JP2005128967A - Medical device, method and program for detecting movement and computer readable recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a medical device for detecting movement capable of preventing a care receiver from rolling off a bed. <P>SOLUTION: A medical method for detecting movement includes a step for successively imaging a care receiver on a resting floor part from one or more positions and/or angles so that the movement of the care receiver can be detected; a step for setting, with respect to the images of the care receiver, a plurality of detection areas A in the vicinity of the part of the care receiver's body which is very likely to be moved when he or she behaves in a dangerous way, and for setting a threshold for changes in part of each image as a reference for determining whether or not an abnormality has occurred in the plurality of detection areas A; a step for comparing each image with the previous image in order to detect whether or not a change exceeding the threshold is occurring in at least either of the plurality of detection areas A; and a step for reporting the occurrence of an abnormality to a base station 23 and displaying the image of the care receiver if a change exceeding the threshold is detected. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a medical motion detection device, a medical motion detection method, a medical motion detection program, and a computer-readable recording medium that detect an abnormal motion of a cared person for medical purposes.

  In the medical field, a system for protecting the caregiver's safety is required. In particular, as elderly people, bedridden elderly people, inpatients, etc. (hereinafter referred to as “caregivers”) become a part of nursing care in hospital beds, nursing homes for the elderly, health care facilities for the elderly, hospitals, etc. Although the importance of assistance for helping to move up with a care recipient who needs care has increased, accidents in which these persons get up from the bed without assistance and fall are a major problem. Specifically, there are many accidents in which a care recipient falls accidentally from the bed when getting out of the bed for a bag or toilet. In recent years, it has become a serious problem that a cared person with cognitive impairment falls from a bed, breaks a bone, falls asleep and falls into a vicious circle. For cognitive impairment, demented elderly with behavioral and problem behaviors, patients with higher brain dysfunction with symptoms such as unilateral neglect or apraxia or agnosia that lose sensation in a part of the body, frontal lobe with attention disorder Symptoms, delirium, etc. caused by a decrease in consciousness level. These cognitively impaired people themselves think that they can move without any problems, so when they get out of bed, an accident that falls or falls occurs. Also, care recipients with attention disabilities often forget to press the nurse call or have the illusion that they can act alone without assistance. There is an accident that falls immediately after leaving the bed to move to a wheelchair or portable toilet.

  As a result, injuries such as head injury, fracture, bruise, and laceration occur. Furthermore, not only such temporary damage but also secondary damage such as disuse syndrome, fear, anxiety and loss of confidence due to reduced activity caused by rest after primary damage and recovery from primary damage. Cause problems such as loss of trust in nurses and other staff. The above is a vicious circle, and as shown in Fig. 1, even a cognitively impaired person who was physically healthy can be physically and physically affected by a bed fall accident while receiving nursing care in the hospital. It is mentally damaged, the activity declines due to the need for rest, the progression of disuse syndrome, and the worsening of cognitive impairment. As a result, the number of bed fall accidents increases further, eventually accumulating damage. It falls into a vicious circle of being bedridden. In this way, the contradiction that the medical condition gets worse is actually caused by the hospital life to recover the original health.

  Such a fall accident occurs especially at night. In the daytime, there are many assistants, caregivers, and nurses, so monitoring is sufficient, and accidents can be avoided by preventing fall accidents and early detection of dangerous behavior. However, since the number of caregivers and nurses drastically decreases at night, sufficient monitoring cannot be performed. According to the regulations, at least two nurses are required for 50 to 60 beds in hospitals with full care. However, this number of people is physically difficult to deal with, and the physical burden on the nurse is large due to the so-called nurse call and patrol, etc., and the nurse center tends to be out of service for a long time. However, the current situation is that it is not possible to prevent this problem sufficiently, and the discovery after the accident is delayed. On the other hand, although it is conceivable to increase the number of nurses and strengthen the monitoring system, the reality is that it is difficult to realize because the burden due to the increase in cost and the securing of personnel are not easy. However, attaching a caregiver to each care recipient and monitoring for 24 hours has a large financial burden on the care service and a physical burden on the caregiver due to family care, and it is impossible to realize it.

  On the other hand, in order to physically prevent such a nighttime fall accident, the cared person's torso and limbs can be tied with restraining belts, or restrained clothes can be forcibly taken away. Some methods, such as administering sleeping pills to prevent awakening, have been adopted. However, such a means deprives the cared person of freedom of the body and action and reduces physical strength. Moreover, since it will not only harm the independence of the cared person but also deprive human dignity, it has been suspended in recent years from the viewpoint of medical care that respects the dignity of the cared person.

