KR102565175B1 - Ward alarm and monitoring method for unexpected accident using heat sensor - Google Patents

Abstract

The present invention is configured to simultaneously determine the abnormal body temperature of a patient in each bedroom and whether or not a fall accident occurs using a heat-detection sensor, thereby not only improving the problem of invasion of personal privacy by not using a separate camera and thermal imaging camera. In addition, installation cost is reduced, low-power operation is possible, and after the local management server analyzes the heat-sensing information received from the data collection devices to recognize the object, the temperature values and location information of the recognized object are It is configured to determine the patient's abnormal body temperature and whether a fall accident has occurred, so that the calculation process is simple and the accuracy of object recognition can be increased. It transmits unexpected situation data including the contents of the emergency situation, ward and room identification numbers, and when the management program of the manager terminal receives the unexpected situation data from the local management server, by outputting warning alarm information to the outside, Remote monitoring is possible, improving work efficiency and maximizing the efficiency of the nursing management system. In the event of an unexpected situation, rapid follow-up is possible, which can increase the safety and reliability of medical services. It's about how.

Description

Ward alarm and monitoring method for unexpected accident using heat sensor}

The present invention relates to a method for alarming and monitoring an emergency situation in a ward using a heat-sensing sensor, and more particularly, detects and monitors the body temperature and fall of each patient in each ward using the heat-sensing sensor and simultaneously detects and monitors an emergency When a situation is detected, it is output to the outside through the output means, so that the patient's condition can be remotely monitored to improve management efficiency and convenience, as well as to reduce unnecessary manpower consumption, and to quickly respond to unexpected situations. It relates to a method for alarming and monitoring an unexpected situation in a ward using a heat-sensing sensor capable of increasing service safety and reliability and enabling low-cost installation and operation.

In general, a hospital is a place for providing lodging and treatment to patients for various treatment purposes, and each ward is equipped with one to a plurality of bedrooms.

Since these hospitals prioritize the safety of patients, they periodically examine and check the patient's condition, but various unexpected situations such as sudden abnormal patient conditions and falls in the bedroom are occurring invariably.

In particular, a fall is an accident in which a patient suddenly falls to a low position regardless of his or her intention. According to a domestic study, 70.8% of falls occur in hospitals, and most often occur between 22:00 and 06:00. reported

Accordingly, various studies are being conducted on a patient monitoring system for remotely monitoring a patient's condition such as a fall and at the same time outputting an unexpected situation to the outside so that a prompt response can be made.

1 is a block diagram showing a patient fall prevention monitoring device disclosed in Korean Patent Registration No. 10-1712191 (Title of Invention: Patient Fall Prevention Monitoring Device).

The patient fall prevention monitoring device (hereinafter referred to as the prior art) 100 of FIG. 1 uses a Raspberry Pi 111 based on an embedded board and an optical sensor 112 having a camera to The monitoring unit 101 photographs the uncomfortable patient's bed and processes the signal, and the data processed by the monitoring unit 101 is analyzed by the OpenCV program 122 to infer the movement of the patient and the server (123) and stored in the database 124, the image processing unit 102 corresponding to the client, and black pixels and white pixels according to the patient sleeping in the right posture, sleeping on the side, and sitting postures inferred by the image processing unit 102 The posture determination unit 103 that determines the patient's bed posture by calculating the number of pixels on the screen for each pixel, and the data processed by the image processing unit 102 are transferred to the management center 105 using TCP/IP communication or web sockets. When a picture processed by the embedded board 121 of the image processing unit 102 is transmitted as html, the video communication unit 104 transmits to , and the management center 105 receives, monitors, and manages the picture in the PC.

In addition, the difference image analysis performed by the OpenCV program 122 analyzes the difference images constructed by the contrast difference between the adjacent (i-1) th image and the i th image to infer the movement form of the moving object, and The method of inferring the movement type for is, if there is a background screen taken in advance and a screen to be detected, the image is converted to gray scale, and the converted gray scale screen is converted to black and white through binarization. A screen is created, morphology image processing is performed to reduce the noise of the binarized screen, and moving average image processing is performed so that even if the background image changes, a non-moving object can be applied to the background image.

Unlike the conventional contact method or count method, the prior art 100 configured as described above can significantly reduce malfunction or delay in information transmission because it rapidly and accurately monitors the patient's fall and departure through non-contact real-time image analysis. Since information on this uncomfortable inpatient can be monitored in real time by hospital personnel, it has the advantage of preventing accidents caused by patients falling or leaving the bed in advance and performing efficient nursing management tasks.

However, the prior art 100 analyzes the image obtained by the photographing of the optical sensor (camera) 112, in detail, generates a difference image, and then displays the black pixels and white pixels of the generated difference image on the screen. Since it is configured to determine the patient's bed posture by calculating the number of pixels, real-time image analysis for each of a plurality of bedrooms results in an excessively complex amount of computational processing and a problem in that computational processing time is consumed.

In addition, the prior art 100 has a disadvantage in that the installation cost is high because the expensive camera 112 must be installed at the top of each bedroom.

In addition, the prior art 100 can monitor and detect whether or not a fall has occurred, but has a structural limitation in that it cannot monitor and detect sudden situations such as abnormal body temperature.

In addition, the prior art 100 has a disadvantage in that the patient's personal privacy problem is caused because the patient's face and life are captured by the camera 112 as it is, regardless of the patient's intention.

In order to overcome this problem, Korean Patent Registration No. 10-2052883 (Title of Invention: Fall Prediction System Using a Thermal Imaging Camera and Fall Prediction Method Using a Thermal Imaging Camera) discloses a step of photographing a person with a thermal imaging camera. ; Analyzing each pixel data constituting the captured image, obtaining x-axis histogram and y-axis histogram data of the captured image based on the pixel data, and based on the x-axis histogram and y-axis histogram data , determining whether the person is in a stable state or a fall risk state, and operating a notification device based on the person's condition, thereby protecting the patient's privacy and predicting the patient's fall. A fall prediction method using a thermal imaging camera has been disclosed, but the installation cost is very high because an expensive and high-powered thermal imaging camera must be installed to photograph each bedroom. Since the thermal image obtained by the thermal image is analyzed in real time to determine the state of a person, the calculation processing amount is high, and when there are a large number of objects to be monitored, a problem occurs that the calculation processing speed is delayed.

The present invention is intended to solve this problem, and an object of the present invention is configured to simultaneously determine the abnormal body temperature and the occurrence of a fall accident of a patient in each bedroom using a heat-sensing sensor, thereby enabling a separate camera and a thermal imaging camera. This is to provide a method for alarming and monitoring unexpected situations in a ward, which can not only improve personal privacy issues, but also reduce installation costs, and enable low-power operation.

In addition, another problem of the present invention is that the local management server analyzes the heat-sensing information received from the data collection devices to recognize the object, and then uses the temperature values and location information of the recognized object to determine the patient's ideal body temperature and It is configured to determine whether a fall accident has occurred, thereby providing a ward emergency situation warning and monitoring method that can increase the accuracy of object recognition while simplifying calculation processing.

In addition, another problem of the present invention is that when the local management server determines that the patient's abnormal body temperature and a fall accident have occurred, it transmits the unexpected situation data including the contents of the unexpected situation and the identification number of the ward and room to the administrator's terminal. When the terminal's management program receives emergency situation data from the local management server, it outputs warning alarm information to the outside, enabling remote monitoring of the patient's condition, improving work efficiency and maximizing the efficiency of the nursing management system. In the event of an unexpected situation, prompt follow-up is possible to provide a ward emergency warning and monitoring method that can increase the safety and reliability of medical services.

