evaluation items of the workload and the overload risk that is centrally aggregated as a warning score, and to determine more appropriate points addition conditions and to set and modify them. Depending on the specifications of this system, the administrator M can select a safety margin presented from the data accumulation table of the server 3, thereby allowing the system to arbitrarily set and change the rate of change in the biometric data that is the condition for adding the evaluation points, that is, the comparison value of the item evaluation table. For example, for a worker W who is engaged in hard work with high exercise intensity under dangerous working conditions with a high heat index (WBGT value), by selecting the optimal data accumulation table from a predetermined setting screen, it is possible to switch to an appropriate item evaluation table linked to it. As another specification, it is also possible to perform machine learning based on the large-scale data accumulated in the server 3 regarding the correlation between the fluctuation of the biometric data and the evaluation points and warning points that are risk-evaluated based on the warning score algorithm and the data accumulation table, and to automate the setting of the comparison value that is the condition for adding the points.

[0034] FIG. 4 also illustrates a mechanism for correcting the warning score algorithm to improve its accuracy. In order to issue a warning with a more sufficient safety margin or to suppress issuance of a warning that is deemed excessive compared to the work content, the system is equipped with a correction function for workload judgment based on the exercise intensity during work as necessary. This calculates the exercise intensity of the work from the maximum heart rate previously set according to the age of the worker W and the heart rate (H) of the worker W measured by the sensor 1, and multiplies the evaluation score by a predetermined correction coefficient exceeding 1 when the rate of change of the moving average value of the exercise intensity per predetermined unit time increases for a predetermined number of consecutive times and the rate of change of the moving average value of the respiratory rate (R) decreases for the same predetermined number of consecutive times. This exercise intensity correction function is for ensuring the urgency and necessity of issuing a warning, and is referred to as the exercise intensity algorithm in the system. That is, in a situation where the exercise intensity during work increases, the respiration rate (R) of the worker W generally tends to increase with the exercise intensity. Conversely, in a state where the exercise intensity increases while the respiration rate (R) continues to decrease, there is a strong possibility that the body of the worker W is not adapted to the increased load due to the exercise intensity, and it is meaningful to multiply the evaluation score by a correction coefficient exceeding 1 to preliminarily issue an early warning and strengthen the safety protection of the worker W. In addition, when the moving average value of the respiration rate (R) increases with the increase in exercise intensity and reaches the comparison value in the item evaluation table, the evaluation score is added by the normal warning score algorithm. On the other hand, when the rate of change of the moving average value of the respiration rate (R) changes gently in response to the rate of change of the exercise intensity, it is possible to slow down the rate of addition by multiplying the evaluation score by a coefficient less than 1 as the data fluctuation within the expected range, and to avoid frequent issuance of overly sensitive warnings that deviate from the actual workload. In other specifications, when both exercise intensity and respiration rate (R) are on the rise, and activity data indicating that the worker W is lying down or stationary is continuously confirmed at the same time, the evaluation score is multiplied by a correction coefficient exceeding 1. Furthermore, when exercise intensity is on the decline and core body temperature (C) or respiration rate (R) is on the rise, an exercise intensity correction function is added that multiplies the evaluation score by a correction coefficient exceeding 1.

[0035] For this reason, the workload judgment program used in the health management system of the present invention is basically as follows. First, there is a first process of obtaining time series data from four types of biological data related to the deep body temperature (C), heart rate (H), stress index (L), and respiration rate (R) of the worker W who is the sensor wearer. Next, there is a second process of obtaining a moving average value per a predetermined unit time from the time series data. The third process has the following two procedures that are executed in synchronization. One is a procedure of converting the rate of change obtained by comparing the moving average value of the deep body temperature (C), heart rate (H), and stress index (L) with the moving average value at the present time before a predetermined time into an evaluation score based on the item evaluation table. The other is a procedure of converting the difference number obtained by comparing the moving average value of the respiration rate (R) with the set reference value before a predetermined time into an evaluation score based on the item evaluation table. Furthermore, there is a fourth process of issuing a first warning when the warning score obtained by summing up the four evaluation scores related to the deep body temperature (C), heart rate (H), stress index (L), and respiration rate (R) obtained in the third process exceeds a first threshold value, and a second warning when it exceeds a second threshold value. Depending on the specifications, the workload assessment program may also include an emergency notification process as follows: When the rate of change of the moving average values ​​of deep body temperature (C) and heart rate (H) is continuously measured as data showing the same tendency of either increase or decrease, each evaluation point is added. When this further exceeds a predetermined number of times, a second warning is issued in the fifth process. There is also a sixth process of obtaining posture data of the sensor wearer, and adding a warning point when posture data showing a recumbent state exceeds a predetermined time. And there is a seventh process of adding a point to the evaluation point when the moving average value of the respiratory rate (R) exceeds a predetermined number of times. These processes are synchronized with the third and fourth processes and are executed simultaneously in parallel. Furthermore, this workload assessment program is equipped with an exercise intensity correction function that includes an eighth process that analyzes exercise intensity from the maximum heart rate corresponding to the age of the sensor wearer and the measured heart rate (R), and multiplies the evaluation score by a predetermined coefficient exceeding 1 when the rate of change in the moving average of the exercise intensity has an increasing tendency for a predetermined number of consecutive times and the rate of change in the moving average of the respiratory rate (R) has a decreasing tendency for the same predetermined number of consecutive times.