  In order to solve such problems in nursing care and the like, an abnormality monitoring device disclosed in Patent Document 1 has been developed. This abnormality monitoring apparatus can promptly notify a caregiver of an abnormality of a care receiver without constantly monitoring the care receiver. This anomaly monitoring apparatus compares an infrared illuminator that irradiates infrared light, an imaging means attached to a ceiling, a wall surface, etc., and an image taken by the imaging means with an image taken before that image. A change detection type image processing unit that obtains information necessary for human body detection and determines whether the cared person is moving or stationary based on pre-stored knowledge for human body detection; After it is determined that the person has moved, a determination processing unit is provided that determines that an abnormality has occurred in the cared person when it is determined that the cared person has been stationary for a predetermined stationary abnormality detection time.

  The abnormal state of the cared person detected by the abnormality monitoring device corresponds to a state where the cared person falls and cannot move, or a state where the cared person falls from a bed or a chair and stands stationary. That is, as shown in FIG. 2, after the movement of the cared person is detected, an abnormality in the movement of the cared person is detected only for a specific state in which the movement of the cared person suddenly stops. . For this reason, the abnormality monitoring device detects the care recipient's movement and then detects that the care recipient is in an abnormal state only when the care recipient's movement suddenly stops, that is, when there is no change between images. Thus, there is a problem that only a very limited state in which it is determined that an abnormality has occurred in the cared person when it becomes stationary can be dealt with.

  Furthermore, Patent Document 2 discloses a detection device that detects the movement of a care recipient. As shown in FIG. 3, this detection device learns the movement of a cared person using a neural network genetic algorithm (GA), so that the cared person does not always monitor the cared person through the video camera. When the caregiver's condition can be grasped and the abnormal movement occurs in the movement of the care recipient, the caregiver is notified promptly and the burden on the caregiver is reduced. However, since this method requires a system for performing advanced learning using a genetic algorithm of a neural network, there is a problem that a complicated and expensive system is required.

  Furthermore, all of the above methods detect the caregiver's movement before and after a fall accident, and detect the occurrence of the accident from the stationary state after the accident, and can only detect an abnormality after the fall accident. It was not intended to prevent accidents. As mentioned above, in order to break the bedridden due to repeated bed fall accidents and the vicious circle of sickness deterioration, a system that detects and prevents abnormalities before the accident occurs is indispensable. It was not put into practical use.

  Patent Document 3 discloses a safety management system for preventing an accident such as a fall when the care recipient tries to leave the bed by himself / herself. As shown in FIGS. 4 (a) and 4 (b), this safety management system is provided with a plurality of sensors on the bed 6, and the base station grasps that the cared person is leaving the bed 6, and By operating the central control unit, the infrared CCD camera 14 installed in the room is operated, and the state of the bed where the sensor call is made is captured by the infrared CCD camera 14, and this is provided in the central control unit. The system which displays the image which is being displayed on a liquid crystal monitor, and visually observes the condition of a bed is disclosed.

  However, in this system, the movement of the cared person is detected by a physical sensor, but the weight sensors 7 and 8 are installed below the mattress 17 of the bed 6 and sense the movement of the cared person's weight. Infrared photoelectric sensors 9a and 9b installed between the bed head side 6a and the bed foot side 6b for sensing the movement of the cared person, and infrared pyroelectric sensors installed under the bed 6 for sensing the movement of the cared person The sensor 10, the touch sensor 11 that is installed on the bed shelf 18 and senses that the cared person has touched, and the string member 12 that is stretched between the bed head side 6a and the bed foot side 6b, and is cared for. Since it is necessary to install a large number of sensors such as the bed sensor 13 that senses the movement of the person, the probability of a malfunction such as the person being cared for hitting his / her arm by turning over other than when an abnormality occurs increases. In addition, the installation of a large number of sensors and the work of wiring the drive circuit thereof are complicated, and there are also disadvantages that the installation and maintenance costs are high and the system is expensive.