The solution of the present invention for solving the above problems is installed on the upper part of each bedroom of at least one ward, and measures the temperature by subdividing the preset detection area (S) into a plurality of cells, and the temperature value for each cell is heat- In the ward emergency alerting and monitoring method (S1) for monitoring whether an emergency occurs in each ward using heat-detecting sensors that measure detection information: The ward emergency alerting and monitoring method (S1) is A local management server including a database unit in which a reference table, which is data matched with optimal constants (TH) for each range of temperature values of , is preset and stored receives heat-sensing information from the heat-sensing sensors, and the ward Step 10 (S10) of receiving a ward temperature value from a ward thermometer installed in the ward to measure the temperature value of the ward; After the local management server analyzes the heat-sensing information received through step 10 (S10) and extracts the temperature value for each cell, the extracted temperature value for each cell is compared with the preset body temperature range to be included in the body temperature range and continuously Step 20 (S20) of recognizing the cells as the patient object and detecting the cell information of the patient object; After the local management server extracts the temperature value for each cell extracted in step 20 (S20), it calculates the patient body temperature, which is the average value of the temperature values for each cell of the extracted patient object, and puts the calculated patient body temperature within the predetermined normal body temperature range. 1) determining that the patient's body temperature is normal when the patient's temperature value is within the normal body temperature range, and 2) determining that an abnormal body temperature has occurred in the patient when the patient's body temperature value is out of the normal body temperature range 30 (S30); Step 50, when the local management server determines that an unexpected situation has occurred when the patient's body temperature is out of the normal body temperature range in step 30 (S30), and generates unexpected situation data including the details of the unexpected situation and ward and room identification information ( S50); Step 60 (S60) of the local management server transmitting the emergency situation data generated in step 50 (S50) to manager terminals; It proceeds at a predetermined period (T), and when the local management server receives the temperature value of each ward transmitted through the step 10 (S10), the standard stored in the database unit After extracting the table, searching the extracted reference table, extracting optimal constants (TH) matching the input temperature values of each ward, and then storing the extracted optimal constants (TH) of each ward in the database unit Step 80 (S80) is included, and in step 20 (S20), when the local management server receives heat-sensing information through step 10 (S10), the temperature value for each cell is extracted from the received heat-sensing information. Then, the optimal constant (TH) stored in the database unit is multiplied by the extracted temperature value for each cell to calculate the optimal temperature value for each cell, and the calculated optimal temperature value for each cell is compared with the preset body temperature range and included in the body temperature range while recognizing successive cells as patient objects, the optimal constant (TH) stored in the reference table has a smaller value as the temperature value of the corresponding ward increases, and a higher value as the temperature value of the ward decreases, The step 20 (S20) includes a step 21 (S21) of the local management server receiving the heat-sensing information received through the step 10 (S10); After the local management server extracts the temperature value for each cell from the heat-sensing information input through the step 21 (S21), the optimal constant (TH) stored in the database unit is multiplied by the extracted temperature value for each cell to obtain the temperature for each cell. After calculating the optimal value, the calculated optimal temperature value for each cell is compared with the preset body temperature range, and consecutive cells included in the body temperature range are recognized as patient objects, and cell information of the patient object is detected Step 22 (S22) ; Step 23 (S23) of the local management server detecting the cell-location information of the patient object recognized by the step 22 (S22) and then storing the detected cell-location information of the patient object in the database unit; Step 24 (S24) includes: step 241 (S241) of receiving, by the local management server, the cell-location information of the patient object detected by step 23 (S23); Step 242 (S242) of extracting, by the local management server, cell-location information of the corresponding patient object stored in the database unit, data from a current point in time to a preset previous time (t); Step 243 (S243) of determining the longitudinal direction of the patient by the local management server analyzing cell-location information of the patient object input through step 241 (S241); Step 245 (S245) of determining, by the local management server, areas adjacent to both ends in the longitudinal direction of the patient object determined in step 243 (S243) as candidate areas; step 246 (S246) of calculating, by the local management server, the size of the width (cell size) of each candidate area determined in step 245 (S245); The local management server compares the width of each candidate region calculated in step 246 (S246) with a preset second reference value, which is the maximum value of the width for determining that the recognized candidate region is a facial region. , 1) When one of the candidate regions has a width less than the second reference value and the other has a width greater than or equal to the second reference value, a candidate region having a width less than the second reference value is determined as a facial region; , 2) Step 247 (S247) of repeating the process after going to step 21 (S21) when the widths of all candidate regions are equal to or greater than the second reference value; In step 247 (S247), the process is performed when the widths of all candidate regions are less than the second reference value, and the local management server uses the data extracted in step 242 (S242) to determine the size of the previous point in time based on the current point in time. direction, the cell-location information of the patient object is analyzed, each candidate region is detected, and then the width of each candidate region is detected, and the width of each candidate region is detected using the width and size values of each candidate region. Step 248 of calculating a size change rate (%) (S248); Step 249 (S249) of the local management server comparing the change rate (%) of the width size of each candidate region calculated in step 248 (S248) and determining that the candidate region having a low change rate (%) is the facial region. ; Step 250 (S250) of the local management server, when the facial region is determined in step 247 (S247) or step 249 (S249), finally determining the determined facial region as the facial region of the corresponding patient; The local management server includes a step 251 (S251) of determining the average value of the optimum temperature values of the cells corresponding to the facial region determined in the step 250 (S250) as the patient's body temperature, and the step 30 (S30) includes the The local management server uses the patient's body temperature determined in step 251 (S251) to determine whether or not the body temperature is abnormal.

delete

delete

In addition, in the present invention, the step 24 (S24) further includes a step 244 (S244) performed after the step 243 (S243), and the step 244 (S244) is performed by the local management server to provide cell-location information of the patient object. And, after calculating the length (L, cell-size) of the patient object by utilizing the longitudinal direction of the patient object determined in step 243 (S243), the length (L) of the calculated patient object is used as the recognized object It is compared with a preset reference value, which is the minimum value of the length of an object that can be determined as a body. 1) If the length (L) of the patient object is less than the reference value, it is determined that there was an error in determining the length direction of the patient, 2) The length of the patient object If (L) is equal to or greater than the reference value, determining that there is no error in determining the longitudinal direction of the patient (S2441); The local management server compares whether there are many Y-axis components or X-axis components in the longitudinal direction of the bed in the longitudinal direction of the patient object determined in step 243 (S243), 1) if the length of the patient object If there are many X-axis components in the direction, it is determined that an error has occurred in determining the longitudinal direction, and 2) if it is determined that there are many Y-axis components in the longitudinal direction of the patient object, step 2442 of determining that there is no error in determining the longitudinal direction (S2442 ); 1) If the local management server determines that no error has occurred in both steps 2441 (S2441) and 2442 (S2442), it verifies that there is no error in determining the longitudinal direction of the patient object in step 243 (S243), 2) When it is determined that an error has occurred in at least one of the steps 2441 (S2441) and the step 2442 (S2442), step 2443 of finally determining that an error has occurred in determining the longitudinal direction of the patient object in step 243 (S243) (S2443), and the step 245 (S245) proceeds to the next step when the local management server verifies that there is no error in determining the longitudinal direction of the patient object in the step 244 (S244) and the step 243 (S243). Step 21 (S21) proceeds to the next step when the local management server determines that an error has occurred in determining the longitudinal direction of the patient object by the longitudinal direction determining module in step 243 (S243) in step 244 (S244). It is desirable to proceed

In addition, in the present invention, bedroom area information, which is location information of cells corresponding to bedrooms in the sensing area S of each of the heat-detecting sensors, is preset and stored in the database unit, and the ward emergency situation alarming and monitoring method ( S1) further includes a step 40 (S40) performed after the step 30 (S30), and the step 40 (S40) includes the local management server receiving the bedroom area information of the corresponding heat-sensing sensor stored in the database unit. After extraction, the cell location of the patient object and the cell location of the bedroom are compared using the cell information of the patient object detected in step 20 (S20) and the extracted bedroom area information, and the cell location of the patient object is determined by the bedroom area information. When the elapsed time away from the cell location is greater than or equal to a threshold value, it is determined that a fall accident has occurred in the patient, and in step 50 (S50), the local management server determines that a fall accident has occurred in the patient in step 40 (S40). At this time, it is desirable to generate emergency situation data.