[0036] Finally, we will use the security industry, where we have conducted demonstration tests, as an example to explain the operation and functions of the app and the specific usage of this system. However, this is only one example, and the system is not limited to this. Hereinafter, the management manager will be referred to as Manager M, and the security guard wearing Sensor 1 will be referred to as Worker W. The manager of the management office supervises many security guards and performs various duties such as labor management and health and safety management. There is daily communication between the two parties, such as instructions and reports, but the work sites are located far from the management office, and security guards do not always stop by the management office before starting work, but often go directly from their homes to the sites and then return directly from the sites to their homes. This system was introduced to support the proper execution of management duties by the management director and to optimize health management of security guards. The management director can access the system via the Internet and obtain various information provided at the management office. Security guards can also make various reports on-site without having to stop by the management office.

[0037] The management manager will perform initial system settings as necessary to optimize the system for the operations of his / her company. The management manager will launch a browser on the access terminal 5 connected to the Internet, display the login screen of the system, and log in by entering his / her ID and password. After logging in, a home screen like that shown in Figure 7 will be displayed, where the status of "Work Information," "Warnings" (shown as "Alerts" in Figure 7), and "Emergency Calls" can be confirmed. The top of the home screen has a home button (shown as "Coat" in Figure 7) and several pull-down menus, among which the master maintenance menu allows you to select the required management menu and perform initial system settings. The management manager registers a user in the user management menu of FIG. 8, and can set the identification code (ID), password, name, date of birth, contact information such as telephone number and email address, user authority, and measurement parameters of the sensor 1. The management manager further sets the identification code and address of the work site in the work place management menu of FIG. 9, and sets the work content and its identification code, such as clerical work, environment management, and traffic control security, in the work content management menu of FIG. 10. These work contents may also be associated with a data accumulation table and exercise intensity algorithm. In the measurement parameter management menu of FIG. 11, for example, correction coefficients corresponding to the body type of the security guard can be set to improve the accuracy of biometric data acquisition by the sensor 1. In the attendance setting menu of FIG. 12, documents such as a work schedule, daily work hours and break times, etc. can be set. In the email setting menu of FIG. 13, contact information such as telephone numbers of the management manager and security guards, and email addresses for sending warning data from the server 3 to both terminals can be set. This completes the initial settings, but these settings can be checked in a list, and the settings can also be changed or deleted.

[0038] When the guard arrives at the scene, he or she wears the sensor 1 on the chest by attaching it to the attachment before starting work. The guard taps the icon of the gateway terminal 2 to start the app, and enters the ID and password on the login screen to complete the login to the system. The gateway terminal 2 displays a work start screen as shown in Figure 14, so the guard enters the work data and report data. The work data here includes the "start time" of the work, as well as the name of the city, town, village, or site where the work will be performed, which are selected from the areas labeled "work location" and "work content." These are registered in the initial settings mentioned earlier, and the guard selects the appropriate one from among them. The report data is selected by tapping the one that corresponds to the guard's physical condition at the start of work from the five icons to the right of the area labeled "physical condition." The icons in Figure 14 are displayed in five levels, from the left, as "very good, good, normal, not very good, and poor." These initial operations of the gateway terminal 2 automatically pair the sensor 1 with the gateway terminal 2, enabling data communication. At the same time, in addition to the biometric data acquired by the sensor 1 and continuously transmitted, the gateway terminal 2 also transmits to the server 3 work data indicating the start time, work location, and work content of the work, and report data on physical condition.