  Similarly, Patent Document 4 discloses a dynamic detection system that detects a series of movements from a sleeping position of a cared person to the start of walking for each dynamic. As shown in FIG. 5, the dynamic detection system includes (a) a sleeping position, (b) a long sitting position, (c) an end sitting position, (d) a standing position, and (e) a walking start time. The respective dynamics are detected step by step for each dynamic, and the detection signal is output as a dynamic signal as display information or an alarm sound. This dynamic detection system includes two upper optical sensors 19, two pressure-sensitive sheets 20, and a lower optical sensor 21 as a dynamic detection device. Also in this method, a plurality of sensors are used, and it is necessary to construct wirings for driving these sensors for each hospital bed, which causes a problem of introducing and maintaining equipment. Moreover, the problem of malfunction is unavoidable because many sensors are used. Furthermore, there is a problem that an abnormality can be detected only with an operation pattern as shown in FIG. Depending on the cared person, the body type, height, heel, etc. vary widely, and accurate detection is not expected at certain fixed positions. Furthermore, the posture and procedure of getting up vary depending on the symptoms of the cared person. For example, a care receiver who has a broken right foot and a care receiver who does not have a left hand sensation naturally have different procedures for getting up. In the system of Patent Literature 4, there is a problem in that it is impossible to perform settings according to the movement of each care recipient, and as a result, an accurate operation cannot be obtained.

  In addition, in both Patent Documents 3 and 4, since the sensor is fixed, the position of the sensor once installed cannot be easily changed, and the bed must be fixed according to the position of the sensor. The change cannot be made, and caregivers and caregivers are inconvenienced. In addition, there are problems of increased costs and complicated hospital systems due to the installation of sensors in each bed.

Although various methods have been proposed and developed in this way, they are far from being able to provide practical solutions that can prevent accidents of care recipients from falling. As a problem that could not be done, there was even a state of compliment, and a practical technique that could be expected to prevent falling was eagerly desired from the field of care.
JP 11-341474 A JP 2002-230555 A JP 2000-105885 A JP 2001-327549 A

  The present invention has been made as a practical solution based on the opinions of medical sites in order to solve such problems. The main object of the present invention is to provide an efficient medical motion detection device, medical motion detection method, medical motion detection program, and computer-readable recording medium that can prevent an accident that a care recipient falls from the bed. Is to provide.

  In order to achieve the above object, the medical motion detection device according to claim 1 of the present invention is provided with one or more arranged at one or more different positions in order to image the movement of the cared person G from one or more angles. An imaging unit 1, an area setting unit 3 capable of setting a plurality of arbitrary detection areas A with respect to the image of the cared person G captured by the imaging unit 1, and a plurality of areas set by the area setting unit 3 As a reference for discriminating the occurrence of abnormality of the care recipient G in the detection area A, the threshold setting unit 4 for setting the threshold value of the change occurring in a part of the image to be adjustable, and the threshold setting unit 4 are set. Based on the detected threshold value, the abnormality detection unit 2 for detecting whether or not an abnormality has occurred in the plurality of detection regions A set by the region setting unit 3, and based on the abnormality detection by the abnormality detection unit 2 And a communication unit 16 for notifying the base station 23 of the occurrence of an abnormality. That.

  The medical motion detection device according to claim 2 is the medical motion detection device according to claim 1, and is further set by the region setting unit 3 based on a threshold value set by the threshold value setting unit 4. When it is determined that an abnormality has occurred in the plurality of detection areas A, a recording unit 5 that starts recording an image of the care receiver G is provided.

  Furthermore, the medical motion detection device according to claim 3 is the medical motion detection device according to claim 1 or 2, wherein when the abnormality detection unit 2 detects an abnormality, the communication unit 16 performs the base station 23. Is notified of the occurrence of an abnormality and the terminal 25 carried by the caregiver is notified of the occurrence of the abnormality.

  Furthermore, the medical motion detection device according to claim 4 is the medical motion detection device according to any one of claims 1 to 3, wherein when the abnormality detection unit 2 detects an abnormality, the care receiver G It is further provided with an alerting unit 28 for issuing a message for alerting the user.

  Furthermore, the medical motion detection device according to claim 5 is the medical motion detection device according to any one of claims 1 to 4, wherein the threshold set by the threshold setting unit 4 is the region setting unit. In the detection area A set in 3, the number of cells is set so that it is determined that an abnormality has occurred when the number of cells constituting the detection area A changes by a predetermined number or more.

  Furthermore, the medical motion detection device according to claim 6 is the medical motion detection device according to any one of claims 1 to 5, wherein the abnormality detection unit 2 is set by the region setting unit 3. When abnormality is detected in a predetermined number of detection areas A among the plurality of detection areas A, it is determined that an abnormality has occurred.