According to the present invention having the above problems and solutions, it is configured to simultaneously determine the abnormal body temperature of a patient in each bedroom and whether or not a fall accident occurs using a heat-sensing sensor, so that a separate camera and thermal imaging camera are not used, and personal privacy Not only can the infringement problem be improved, but the installation cost is reduced and low-power operation is possible.

In addition, according to the present invention, after the local management server recognizes an object by analyzing the heat-sensing information received from the data collection devices, the patient's abnormal body temperature and fall accident occur using the temperature values and location information of the recognized object. By being configured to determine whether or not, the calculation process is simple and the accuracy of object recognition can be increased.

In addition, according to the present invention, when the local management server determines that the patient's abnormal body temperature and a fall accident have occurred, it transmits the unexpected situation data including the contents of the unexpected situation and the ward and room identification numbers to the administrator terminal, and the management program of the administrator terminal When emergency situation data is received from this local management server, it is possible to remotely monitor the patient's condition by outputting warning alarm information to the outside, thereby improving work efficiency and maximizing the efficiency of the nursing management system at the same time. In the event of an occurrence, prompt follow-up is possible, increasing the safety and reliability of medical services.

1 is a block diagram showing a patient fall prevention monitoring device disclosed in Korean Patent Registration No. 10-1712191 (Title of Invention: Patient Fall Prevention Monitoring Device).
2 is a block diagram showing a system for warning and monitoring an emergency situation in a ward, which is an embodiment of the present invention.
3 is an exemplary view of FIG. 2 .
FIG. 4 is a plan view illustrating the data collection device and heat-detecting sensors of FIG. 2 .
FIG. 5 is a side view illustrating the heat-sensing sensor of FIG. 4 .
6 is an exemplary view of a GUI displayed by the management program of FIG. 2 .
FIG. 7 is a block diagram illustrating a local management server of FIG. 2 .
8 is a block diagram illustrating a patient object recognition unit of FIG. 7 .
FIG. 9 is a block diagram illustrating a facial area detection module of FIG. 8 .
10 is a block diagram illustrating a verification module of FIG. 9 .
11 is a block diagram illustrating a fall determination unit of FIG. 7 .
12 is a block diagram illustrating an optimum constant setting unit of FIG. 7 .
13 is a flow chart showing an operation process of an emergency situation warning and monitoring system in a hospital ward according to an embodiment of the present invention.
FIG. 14 is a flow chart showing the patient object recognition step of FIG. 13 .
FIG. 15 is a flowchart showing the facial area detection step of FIG. 14 .
FIG. 16 is a flow chart showing the verification step of FIG. 15 .

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

FIG. 2 is a configuration diagram showing an emergency situation warning and monitoring system in a hospital ward according to an embodiment of the present invention, and FIG. 3 is an exemplary view of FIG. 2 .

A ward emergency alarm and monitoring system 1, which is an embodiment of the present invention, detects and monitors the body temperature and fall of each patient in each ward using a heat-sensing sensor 9, and at the same time detects an unexpected situation, By outputting this to the outside through the output means, the patient's condition can be remotely monitored to increase management efficiency and convenience, as well as to reduce unnecessary manpower consumption, and to respond quickly to unexpected situations, improving the safety and reliability of medical services. can be increased, and low-cost installation and operation are possible.

In addition, as shown in FIGS. 2 and 3, the ward emergency situation alarm and monitoring system 1 is installed on the ceiling directly above each of the bedrooms of each ward and measures the temperature of the detection area S directly below the heat- Sensors (9), and data collection devices (4) installed in each ward to collect the temperature values measured by the heat-detection sensors (9) of the corresponding ward, and from the data collection devices (4) A local management server (3) that analyzes the received collected information (temperature value information) to determine the patient's fever status and fall status in each bedroom, and manager terminals (5), which are terminals used or possessed by a manager (nurse) And, it is an application installed in the manager terminals (5), receives the fever status and fall status of each patient transmitted from the local management server (3), visualizes them, displays them on the monitor of the manager terminal (5), and displays them locally When the unexpected situation occurrence data is received from the management server 3, the management program 7 that displays warning alarm information through the manager terminal 5, the local management server 3, the data collection devices 4, and the manager It consists of a communication network 10 providing a data movement path between the terminals 5.

At this time, in the present invention, the data collection devices 4 simply transmit the heat-sensing information measured by the heat-detecting sensors 9 of the corresponding ward 200 to the local management server 3, and the local management server 3 ) analyzes the heat-sensing information received from the data collection devices 4 and performs calculation processing to determine the patient's abnormal body temperature and whether or not to fall. The arithmetic processing for determining can be configured to be performed directly in the data collection device 4.

The communication network 10 supports data communication between the local management server 3, data collection devices 4 and manager terminals 5, and in detail, a wired communication network, a wide area communication network (WAN), a mobile communication network, etc. may consist of

FIG. 4 is a plan view illustrating the data collection device and heat-detecting sensors of FIG. 2 , and FIG. 5 is an exemplary side view illustrating the heat-detecting sensor of FIG. 4 .

As shown in FIGS. 4 and 5, the heat-detection sensors 9 are installed on the ceiling 220 directly above each bedroom 210 disposed in the corresponding ward 200, and the lower detection area S Measure the surface temperature. That is, the heat-detection sensors 9 measure the surface temperature of the preset detection area S.

In addition, the heat-detecting sensors 9 can measure the sensing area S by subdividing it into a plurality of cells S'. At this time, the temperature value for each cell S' generated by the heat-detecting sensor 9 is referred to as heat-sensing information.

In addition, it is preset that the sensing area S is subdivided into a plurality of cells by the thermal sensors 9 .

In addition, the heat-sensing sensors 9 transmit the generated heat-sensing information to the data collection device 4 of the corresponding ward 200 in real time.

The ward thermometer 11 measures the temperature of the corresponding ward 200 and transmits the measured temperature value of the ward to the data collection device 4 of the corresponding ward 200 at predetermined intervals T.

The data collection device 4 is installed in each ward 200 to collect heat-sensing information from the heat-sensing sensors 9 of the corresponding ward 200 and at the same time, ward thermometers 11 at predetermined intervals T. ), collects ward temperature values, and transmits collected data consisting of the collected heat-sensing information or collected data matching the collected heat-sensing information and ward temperature values to the local management server 3.

At this time, the collected data generated by the data collection device 4 is composed of identification information of the corresponding ward, identification information of each heat-sensing sensor, heat-sensing information, and ward temperature value.

Manager terminals 5 are digital devices possessed by managers such as nurses or installed at positions such as desks where managers are placed, and in detail, may be composed of desktop-PCs, laptops, tablet PCs, smartphones, etc. .

In addition, a management program 7 is installed in the manager terminals 5 .

6 is an exemplary view of a GUI displayed by the management program of FIG. 2 .

The management program 7 is software, application, or application program installed in the manager terminal 5 and providing an emergency alarm and monitoring service in a ward in conjunction with the local management server 3.