[0039] The security guard performs tasks such as traffic control and patrol security while wearing the sensor 1. During breaks, the guard can temporarily disable data communication with the system by operating the gateway terminal 2. When resuming work after an interruption, the security guard re-pairs the sensor 1 with the gateway terminal 2 and inputs the work data for resuming work. Generally, biometric data immediately after starting work is often unstable with a large fluctuation range. However, the guard's physical and mental condition begins to settle down shortly after starting work, and the values ​​of the biometric data sent from sensor 1 also stabilize. The results of demonstration experiments have shown that this occurs approximately 10 to 20 minutes after the start of work. In addition, a data collection table that collects a large amount of data by individual or by work environment or content makes it possible to estimate trends in biometric data under multiple different conditions and over a repeated time axis, so the accuracy of evaluating newly acquired biometric data also improves over time.

[0040] The biological data of the security guard acquired by the sensor 1 is transmitted to the gateway terminal 2 by wireless communication. The biological data at this time includes body surface temperature, deep body temperature (C), electrocardiogram, heart rate (H), stress index (L), respiratory rate (R), etc. In addition to the biological data, environmental data such as environmental temperature and relative humidity, and activity data indicating the number of steps and movement, etc. are also acquired by the sensor 1 and transmitted to the gateway terminal 2. The security guard working can check these on the work screen of the gateway terminal 2. In the center of the screen in Fig. 15, the warning score transmitted from the server 3 and the corresponding workload judgment result ("Safe", "Caution", "Warning") are displayed. In this example, the warning score is 6 points, so the mark to the left of it is "Safe". Below that, the latest heart rate (H) and deep body temperature (C) transmitted from the sensor 1 are displayed. When the underlined "Graph display" section on the heart rate (H) or core body temperature (C) display is tapped, it switches to the linked screen shown in Figure 16, which displays a graph of the changes in the vital signs. Note that the top of the screen in Figure 15 also allows you to check the communication status with Sensor 1 and the remaining battery level. The "Measuring" display on the top left shows the communication status with Sensor 1, and the battery icon on the top right shows the remaining battery level of Sensor 1. When a warning is issued or a security guard feels unwell, he or she can tap "Health Report" at the bottom right of the screen, select the appropriate report from the displayed report menu ("Request for medical assistance" (x), "Suspend work" (△), "Continue work" (◯)) that best suits their condition, and send the report data for the report or request to the management director via server 3. This report data is managed by linking it to the data collection table on server 3, for example by quantifying a request for medical assistance as "3" and continuing work as "1".

[0041] The data acquired by the sensor 1 and sent to the gateway terminal 2, and the data input from the gateway terminal 2, are sent from the gateway terminal 2 to the server 3, which is often constantly running, and analysis processing such as workload judgment using a warning score algorithm is performed. At this time, information on the guard's whereabouts from the acceleration sensor of the sensor 1 and the position information of the work site from the GPS of the gateway terminal 2 are also sent. In addition, by receiving the working data, it is possible to immediately confirm the arrival of the guard at the site and the start of work. The management manager can access the system at any time, just as he did at the time of the initial setup. This can be done each time the manager starts daily work, or the system can be connected all the time. Immediately after accessing the system, the manager checks the access terminal 5 to see if there are any abnormalities in the system's operation status, such as data acquisition and communication, and in the security guard's biometric data. If no abnormalities are found, the manager can check the access terminal 5 as necessary thereafter, and there is no need to constantly monitor it.

[0042] After the management manager logs in, the home screen of FIG. 7 is displayed on the screen of the access terminal 5. At the top of the home screen, from the left, the menu display of the home button, work viewer, work management, and master maintenance can be confirmed. The pull-down menu of the work viewer allows the user to check the contents of warnings sent from the server 3 and emergency calls, which are one of the report data sent from the gateway terminal 2. These can also be checked from the links in the "Warning" and "Emergency Call" panels displayed at the bottom of the home screen. The "Work Information" panel displays the number of security guards working, and both the "Warning" and "Emergency Call" panels display the number of cases and the number of cases that have not been confirmed by the management director. At this time, the number of unconfirmed cases is displayed by flashing, so that the management director can recognize the issuance of a warning without delay. When the management director accesses the alarm confirmation screen, the operator's name, contact information, warning type, and confirmer are displayed in a list by time of occurrence, as shown in Figure 17. Similarly, the emergency call confirmation screen displays the sender, contact information, content, and confirmer by time of transmission in a list.