  Furthermore, the medical motion detection device according to claim 7 is the medical motion detection device according to any one of claims 1 to 6, wherein the detection region A set by the region setting unit 3 is a rectangular region. , At least one of a circular region and a free region.

  The medical motion detection method according to claim 8 is a step of continuously imaging the cared person G on the bedside from one or more positions and / or angles so that the movement of the cared person G can be detected. A plurality of detection areas A are set in the vicinity of a part that is frequently moved when the cared person G performs dangerous actions on the captured image of the cared person G, and abnormalities are detected in the plurality of detection areas A. As a criterion for determining whether or not it occurs, a step of setting a threshold value of a change that occurs in a part of the image, and a change that exceeds the threshold value occurs in at least one of the plurality of detection areas A among the captured images. A step of comparing with each of the previous images in order to detect whether or not there is a change, and if a change exceeding the threshold is detected, the base station 23 is notified of the occurrence of an abnormality and the image of the care receiver G is displayed. Step and No.

  Furthermore, the medical motion detection method according to claim 9 is the medical motion detection method according to claim 7, wherein when a change exceeding a threshold is further detected, an abnormality occurs in the terminal 25 carried by the caregiver. And a step of displaying the image of the care receiver G on the monitor of the terminal 25.

  Furthermore, the medical motion detection program according to claim 10 has a function of continuously imaging the cared person G on the bedside from one or more positions and / or angles so that the movement of the cared person G can be detected. A plurality of detection areas A are set in the vicinity of a part that is frequently moved when the care receiver G performs a dangerous action on the captured image of the care receiver G. As a criterion for determining whether or not an abnormality has occurred, there is a function for setting a threshold value of a change that occurs in a part of an image, and a change that exceeds the threshold value in at least one of a plurality of detection areas A in a captured image. In order to detect whether or not it has occurred, a function for comparing with each of the previous images, and when a change exceeding the threshold is detected, the base station 23 is notified of the occurrence of an abnormality and the image of the care receiver G is displayed. Function and threshold When a change exceeding 1 is detected, the terminal 25 carried by the caregiver is notified of the occurrence of an abnormality, and the function of displaying the image of the care recipient G on the monitor of the terminal 25 is realized by a computer. It is a motion detection program.

  A computer-readable recording medium according to an eleventh aspect of the present invention stores the medical motion detection program according to the tenth aspect. Recording media include CD-ROM, CD-R, CD-RW, flexible disk, magnetic tape, MO, DVD-ROM, DVD-RAM, DVD-R, DVD-RW, DVD + R, DVD + RW, Blu-ray, AOD And other media that can store programs, such as magnetic disks, optical disks, magneto-optical disks, semiconductor memories, and the like. The above-mentioned program includes a program distributed by downloading through a network line such as the Internet.

  According to the medical motion detection device, the medical motion detection method, the medical motion detection program, and the computer-readable recording medium of the present invention, it is possible to prevent an accident that the care receiver falls from getting up from the bed. At the same time, early detection at the time of an accident becomes possible. The system does not focus on early detection after an accident such that the caregiver suddenly stops moving after the accident, but detects an abnormality in a state where there is a risk of an accident and notifies the care recipient That's why. As a result, the caregiver can be dispatched promptly to the caregiver who has taken the problem action to prevent an accident. Further, as a means for detecting an abnormality, a method is adopted in which a detection area is designated for each care recipient and an abnormality is determined when any change occurs in this area. As a result, optimal settings according to the characteristics of the cared person can be made, and the accuracy of abnormality detection can be greatly improved. In conventional systems, it is difficult to distinguish between abnormal behavior and non-behavior from the movement of the cared person, and there are many false detections that are not suitable for practical use. For example, a method of incorporating learning using a neural network has been studied. However, since an extremely sophisticated and complex system is required, there is a disadvantage that the introduction and maintenance costs are high. In the present invention, the detection region is set in the vicinity of the part to be moved when moving to the dangerous action according to each patient, thereby succeeding in drastically reducing erroneous detection. In particular, in the present invention, since a plurality of detection areas can be set for each care recipient, abnormality detection can be performed with optimal settings according to the patient, and erroneous detection is prevented and reliable abnormality detection is realized.