In addition, the management program 7 visualizes the unexpected situation data received from the local management server 3 through prepared GUIs (Graphic User Interfaces) and displays them on the monitor of the manager terminal 5. At this time, the unexpected situation data consists of the contents of the unexpected situation (abnormal body temperature or fall), identification information of the ward where the unexpected situation occurred, and identification information of the bedroom where the unexpected situation occurred.

In addition, as shown in FIG. 6, the GUI 910 provided by the management program 7 is arranged in the ward display area 9111 displayed corresponding to the location of the actual ward 200 and the corresponding ward 200. State display units 911 in which bedroom display boxes 9112 corresponding to each of the bedrooms 210 are graphically displayed and displayed, and unexpected situation display boxes 912 displaying identification numbers of wards and bedrooms in which an unexpected situation has occurred made up of

In addition, the GUI 910 displays 1) bedroom display boxes 9112 in which abnormal body temperature and unexpected falls do not occur in the first color, and 2) true room display boxes 9112 in which abnormal body temperature occurs in a second color. 3) By displaying the bedroom indicator box (9112) where a fall accident occurred in a third color, the administrator can intuitively recognize whether an unexpected situation has occurred and the location of the ward and bedroom where the unexpected situation occurred quickly and accurately. This enables rapid response to emergencies.

FIG. 7 is a block diagram illustrating a local management server of FIG. 2 .

As shown in FIG. 7, the local management server 3 includes a control unit 31, a database unit 32, a data transmission/reception unit 33, a patient object recognition unit 34, and an abnormal body temperature determination unit 35. , a fall determination unit 36, an unexpected situation data generation unit 37, and an optimal constant setting unit 38.

The control unit 31 is the O.S (Operating System) of the local management server 3, and manages the control objects 32, 33, 34, 35, 36, 37, and 38. and control.

In addition, when the controller 31 receives collected data from the data collection devices 4 through the data transceiver 33, the controller 31 controls the data transceiver 33 to transmit the received collected data to the manager terminal 5. While controlling the data transmission/reception unit 33, the received collected data is stored in the database unit 32, heat-sensing information of the collected data is input to the patient object recognition unit 34, and the ward temperature value is included in the collected data. If is included, the ward temperature value is input to the optimal constant setting unit 38.

In addition, the control unit 31 generates unexpected situation data when it is determined that an abnormal body temperature has occurred in the patient by the abnormal body temperature determination unit 35 or when it is determined that a fall accident has occurred in the patient by the fall determination unit 36 Section 37 is executed.

In addition, when the unexpected situation data is generated by the unexpected situation data generator 37, the control unit 31 controls the data transmission/reception unit 33 to transmit the generated unexpected situation data to the manager terminals 5.

In the database unit 32, the location and identification information of each ward 200 and the location and identification information of the bedroom 210, which is the detection area S of each of the heat-detecting sensors 9 of each ward 200, are stored. It is set and saved.

In addition, in the database unit 32, the detection area S of each heat-detecting sensor 9 and the location information of the cells subdivided in the corresponding sensing area S are matched with the sensing area information and the bedroom in the sensing area S. Bedroom area information, which is location information of cells corresponding to 210, is preset and stored.

Also, collected data transmitted from the data collection devices 4 are stored in the database unit 32 .

In addition, the database unit 32 stores the unexpected situation data generated by the unexpected situation data generating unit 37 .

In addition, the optimal constant (TH, Threshold) set by the optimal constant setting unit 38 is stored in the database unit 32 .

In addition, the database unit 32 stores a reference table, which is data in which optimal constants (TH) are matched for each range of ward temperature values. At this time, the optimal constant (TH) for each range of ward temperature values in the reference table can be determined through a number of repeated experiments, and the optimal constant (TH) decreases as the ward temperature value increases and increases as the ward temperature value decreases. have

For example, when the ward temperature value is high, the temperature value of each cell detected by the thermal sensor 9 increases as a whole, making it difficult to accurately measure the patient's body temperature.

The data transmission/reception unit 33 transmits/receives data with the data collection devices 4 and the manager terminals 5.

The program management unit 34 manages and controls updates and GUI changes of the management program 7 .

8 is a block diagram illustrating a patient object recognition unit of FIG. 7 .

As shown in FIG. 8, the patient object recognition unit 34 includes a heat-sensing information input module 341, an object recognition module 343, an object location information extraction/storage module 345, a facial region detection module ( 347).

The heat-sensing information input module 341 receives the heat-sensing information transmitted from the data collection device 4 under the control of the control unit 31 .

The object recognition module 343 first extracts the temperature value for each cell from the input heat-sensing information, and then sets the optimum constant TH by the optimum constant setting unit 38 of FIG. 10 and stores it in the database unit 32. , Threshold) is multiplied by the temperature value for each cell to calculate the optimal temperature value for each cell.

In general, since the heat-sensing sensor 9 is installed on the ceiling of a ward and senses heat at a predetermined distance from an object, it is affected by the temperature of the ward. For example, when the temperature of the corresponding ward is low, the temperature value of the cells in the detection area S measured by the heat-detection sensor 9 is lowered as a whole, and when the temperature of the corresponding ward is high, heat-sensing The temperature values of the cells measured by the sensor 9 increase as a whole.

Accordingly, according to the present invention, the optimal constant setting unit 38 of FIG. 10, which will be described later, periodically sets an optimal constant (TH) capable of correcting the error of the temperature value for each cell in consideration of the temperature value of the ward, and the object recognition module ( 343) can increase the accuracy and reliability of object recognition by correcting the error of the temperature value of each cell due to the temperature of the ward by applying the set optimal constant (TH) when recognizing the patient object using the temperature value of each cell.

In addition, the object recognition module 343 compares the calculated optimal temperature value for each cell with a preset body temperature range, recognizes consecutive cells included in the body temperature range as a patient object, and detects cell information of the patient object.

The object location information extraction/storage module 345 detects the cell-location information of the patient object recognized by the object recognition module 343 and stores it in the database unit 32 .

FIG. 9 is a block diagram illustrating a facial area detection module of FIG. 8 .

As shown in FIG. 9, the facial region detection module 347 includes an object location information input module 3471, an accumulated data extraction module 3472, a longitudinal direction determination module 3473, a verification module 3474, and a candidate region. field determination module 3475, width calculation module 3476 of each candidate region, comparison and determination module 3477, each candidate region tracking module 3478 based on accumulated data, width variation comparison and determination module 3479, face It consists of an area determination module 3480 and a patient body temperature determination module 3481.

The object location information input module 3471 receives the cell-location information of the patient object detected by the object location information extraction/storage module 345.

The cumulative data extraction module 3472 extracts cell-location information of a corresponding patient object stored in the database unit 32 .

At this time, the accumulated data extraction module 3472 extracts data from the current point in time to a preset previous time (t).

The longitudinal direction determining module 3473 analyzes the cell-location information of the patient object input through the object location information input module 3471 to determine the longitudinal direction of the patient.

In general, since the length of a human body is significantly larger than the width, the longitudinal direction determination module 3473 can determine the width direction and length direction of the patient object by analyzing the cell-location information of the patient object.

10 is a block diagram illustrating a verification module of FIG. 9 .

The verification module 3474 of FIG. 10 verifies whether or not there is an error in the longitudinal direction of the patient determined by the longitudinal direction determination module 3473.

Also, as shown in FIG. 10 , the verification module 3474 includes a first verification module 34741 , a second verification module 34742 , and an error determination module 34743 .

The first verification module 34741 calculates the length (L, cell-size) of the patient object by utilizing the cell-location information of the patient object and the longitudinal direction of the patient object determined by the longitudinal direction determination module 3473. Afterwards, the calculated length (L) of the patient object is compared with a preset reference value.

At this time, the reference value is defined as the minimum value of the length of an object that can be determined that the recognized object is a body.