[0043] Clicking directly on the "Work Viewer" menu at the top of the home screen, or on the "Go to Work Viewer" link in the "Work Information" panel, will display the Work Viewer, which allows users to instantly grasp the work status of multiple guards on a map, as shown in Figure 18. The number of unconfirmed emergency calls and warnings is displayed in the upper left corner of the screen, allowing the management office manager to quickly recognize any abnormalities with the guards. Below that, the guards are displayed in a list on the left side of the screen, and a worker tab for each individual guard is displayed on the map on the right. The worker list on the left shows the worker's name, deep body temperature (C), heart rate (H), start and end of work, etc., and when you click the worker tab displayed to the right of the worker list, the map display on the right side of the screen changes to a map display centered on the work site where the guard is working. When you click the sensor tab displayed further to the right of the worker tab, the sensor information screen, which shows a list of the guard's biometric data, is displayed as shown in Figure 19. This screen displays data such as the work date, worker's name, contact information, work location, work time, work content, and physical condition. In addition, the biometric data acquired by sensor 1, such as the guard's body temperature data (surface body temperature and deep body temperature (C)), environmental data (ambient temperature, relative humidity), stress index (L), respiratory rate (R), and heart rate (H), can be confirmed in graph and list format. In the sensor information screen in Fig. 19, the graphs below the body temperature data are omitted, but the graphs of vital signs data such as heart rate (H) follow. Below these multiple graphs is a list display screen as shown in Fig. 20, which shows a chronological list of vital signs data, which are also one-minute moving average values. The information displayed in Fig. 20 is the time of data transmission, warning score, exercise intensity, living and working information that suggests a fall, deep body temperature (C), heart rate (H), stress index (L), respiratory rate (R), number of steps, surface temperature, ambient temperature, relative humidity, and a breakdown of the evaluation score (C / H / L / R).

[0044] When a warning is issued from the server 3, the management site manager can immediately know that a warning has been issued by the flashing of the screen or a warning sound on the home screen of the access terminal 5 that received the warning. At this time, the security guard at the work site also receives the warning from the server 3 at the gateway terminal 2 at almost the same time, and quickly recognizes the issuance of the warning by a warning sound or the like. At this time, the management site manager and the security guard will check and contact each other, and the security guard can send report data regarding the response to the warning to the management site manager via the server 3 by tapping the "Health Report" banner at the bottom right of the screen of the gateway terminal 2. The report can be made by selecting the appropriate one from the report menu display ("Request for rescue" (x), "Suspend work" (△), "Continue work" (◯)), and requests and reports can be made quickly by just touching the screen, and responses such as stopping or temporarily suspending work can be discussed. In addition, when a warning is issued, a pop-up display that prompts the security guard to check and respond can be displayed on the screen of the gateway terminal 2, and the warning sound can be stopped by the security guard responding to it. On the other hand, when the day's security work is completed without any work stoppage due to the issuance of such a warning, the security guard taps the work completion banner at the bottom left of the screen. After that, work data is sent to the server 3 to communicate the time of work completion, and this is recorded in the roster, allowing the management office manager to confirm that the security guard's work has been completed. The security guard removes the sensor 1, turns off the power to the gateway terminal 2, and leaves the work site. On the other hand, when the management office manager performs the logout process after completing work, he can select logout from the pull-down menu to the right of the user status display (user: system administrator) displayed at the right end of the home screen in Fig. 7.