  Furthermore, since the actual caregiver's state can be visually confirmed when an abnormality is detected, whether or not an abnormality has occurred can be easily determined from a remote place without going to the site, and useless behavior can be eliminated. Furthermore, by storing the log in the recording unit 5, it is possible to grasp and analyze the behavior pattern of the cared person, and it is possible to obtain a secondary effect that it can be used as subsequent treatment or clinical medical data. By using the present invention in this way, it is possible to prevent an accident that the care recipient falls from the bed at night, to improve the safety of the care recipient, and to reduce the burden on the caregiver. It is possible to break the vicious circle caused by a fall accident, and to restore not only the body and mental recovery of the cared person but also the trust of the medical system.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the embodiments described below exemplify a medical motion detection device, a medical motion detection method, a medical motion detection program, and a computer-readable recording medium for embodying the technical idea of the present invention. However, the present invention does not specify the medical motion detection device, the medical motion detection method, the medical motion detection program, and the computer-readable recording medium as follows.

  Further, the present specification by no means specifies the members shown in the claims to the members of the embodiments. In particular, the dimensions, materials, shapes, relative arrangements, and the like of the component parts described in the embodiments are not intended to limit the scope of the present invention unless otherwise specified, and are merely explanations. It's just an example. Note that the size, positional relationship, and the like of the members shown in each drawing may be exaggerated for clarity of explanation. Furthermore, in the following description, the same name and symbol indicate the same or the same members, and detailed description thereof will be omitted as appropriate. Furthermore, each element constituting the present invention may be configured such that a plurality of elements are constituted by the same member and the plurality of elements are shared by one member, and conversely, the function of one member is constituted by a plurality of members. It can also be realized by sharing.

  In this specification, a medical motion detection apparatus, a medical motion detection method, a medical motion detection program, and a computer-readable recording medium include a system for performing abnormality detection and notification of a care receiver, and abnormality detection and its The present invention is not limited to an apparatus and method that perform input / output, display, calculation, communication, and other processes related to notification in hardware. An apparatus and method for realizing processing by software are also included in the scope of the present invention. For example, software, programs, plug-ins, objects, libraries, applets, scriptlets, compilers, modules, macros that run on specific programs, etc. are incorporated into general-purpose circuits or computers to detect anomalies, notify them, or process related to them General-purpose or dedicated computers, workstations, terminals, portable electronic devices, PDC, CDMA, W-CDMA, FOMA (registered trademark), GSM, IMT2000, fourth generation mobile phones, PHS, PDA, A pager, a smartphone, and other electronic devices are also included in at least one of the medical motion detection apparatus, the medical motion detection method, the medical motion detection program, and the computer-readable recording medium of the present invention. In this specification, the program itself is included in the medical motion detection apparatus. In addition, this program is not limited to the one used alone, but is provided as a service in an environment that functions as a part of a specific computer program, software, service, etc., an aspect that is called and functions when necessary, and an environment such as an OS. It can also be used as a mode, a mode that operates resident in the environment, a mode that operates in the background, and other support programs.

[Connection, communication type]
Connection between terminals such as computers used in the embodiments of the present invention, servers, computers connected to these terminals, operation, control, input / output, display, various processing, and other peripheral devices such as printers For example, serial connection such as IEEE1394, RS-232x, RS-422, USB, etc. Can do. The connection is not limited to a physical connection using a wire, but may be a wireless connection using a wireless LAN such as IEEE802.11x or OFDM, a radio wave such as Bluetooth, infrared light, optical communication, or the like. Furthermore, a memory card, a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like can be used as a recording medium for exchanging data or storing settings. In this specification, the connection typically means an electrical connection, and does not necessarily mean only a physical connection. Recently, transmission and reception of data and energy using electro-optical elements such as OEIC (Optoelectronic Integrated Circuit) has been realized. Electromagnetic, pressure, sound wave, radio wave, heat, etc. including such electricity and light. The state of being “connected” so that signal data transmission / reception and transmission / reception of various types of energy can be performed is also referred to as the connection according to the present invention, and direct connection or indirect connection does not matter. Furthermore, it is not necessary to be always connected, and the switch circuit or the switching circuit is configured to be connected only when necessary (for example, only when charge, electricity, or current passes) according to the driving state of the drive circuit. Also good.

  FIG. 6 shows a configuration example of a medical motion detection apparatus according to an embodiment of the present invention. The motion detection device shown in this figure includes a plurality of imaging units 1, an abnormality detection unit 2, an area setting unit 3, a threshold setting unit 4, a display unit 15, a communication unit 16, and a recording unit 5. . In this medical motion detection apparatus, the anomaly detection unit 2 detects this when the cared person G starts to move down from the bed at night, and the recording unit 5 starts recording and immediately the nurse. By notifying caregivers, etc., they try to prevent a fall accident.