In addition, the first verification module 34741 determines that 1) if the length (L) of the patient object is less than the reference value, there is an error in determining the longitudinal direction of the patient, and 2) if the length (L) of the patient object is greater than the reference value, the patient It is judged that there was no error in determining the longitudinal direction of

When the longitudinal direction of the bed is assumed to be the Y-axis, the second verification module 34742 determines whether the longitudinal direction of the patient object determined by the longitudinal direction determination module 3473 has many Y-axis components or X-axis components (in the Y-axis). close or close to the X-axis).

In addition, the second verification module 34742 determines that 1) if there are many X-axis components in the longitudinal direction of the patient object, an error has occurred in determining the longitudinal direction, and 2) if there are many Y-axis components in the longitudinal direction of the patient object. If it is determined, it is determined that there is no error in the determination of the longitudinal direction.

The error occurrence determination module 34743 1) determines the longitudinal direction of the patient object in the longitudinal direction determining module 3473 when it is determined that no errors have occurred in both the first verification module 34741 and the second verification module 34742. 2) If it is determined that an error has occurred in at least one of the first verification module 34741 and the second verification module 34742, the longitudinal direction determination module 3473 determines the longitudinal direction of the patient object. It is finally determined that an error has occurred in

At this time, the control unit 30 executes the candidate region determination module 3475 when 1) it is finally verified in the verification module 3474 that no error has occurred, and 2) when it is finally determined that an error has occurred, The heat detection information input module 341, the object recognition module 343, and the object location information extraction/storage module 345 are re-executed.

The candidate region determination module 3475 is executed when the verification module 3474 verifies that there is no error in determining the longitudinal direction of the patient object by the longitudinal direction determination module 3473.

Also, the candidate regions determination module 3475 determines regions adjacent to both ends of the patient object in the longitudinal direction as candidate regions.

For example, since the regions adjacent to both ends of the human body in the height direction consist of the outside of the face and the legs, the candidate regions determined by the candidate regions determination module 3475 consist of the face region and the leg region.

The width calculation module 3476 of each candidate region calculates the size of the width (cell size) of each candidate region determined by the candidate regions determination module 3475.

The comparison and determination module 3477 compares the size of the width of each candidate region calculated by the width calculation module 3476 of each candidate region with a preset second reference value. At this time, the second reference value is defined as the maximum value of the width at which the recognized candidate region can be determined as the facial region.

Further, the comparison and determination module 3477 may perform a candidate having a width less than the second reference value when the width of one of the candidate regions is less than the second reference value and the width of the other candidate region is greater than or equal to the second reference value. The area is determined as a facial area.

In addition, the comparison and determination module 3477 executes each candidate region tracking module 3478 based on accumulated data when the widths of all candidate regions are smaller than the second reference value.

In addition, the comparison and determination module 3477 extracts/stores the heat detection information input module 341, object recognition module 343, and object location information of FIG. Rerun module 345.

For example, when the patient is lying down, the patient's leg area is generally covered with a blanket, and accordingly, when the patient is in a state where the blanket is covered with both legs spread, the width of the leg area is the second Although detection is greater than the reference value, a human facial region has a characteristic of not being detected as exceeding a predetermined size. Therefore, in the present invention, the comparison and determination module 3477 is used to detect the facial region among candidate regions.

As another example, when the patient's legs are overlapped, since the width of the leg region is detected to be less than the second reference value, after the comparison and determination module 3477, each candidate region tracking module 3478 based on accumulated data is executed. .

Each candidate region tracking module 3478 based on accumulated data is executed when the width of all candidate regions in the comparison and determination module 3477 is less than the second reference value.

In addition, each candidate region tracking module 3478 based on accumulated data utilizes the data extracted by the cumulative data extraction module 3472 to analyze the cell-location information of the patient object in the direction of the previous point of view based on the current point of view. , After detecting each candidate region, the size of the width of each candidate region is detected.

In addition, the cumulative data-based tracking module 3478 of each candidate region utilizes the detected width values of each candidate region to calculate a width change rate (%) of each candidate region.

The width variation comparison and determination module 3479 compares the width variation rate (%) of each candidate area calculated by each candidate area tracking module based on the cumulative data, and selects a candidate area with a low width size variation rate (%) as a facial area. decide that

For example, the face region has a small change rate (%) of width size even when the posture is changed, but the leg region has a high change rate (%) compared to the face region even when the posture is not changed.

When the facial region is determined in 1) the comparison and determination module 3477 or 2) the width variation comparison and determination module 3479, the facial region determination module 3480 finally determines the determined facial region as the facial region of the corresponding patient. .

The patient body temperature determination module 3481 determines the average value of the optimum temperature values of the cells corresponding to the facial region determined by the facial region determination module 3480 as the patient's body temperature.

The patient object recognition unit 34 configured as described above inputs the patient's body temperature information recognized by the object recognition module 343 to the abnormal body temperature determination unit 35, and the cell information of the patient object to the fall determination unit ( 36).

The abnormal body temperature determining unit 35 compares the patient's body temperature detected by the patient's body temperature determination module 3481 with a predetermined normal body temperature range, and 1) determines that the patient's body temperature is normal when the patient's body temperature is within the normal body temperature range. 2) When the patient's body temperature is out of the normal body temperature range, it is determined that abnormal body temperature has occurred in the patient.

At this time, the control unit 31 executes the sudden situation data generation unit 37 when it is determined by the abnormal body temperature determination unit 35 that abnormal body temperature has occurred in the patient.

11 is a block diagram illustrating a fall determination unit of FIG. 7 .

As shown in FIG. 11, the fall determination unit 36 includes a data input module 361, a bedroom area information extraction module 362, a location comparison module 363, and a fall determination module 364.

The data input module 361 receives cell information of the patient object from the patient object recognition unit 34 and receives heat-sensing information of a corresponding heat-sensing sensor.

The bedroom area information extraction module 362 searches the database unit 32 and extracts the bedroom area information of the corresponding thermal sensor 9 stored in the database unit 32 .

At this time, the bedroom area information is location information of cells corresponding to the bedroom within the corresponding detection area S, and is preset and stored in the database unit 32.

The location comparison module 363 utilizes the patient object cell information input through the data input module 361 and the bedroom area information extracted by the bedroom area information extraction module 362 to determine the location of the cell of the patient object and the bedroom. Compare cell positions.

The fall determination module 364 determines that a fall accident has occurred in the patient when the elapsed time in which the cell position of the patient object deviates from the cell position of the bedroom in the position comparison module 363 is greater than or equal to a threshold value.

Returning to FIG. 7 and examining the unexpected situation data generator 37, the abnormal situation data generator 37 determines the abnormal body temperature of the patient by the abnormal body temperature determination unit 35 or falls It is executed when the patient's fall is determined by and generates unexpected situation data. At this time, the unexpected situation data consists of the contents of the unexpected situation (abnormal body temperature, fall accident), the identification number of the corresponding ward, and the identification number of the corresponding bedroom.

In addition, the unexpected situation data generated by the unexpected situation data generator 37 is transmitted to the manager terminals 5 through the data transmission/reception unit 33 under the control of the control unit 31, and management of the manager terminals 5 When the program 7 receives emergency situation data from the local management server 3, it outputs sound warning information to the outside so that the administrator can quickly deal with it.

12 is a block diagram illustrating an optimum constant setting unit of FIG. 7 .

The optimum constant setting unit 38 of FIG. 12 is executed according to the control of the control unit 31 at each preset period T.

As shown in FIG. 12, the optimal constant setting unit 38 includes a ward temperature value input module 381, a reference table extraction module 382, and an optimal constant determining module 383.

The ward temperature value input module 381 receives the temperature value of each ward transmitted from the data collection device 4 .