[0045] As explained above, by using this system, it is possible to quickly and effectively reduce health risks such as heat stroke in harsh work environments, and to realize a better occupational safety and health environment. In addition, by managing the work status and health status of security guards together, it is possible to reduce the burden of cumbersome administrative procedures such as reports and paperwork between the management director and security guards. The system manages a variety of data, such as security guards' biometric data, report data, activity data, and employment data, as well as environmental data from the workplace, and report data when warnings are issued or when employees feel unwell. Figure 21 shows the management screen that is displayed when the Work Management menu is clicked from the home screen. The operator code, operator name, age, and measurement parameters of the selected security guard are displayed on the left side of the screen. From the "Select Operator" pull-down display below, other security guards to be checked can be selected. On the right side, the work location, start time, end time, work confirmation, construction number, etc. of the security guard are displayed by date. Furthermore, by clicking the sensor information tab on the right, the sensor information of the security guard for that day is displayed as shown in Figure 19 and Figure 21, so the management manager can check the necessary information as appropriate. In this management screen, the employment management and health management of security guards are integrated in one place, and by clicking the roster tab, it is also possible to output the forms required for labor management and occupational safety and health management. An example of the form format output in this way is the roster in Figure 22. In addition, by performing machine learning based on the large amount of data accumulated in the data collection table of server 3, the accuracy of the system's workload judgment function for newly acquired biometric data will improve over time. [Explanation of symbols]

[0046] 1 Sensor 2. Gateway terminal 3 Server 4. Information Visualization System 5 Access terminal M Administrator W Worker

Claims

1. A health management system in which a sensor, a gateway terminal, and a server cooperate with each other through application software, The sensor acquires biological data relating to the sensor wearer's core body temperature (C), heart rate (H), stress index (L) and respiratory rate (R) over time and transmits the biological data to the gateway terminal; the gateway terminal transmits to the server a moving average value per predetermined unit time obtained by analyzing the time series data of the biometric data and working data inputted by the sensor wearer from the gateway terminal; The server manages the moving average value and the working data by linking them with an identification code (ID) of the sensor wearer, and provides the moving average value and the working data in a web page that can be accessed by a system user using an access terminal, The application software has an item evaluation table that defines comparison values ​​for determining evaluation points for deep body temperature (C), heart rate (H), stress index (L) and respiratory rate (R), and a workload determination function that converts the rate of change obtained by comparing the moving average value of deep body temperature (C), heart rate (H) and stress index (L) from a predetermined time ago to the current time based on the comparison value, and converts the difference number obtained by comparing the moving average value of respiratory rate (R) from a predetermined time ago to a set reference value based on the comparison value. The health management system is characterized in that the application software has a workload determination function that transmits first warning data to the access terminal and the gateway terminal linked to the identification code (ID) when the warning score obtained by summing up the evaluation points for deep body temperature (C), heart rate (H), stress index (L) and respiratory rate (R) exceeds a first threshold value, and transmits second warning data to the access terminal and the gateway terminal linked to the identification code (ID) when the warning score exceeds a second threshold value.

2. The health management system according to claim 1, characterized in that the application software includes a data accumulation table for accumulating the moving average value and determining the comparison value, and the comparison value determined in the item evaluation table is changeable.

3. The health management system of claim 1 or 2, characterized in that the application software is provided with an emergency notification function that adds to the evaluation score when the rate of change comparing the moving average values ​​of deep body temperature (C) and heart rate (H) continues to show the same trend of either increasing or decreasing, and when this continues for more than a predetermined number of times, sends the second warning data to the access terminal and the gateway terminal linked to the identification code (ID), adds to the warning score when posture data obtained from the sensor wearer indicates a recumbent state for more than a predetermined time, and adds to the evaluation score when the moving average value of the respiratory rate (R) exceeds a predetermined number of times.

4. The health management system according to claim 1 or 2, characterized in that the application software has an exercise intensity correction function that obtains exercise intensity from a maximum heart rate set corresponding to the age of the sensor wearer and the heart rate (H), and multiplies the evaluation score by a predetermined coefficient exceeding 1 when the rate of change of the moving average value of the exercise intensity per predetermined unit time has an increasing tendency for a predetermined number of consecutive times and the rate of change of the moving average value of the respiratory rate (R) has a decreasing tendency for the same predetermined number of consecutive times.

5. A health management system in which a sensor, a gateway terminal, a server, and an information visualization system cooperate with each other through application software, The sensor acquires biological data relating to the sensor wearer's core body temperature (C), heart rate (H), stress index (L) and respiratory rate (R) over time and transmits the biological data to the gateway terminal; the gateway terminal transmits to the server a moving average value per predetermined unit time obtained by analyzing the time series data of the biometric data and working data inputted by the sensor wearer from the gateway terminal; The server manages the moving average value and the working data by linking them with an identification code (ID) of the sensor wearer, The information visualization system provides the moving average value and the working data for each sensor wearer in a map-displayed work viewer on a web page that can be accessed by a system user using an access terminal, The application software has an item evaluation table that defines comparison values ​​for determining evaluation points for deep body temperature (C), heart rate (H), stress index (L) and respiratory rate (R), and a workload determination function that converts the rate of change obtained by comparing the moving average value of deep body temperature (C), heart rate (H) and stress index (L) from a predetermined time ago to the current time based on the comparison value, and converts the difference number obtained by comparing the moving average value of respiratory rate (R) from a predetermined time ago to a set reference value based on the comparison value. The health management system is characterized in that the application software has a workload determination function that transmits first warning data to the access terminal and the gateway terminal linked to the identification code (ID) when the warning score obtained by summing up the evaluation points for deep body temperature (C), heart rate (H), stress index (L) and respiratory rate (R) exceeds a first threshold value, and transmits second warning data to the access terminal and the gateway terminal linked to the identification code (ID) when the warning score exceeds a second threshold value.

6. The health management system according to claim 5, characterized in that the application software includes a data accumulation table for accumulating the moving average value and determining the comparison value, and the comparison value determined in the item evaluation table is changeable.

7. The health management system according to claim 5 or 6, characterized in that the application software is provided with an emergency notification function that adds to the evaluation score when the rate of change comparing the moving average values ​​of deep body temperature (C) and heart rate (H) continues to show the same trend of either increasing or decreasing, and when this continues for more than a predetermined number of times, transmits the second warning data to the access terminal and the gateway terminal linked to the identification code (ID), adds to the warning score when posture data obtained from the sensor wearer indicates a recumbent state for more than a predetermined time, and adds to the evaluation score when the moving average value of respiratory rate (R) exceeds a predetermined number of times.

8. The health management system according to claim 5 or 6, characterized in that the application software has an exercise intensity correction function that obtains exercise intensity from a maximum heart rate set corresponding to an age of the sensor wearer and the heart rate (H), and multiplies the evaluation score by a predetermined coefficient exceeding 1 when the rate of change of the moving average value of the exercise intensity per predetermined unit time has an increasing tendency for a predetermined number of consecutive times and the rate of change of the moving average value of the respiratory rate (R) has a decreasing tendency for the same predetermined number of consecutive times.

9. a first step of acquiring biological data relating to the deep body temperature (C), heart rate (H), stress index (L) and respiratory rate (R) of a sensor wearer over time to obtain time series data; a second step of analyzing the time series data to obtain a moving average value per predetermined unit time; a third step of converting a rate of change obtained by comparing the moving average value of the deep body temperature (C), heart rate (H) and stress index (L) between a predetermined time ago and the present time based on the comparison value, and converting a difference number obtained by comparing the moving average value of the respiratory rate (R) between the predetermined time ago and a set reference value based on the comparison value, thereby obtaining the evaluation points for the deep body temperature (C), heart rate (H), stress index (L) and respiratory rate (R); and a fourth step of issuing a first warning when a warning score obtained by summing up the evaluation points exceeds a first threshold value, and issuing a second warning when the warning score exceeds a second threshold value.

10. A workload judgment program for a health management system as described in claim 9, characterized in that it comprises a fifth step of adding points to the evaluation score when the rate of change comparing the moving average values ​​of deep body temperature (C) and heart rate (H) continues to show the same trend of either increasing or decreasing, and issuing the second warning when this continues for more than a predetermined number of times; a sixth step of adding points to the warning score when posture data obtained from the sensor wearer indicates a recumbent state for more than a predetermined time; and a seventh step of adding points to the evaluation score when the moving average value of respiratory rate (R) exceeds a predetermined number of times.

11. 11. The workload judgment program for a health management system according to claim 9, further comprising an eighth step of performing an exercise intensity correction in which exercise intensity is obtained from a maximum heart rate set corresponding to an age of the sensor wearer and the heart rate (H), and when the rate of change of the moving average of the exercise intensity per predetermined unit time has an increasing tendency for a predetermined number of consecutive times and the rate of change of the moving average of the respiratory rate (R) has a decreasing tendency for the same predetermined number of consecutive times, the evaluation point is multiplied by a predetermined coefficient exceeding 1.

Images