The reference table extraction module 382 extracts the reference table stored in the database unit 32 .

The optimal constant setting module 383 searches the reference table extracted by the reference table extraction module 382, and selects optimal constants TH matched with each ward temperature value input by the ward temperature value input module 381. extract each.

In addition, the optimal constant setting module 383 stores the extracted optimal constants (TH) of each ward in the database unit 32 .

13 is a flow chart showing an operation process of an emergency situation warning and monitoring system in a hospital ward according to an embodiment of the present invention.

The emergency alerting and monitoring method (S1) of a hospital ward in FIG. 13 shows the operation process of the emergency alerting and monitoring system 1 of a ward, which is an embodiment of the present invention.

In addition, as shown in FIG. 13, the ward emergency situation warning and monitoring method (S1) includes the collection data reception step (S10), patient object recognition step (S20), abnormal body temperature determination step (S30), and fall determination step. It consists of (S40), unexpected situation data generation step (S50), data transmission step (S60), monitoring step (S70), and optimal constant setting step (S80).

The collected data receiving step (S10) is a step in which the local management server 3 receives the heat-sensing information measured by the heat-detecting sensors 9 and the ward temperature value measured by the ward thermometer 11. am.

In the patient object recognition step (S20), the local management server 3 extracts the temperature value for each cell from the heat-sensing information received through the collected data receiving step (S10), and then performs the optimal constant setting step (S80) to be described later. The optimal constant (TH) set and stored in the database unit 32 is multiplied by the temperature value for each cell to calculate the optimal temperature value for each cell, and the calculated optimal temperature value for each section is compared with the preset body temperature range to be included in the body temperature range This is a step of recognizing successive cells as patient objects and detecting cell information of the patient objects.

FIG. 14 is a flow chart showing the patient object recognition step of FIG. 13 .

As shown in FIG. 14, the patient object recognition step (S20) includes a heat-sensing information input step (S21), an object recognition step (S22), an object location information extraction/storage step (S23), a facial region detection step ( S24).

The heat-sensing information input step (S21) is a step in which the local management server 3 receives the heat-sensing information received through the collected data receiving step (S10).

In the object recognition step (S22), after the local management server (3) extracts the temperature value for each cell from the inputted heat-sensing information, it is set by the optimal constant setting step (S80) to be described later and stored in the database unit (32). This step is to calculate the optimum temperature value for each cell by multiplying the constant (TH) with the temperature value for each cell.

In addition, the object recognition step (S22) compares the calculated optimal temperature value for each cell with a preset body temperature range, recognizes consecutive cells included in the body temperature range as a patient object, and detects cell information of the patient object.

The object location information extraction/storage step (S23) is a step in which the local management server 3 detects the cell-location information of the patient object recognized by the object recognition step (S22) and stores it in the database unit 32. .

FIG. 15 is a flowchart showing the facial area detection step of FIG. 14 .

As shown in FIG. 15, the face region detection step (S24) includes the object location information input step (S241), the accumulated data extraction step (S242), the longitudinal direction determination step (S243), the verification step (S244), the candidate region Determining step (S245), calculating the width of each candidate region (S246), comparing and judging step (S247), tracking each candidate region based on accumulated data (S248), comparing and determining whether the width is changed (S249), face It consists of a region determination step (S250) and a patient's body temperature calculation step (S251).

The object location information input step (S241) is a step in which the local management server 3 receives the cell-location information of the patient object detected by the object location information extraction/storage step (S23).

In the cumulative data extraction step (S242), the local management server 3 extracts the cell-location information of the corresponding patient object stored in the database unit 32, but extracts data from the current point in time to a preset previous time (t). It is a step.

The longitudinal direction determination step (S243) is a step in which the local management server 3 analyzes the cell-location information of the patient object input through the object location information input step (S241) to determine the longitudinal direction of the patient.

In general, since the length of a human body is significantly larger than the width, the longitudinal direction determination module 3473 can determine the width direction and length direction of the patient object by analyzing the cell-location information of the patient object.

FIG. 16 is a flow chart showing the verification step of FIG. 15 .

The verification step (S244) of FIG. 16 is a step of verifying whether or not there is an error in the longitudinal direction of the patient determined by the longitudinal direction determining step (S243).

As shown in FIG. 16, the verification step (S244) includes a first verification step (S2441), a second verification step (S2442), and an error occurrence determination step (S2443).

The first verification step (S2441) is the length (L, After calculating the cell-size), this is a step of comparing the calculated length (L) of the patient object with a preset reference value.

At this time, the reference value is defined as the minimum value of the length of an object that can be determined that the recognized object is a body.

In addition, in the first verification step (S2441), 1) if the length (L) of the patient object is less than the reference value, it is determined that there is an error in determining the longitudinal direction of the patient, and 2) if the length (L) of the patient object is greater than the reference value, the patient It is judged that there was no error in determining the longitudinal direction of

In the second verification step (S2442), when the local management server 3 assumes that the longitudinal direction of the bed is the Y-axis, whether there are many Y-axis components in the longitudinal direction of the patient object determined by the longitudinal direction determining step (S243) or the X-axis This step compares whether there are many components (closer to the Y-axis or closer to the X-axis).

In addition, in the second verification step (S2442), 1) if there are many X-axis components in the longitudinal direction of the patient object, it is determined that an error has occurred in determining the longitudinal direction, and 2) if there are many Y-axis components in the longitudinal direction of the patient object. If it is determined, it is determined that there is no error in the determination of the longitudinal direction.

In the error occurrence determination step (S2443), if the local management server 3 determines that no error has occurred in both the first verification step (S2441) and the second verification step (S2442), the longitudinal direction determination step (S243) It is verified that there is no error in determining the longitudinal direction of the patient object, and 2) if it is determined that an error has occurred in at least one of the first verification step (S2441) and the second verification step (S2442), the longitudinal direction determination step (S243) This is the step of finally deciding that an error has occurred in determining the length direction of the patient object.

At this time, in the step of determining whether an error has occurred (S2443), 1) if it is finally verified that no error has occurred, the step of determining candidate regions (S246) proceeds as the next step, and 2) if it is finally determined that an error has occurred, the aforementioned FIG. 14 Return to the heat detection information input step (S21) of and repeat the process thereafter.

The step of determining the candidate regions (S245) proceeds when it is verified in the verification step (S244) that there is no error in determining the lengthwise direction of the patient object in the lengthwise direction determination step (S243), and the local management server 3 determines the length of the patient object. This step is to determine regions adjacent to both ends of the direction as candidate regions.

For example, since the regions adjacent to both ends of the human body in the height direction consist of the outside of the face and the legs, the candidate regions determined by the candidate regions determination module 3475 consist of the face region and the leg region.

The step of calculating the width of each candidate region (S246) is a step in which the local management server 3 calculates the size of the width (cell size) of each candidate region determined in the step of determining the candidate regions (S245).

The comparison and determination step (S247) is a step in which the local management server 3 compares the size of the width of each candidate region calculated in the step of calculating the width of each candidate region (S246) with a preset second reference value. At this time, the second reference value is defined as the maximum value of the width at which the recognized candidate region can be determined as the facial region.

In addition, in the comparison and determination step S247, when the width of one of the candidate regions is less than the second reference value and the width of the other candidate region is greater than or equal to the second reference value, the candidate region has a width less than the second reference value. The area is determined as a facial area.

Also, in the comparison and determination step (S247), when the widths of all of the candidate regions are less than the second reference value, each candidate region tracking step (S248) based on accumulated data proceeds to the next step.

Also, in the comparison and determination step S247, when the widths of all candidate regions are equal to or greater than the second reference value, the above-described heat detection information input step S21 of FIG. 14 is returned to repeat the subsequent process.

For example, when the patient is lying down, the patient's leg area is generally covered with a blanket, and accordingly, when the patient is in a state where the blanket is covered with both legs spread, the width of the leg area is the second Although detection is greater than the reference value, a human facial region has a characteristic of not being detected as exceeding a predetermined size. Therefore, in the present invention, a facial region among candidate regions is detected through a comparison and judgment step (S247).

As another example, when the patient's legs are overlapped, since the width of the leg region is detected to be less than the second reference value, after the comparison and determination module 3477, each candidate region tracking module 3478 based on accumulated data is executed. .

The step of tracking each candidate region based on the accumulated data (S248) is performed when the width of all candidate regions is less than the second reference value in the comparison and determination step (S247).

In addition, in the step of tracking each candidate region based on the accumulated data (S248), the local management server 3 utilizes the data extracted by the step of extracting the accumulated data (S242) in the direction of the previous point in time based on the current point in time, This step is to analyze the cell location information, detect each candidate region, and then detect the size of the width of each candidate region.

In addition, in the step of tracking each candidate region based on the accumulated data (S248), a width change rate (%) of each candidate region is calculated by utilizing values of the detected width values of each candidate region.

In the step of comparing and determining whether the width is changed (S249), the local management server 3 compares the width change rate (%) of each candidate region calculated by the tracking step (S248) of each candidate region based on the cumulative data, and the width size change rate This is a step of determining a candidate region with a low percentage as a facial region.

For example, the face region has a small change rate (%) of width size even when the posture is changed, but the leg region has a high change rate (%) compared to the face region even when the posture is not changed.

In the facial region determination step (S250), when the local management server 3 determines the facial region in 1) the comparison and determination step (S247) or 2) the width variation comparison and determination step (S249), the determined facial region is selected for the patient. This is the final decision step for the facial area of the face.

The patient's body temperature calculation step (S251) is a step in which the local management server 3 determines the average value of the optimum temperature values of the cells corresponding to the facial region determined in the facial region determining step (S250) as the patient's body temperature.

In the abnormal body temperature determination step (S30), the local management server 3 compares the patient's body temperature detected in the patient temperature calculation step (S251) with a preset normal body temperature range, 1) when the patient's body temperature is within the normal body temperature range, In this step, it is determined that the patient's body temperature is normal, and 2) when the patient's body temperature is out of the normal body temperature range, it is determined that abnormal body temperature has occurred in the patient.

In the fall determination step (S40), the local management server 3 receives the cell information of the patient object detected in the patient object recognition step (S20), and the bedroom area information of the corresponding heat-detecting sensor stored in the database unit 32. After extracting, it is a step of comparing the cell location of the patient object and the cell location of the bedroom by utilizing the input patient object cell information and the extracted bedroom area information.

In addition, in the step of determining whether or not to fall (S40), it is determined that a fall accident occurred in the patient when the elapsed time when the cell position of the patient object deviated from the cell position of the bedroom is equal to or greater than a threshold value.

The sudden situation data generation step (S250) is performed when a patient having an abnormal body temperature is detected in the abnormal body temperature determination step (S30) or when it is determined that a fall has occurred in the fall state determination step (S40).

In addition, in the emergency situation data generation step (S250), the local management server 3 generates unexpected situation data including the contents of the emergency situation, the identification number of the corresponding ward, and the identification number of the corresponding bedroom.

The data transmission step (S60) is a step in which the local management server 3 transmits the emergency situation data generated in the emergency situation data generation step (S50) to manager terminals.

The monitoring step (S70) is a step of outputting sound warning information to the outside when the manager terminal receives the unexpected situation data through the data transmission step (S60).

The optimal constant setting step (S80) is performed every predetermined period (T), and when the local management server 3 receives the temperature value of each ward transmitted through the collected data receiving step (S10), the database unit 32 After extracting the reference table stored in , searching the extracted reference table and extracting optimal constants (TH) that match the temperature values of each ward. At this time, the reference table means data in which the optimal constant (TH) is matched for each range of temperature values in the ward.

In addition, in the optimal constant setting step (S80), the extracted optimal constants (TH) of each ward are stored in the database unit 32.

As described above, the ward sudden situation alarm and monitoring method (S1), which is an embodiment of the present invention, is configured to simultaneously determine the abnormal body temperature of the patient in each bedroom and whether or not a fall accident has occurred using the silver heat-detecting sensor 9, so that separate By not using cameras and thermal imaging cameras, it not only improves privacy issues, but also reduces installation costs and enables low-power operation.

In addition, in the ward emergency alarming and monitoring method (S1) of the present invention, after the local management server 3 analyzes the heat-sensing information received from the data collection devices 4 to recognize an object, the temperature of the recognized object By using the values and location information, it is configured to determine the abnormal body temperature of the patient and whether or not a fall accident has occurred, thereby increasing the accuracy of object recognition while simplifying calculation processing.

In addition, in the ward unexpected situation alarming and monitoring method (S1) of the present invention, when the local management server 3 determines that the patient's abnormal body temperature and a fall accident have occurred, the manager terminal 5 identifies the contents of the unexpected situation, the ward, and the guest room. Emergency situation data including a number is transmitted, and when the management program 7 of the manager terminal 5 receives the unexpected situation data from the local management server 3, warning alarm information is output to the outside so as to determine the patient's condition. Remote monitoring is possible to improve work efficiency and maximize the efficiency of the nursing management system, and in the event of an unexpected situation, prompt follow-up is possible, increasing the safety and reliability of medical services.

1: Ward emergency alarm and monitoring system
3: local management server 4: data collection device
5: manager terminal 7: management program
9: heat-detection sensor 10: communication network
11: ward thermometer 31: control unit
32: database unit 33: data transmission and reception unit
34: patient object recognition unit 35: abnormal body temperature determination unit
36: Fall determination unit 37: Unexpected situation data generation unit
38: optimal constant setting unit 341: heat-sensing information input module
343: object recognition module 361: data input module
362: Bedroom area information extraction module 363: Location comparison module
364: fall determination module 381: ward temperature value input module
382: Reference table extraction module 383: Optimal constant determination module
S1: Ward emergency alert and monitoring method
S10: Collected data receiving step S20: Patient object recognition step
S21: heat-sensing information input step S22: object recognition step
S23: Object location information extraction/storage step S24: Facial region detection step
S30: Determination of abnormal body temperature S40: Step of determining whether or not to fall
S50: Emergency situation data generation step S60: Data transmission step
S70: Monitoring step S80: Optimal constant setting step
S241: Object location information input step S242: Accumulated data extraction step
S243: longitudinal direction determination step S244: verification step
S245: Step of determining candidate regions S246: Step of calculating the width of each candidate region
S247: Comparison and judgment step
S248: Tracking each candidate area based on accumulated data
S249: Width variation comparison and determination step S250 Facial area determination step
S251: Patient body temperature calculation step

Claims (5)

Installed on the top of each bedroom of at least one ward, and measuring the temperature by subdividing the preset detection area (S) into a plurality of cells, using heat-detection sensors that measure the heat-detection information, which is the temperature value for each cell, , In the ward emergency alarm and monitoring method (S1) for monitoring whether or not an unexpected situation occurs in each ward:
The ward unexpected situation alarm and monitoring method (S1)
A local management server including a database unit in which a reference table, which is data matched with optimal constants (TH) for each range of temperature values of the ward, is set and stored, receives heat-sensing information from the heat-sensing sensors, Step 10 (S10) of receiving a ward temperature value from a ward thermometer installed in the ward and measuring the temperature value of the ward;
After the local management server analyzes the heat-sensing information received through step 10 (S10) and extracts the temperature value for each cell, the extracted temperature value for each cell is compared with the preset body temperature range to be included in the body temperature range and continuously Step 20 (S20) of recognizing the cells as the patient object and detecting the cell information of the patient object;
After the local management server extracts the temperature value for each cell extracted in step 20 (S20), it calculates the patient body temperature, which is the average value of the temperature values for each cell of the extracted patient object, and puts the calculated patient body temperature within the predetermined normal body temperature range. 1) determining that the patient's body temperature is normal when the patient's temperature value is within the normal body temperature range, and 2) determining that an abnormal body temperature has occurred in the patient when the patient's body temperature value is out of the normal body temperature range 30 (S30);
Step 50, when the local management server determines that an unexpected situation has occurred when the patient's body temperature is out of the normal body temperature range in step 30 (S30), and generates unexpected situation data including the details of the unexpected situation and ward and room identification information ( S50);
Step 60 (S60) of the local management server transmitting the emergency situation data generated in step 50 (S50) to manager terminals;
It proceeds at a predetermined period (T), and when the local management server receives the temperature value of each ward transmitted through the step 10 (S10), the standard stored in the database unit After extracting the table, searching the extracted reference table, extracting optimal constants (TH) matching the input temperature values of each ward, and then storing the extracted optimal constants (TH) of each ward in the database unit Including step 80 (S80),
The step 20 (S20) is
When the local management server receives heat-sensing information through the step 10 (S10), after extracting the temperature value for each cell from the received heat-sensing information, the optimal constant (TH) stored in the database unit is extracted for each of the extracted heat-sensing information. The optimal temperature value for each cell is calculated by multiplying with the temperature value for each cell, and the calculated optimal temperature values for each cell are compared with the preset body temperature range to recognize consecutive cells included in the body temperature range as patient objects,
The optimal constant (TH) stored in the reference table has a smaller value as the temperature value of the corresponding ward increases, and a higher value as the temperature value of the ward decreases;
The step 20 (S20) is
Step 21 (S21) of receiving the heat-sensing information received through the step 10 (S10) by the local management server;
After the local management server extracts the temperature value for each cell from the heat-sensing information input through the step 21 (S21), the optimal constant (TH) stored in the database unit is multiplied by the extracted temperature value for each cell to obtain the temperature for each cell. After calculating the optimal value, the calculated optimal temperature value for each cell is compared with the preset body temperature range, and consecutive cells included in the body temperature range are recognized as patient objects, and cell information of the patient object is detected Step 22 (S22) ;
Step 23 (S23) of the local management server detecting the cell-location information of the patient object recognized by the step 22 (S22) and then storing the detected cell-location information of the patient object in the database unit;
Including step 24 (S24),
Step 24 (S24) is
Step 241 (S241) of receiving, by the local management server, the cell-location information of the patient object detected by step 23 (S23);
Step 242 (S242) of extracting, by the local management server, cell-location information of the corresponding patient object stored in the database unit, data from a current point in time to a preset previous time (t);
Step 243 (S243) of determining the longitudinal direction of the patient by the local management server analyzing cell-location information of the patient object input through step 241 (S241);
Step 245 (S245) of determining, by the local management server, areas adjacent to both ends in the longitudinal direction of the patient object determined in step 243 (S243) as candidate areas;
step 246 (S246) of calculating, by the local management server, the size of the width (cell size) of each candidate area determined in step 245 (S245);
The local management server compares the width of each candidate region calculated in step 246 (S246) with a preset second reference value, which is the maximum value of the width for determining that the recognized candidate region is a facial region. , 1) When one of the candidate regions has a width less than the second reference value and the other has a width greater than or equal to the second reference value, a candidate region having a width less than the second reference value is determined as a facial region; , 2) Step 247 (S247) of repeating the process after going to step 21 (S21) when the widths of all candidate regions are equal to or greater than the second reference value;
In step 247 (S247), the process is performed when the widths of all candidate regions are less than the second reference value, and the local management server uses the data extracted in step 242 (S242) to determine the size of the previous point in time based on the current point in time. direction, the cell-location information of the patient object is analyzed, each candidate region is detected, and then the width of each candidate region is detected, and the width of each candidate region is detected using the width and size values of each candidate region. Step 248 of calculating a size change rate (%) (S248);
Step 249 (S249) of the local management server comparing the change rate (%) of the width size of each candidate region calculated in step 248 (S248) and determining that the candidate region having a low change rate (%) is the facial region. ;
Step 250 (S250) of the local management server, when the facial region is determined in step 247 (S247) or step 249 (S249), finally determining the determined facial region as the facial region of the corresponding patient;
Step 251 (S251) of the local management server determining the average value of the optimum temperature values of the cells corresponding to the facial region determined in step 250 (S250) as the body temperature of the corresponding patient;
The step 30 (S30) is
A ward emergency situation warning and monitoring method (S1), characterized in that the local management server determines whether or not the abnormal body temperature is present by utilizing the patient's body temperature determined in the step 251 (S251). delete delete The method of claim 1, wherein the step 24 (S24)
Further comprising a step 244 (S244) performed after the step 243 (S243),
The step 244 (S244) is
After the local management server calculates the length (L, cell-size) of the patient object by utilizing the cell-location information of the patient object and the longitudinal direction of the patient object determined in step 243 (S243), the calculated patient The length (L) of the object is compared with a predetermined reference value, which is the minimum value of the length of an object that can be judged to be a recognized object as a body. determining that there is an error, and 2) if the length (L) of the patient object is greater than or equal to the reference value, determining that there is no error in determining the length direction of the patient 2441 (S2441);
The local management server compares whether there are many Y-axis components or X-axis components in the longitudinal direction of the bed in the longitudinal direction of the patient object determined in step 243 (S243), 1) if the length of the patient object If there are many X-axis components in the direction, it is determined that an error has occurred in determining the longitudinal direction, and 2) if it is determined that there are many Y-axis components in the longitudinal direction of the patient object, step 2442 of determining that there is no error in determining the longitudinal direction (S2442 );
1) If the local management server determines that no error has occurred in both steps 2441 (S2441) and 2442 (S2442), it verifies that there is no error in determining the longitudinal direction of the patient object in step 243 (S243), 2) When it is determined that an error has occurred in at least one of the steps 2441 (S2441) and the step 2442 (S2442), step 2443 of finally determining that an error has occurred in determining the longitudinal direction of the patient object in step 243 (S243) (S2443),
The step 245 (S245) is
When the local management server verifies that there is no error in determining the longitudinal direction of the patient object in step 243 (S243) in step 244 (S244), it proceeds to the next step.
The step 21 (S21) is
When the local management server determines in step 244 (S244) that an error has occurred in determining the longitudinal direction of the patient object by the longitudinal direction determination module in step 243 (S243), it proceeds to the next step. Alarming and monitoring methods (S1). The method of claim 4, wherein the database unit
Bedroom area information, which is location information of cells corresponding to a bedroom in the sensing area S of each of the heat-detecting sensors, is preset and stored,
The ward unexpected situation alarm and monitoring method (S1)
Further comprising a step 40 (S40) performed after the step 30 (S30),
The step 40 (S40) is
After the local management server extracts the bedroom area information of the corresponding heat-sensing sensor stored in the database unit, it uses the cell information of the patient object detected in step 20 (S20) and the extracted bedroom area information to obtain the patient object Comparing the cell position of and the cell position of the bedroom, and when the elapsed time when the cell position of the patient object is deviated from the cell position of the bedroom is greater than a threshold value, it is determined that a fall accident has occurred in the patient,
The step 50 (S50) is
Ward emergency situation warning and monitoring method (S1), characterized in that the local management server generates unexpected situation data when it is determined that a fall accident has occurred in the patient in the step 40 (S40).