[Imaging unit 1]
The imaging unit 1 images the figure of the cared person G on the bedside such as the bed 6 and sends it to the abnormality detection unit 2. The imaging unit 1 can use a camera, a motion sensor, a wire sensor, a beam sensor, an ultrasonic sensor, or the like provided in a hospital room. Preferably, a night vision camera such as a CCD camera or an infrared (IR) camera is used rather than a sensor. In particular, this embodiment is preferably a night vision camera that can monitor the care receiver G even when the bed is turned off in order to prevent a fall accident from a bedside location at a time when there are few caregivers at night. Further, it may be configured to switch to a normal CCD camera in a bright state such as during the day or when the room light is lit. As a result, it is possible to cope with a patient who prefers to sleep in a bright state with the room light turned on, and to prevent accidents during the day.

  The bed place is a place where a cared person is located and is typically a bed. For example, if a cared person does not like a bed, the embodiment of the present invention can be used even when a futon is placed on a tatami mat. Can be applied. If you place a tatami mat in a hospital room and lay a futon on it, there will be no fall accident because there is no height difference like a bed. There are cases where the tatami mat and the floor will trip and fall. Or, an accident may occur, such as a cared person lying down while sleeping and hitting her body against something. Even in such a case, the present embodiment can be applied to detect a fall accident or the like at an early stage and prevent it beforehand. Alternatively, an accident may occur in which a cared person in a wheelchair or a chair sits down and falls or falls.

  Furthermore, this embodiment can be applied not only to a normal bed but also to a special bed such as a gudge bed. The gadget bed is equipped with high / low functions for raising the back and adjusting the height, so that the cared person can be kept in the sitting position. However, an accident may occur in which a cared person who is resting in a sitting position falls off the bed, falls, and fractures. Therefore, the present embodiment can be applied to prevent such an accident. As described above, in this specification, a bed place is used in the sense of including a bed, a futon, a wheelchair, a sitting chair, etc., where a care receiver is placed, lying down, or sitting.

  The imaging unit 1 captures images from different positions with a plurality of cameras so that the figure and movement of the care recipient G can be reliably captured. For example, the bed 6 is fixedly installed at an angle at which the bed 6 can be imaged, such as a ceiling and a wall near the bed 6, or near the bed 6. It is preferable to provide a plurality of cameras, preferably three as shown in FIG. By capturing images from various angles with a plurality of units, the movement of the cared person G can be more reliably captured three-dimensionally. However, a single camera may be used from the viewpoints of cost reduction, installation, and ease of maintenance. The camera is installed at a place where the entire image of the care receiver G is taken, such as the ceiling. Further, the camera may be not only a fixed type but also a movable type capable of operations such as turning and tilting. In particular, by making it possible to operate the camera movement and zoom from the base station 23, it is possible to display a necessary part on the monitor 24 of the base station 23 at the time of occurrence of an abnormality. Detailed confirmation is possible. Furthermore, the state of the bed may be widely monitored by periodically turning the camera. Operations such as camera movement and ON / OFF can also be set by a timer.

  The video of the cared person G captured by the imaging unit 1 is sufficient if the movement of the cared person G can be detected. Therefore, it is possible to consider the care recipient's privacy by intentionally reducing the image quality, not the clear image quality that allows the face to be identified. Furthermore, as will be described later, if the image confirmation at the base station 23 or the terminal 25 carried by the caregiver is possible only in an emergency for risk prevention, that is, only when the abnormality detection unit 2 detects an abnormality, The caregiver's privacy is protected as much as possible. Further, by turning off the power of the monitor 24 of the base station 23 and the monitor of the terminal 25 except when an abnormality is detected, there is also a secondary effect of suppressing the consumption of the backlight of the liquid crystal monitor and extending the life of the device. can get.

[Abnormality detection unit 2]
The image captured by the imaging unit 1 is sent to the abnormality detection unit 2. The abnormality detection unit 2 always determines whether or not a change exceeding the threshold set by the threshold setting unit 4 has occurred in the detection area A set by the region setting unit 3 among the images captured by the imaging unit 1. Monitoring. When a change exceeding the threshold is detected, it is determined that an abnormality has occurred, and a predetermined operation is performed.

[Area setting unit 3]
The area setting unit 3 designates a plurality of detection areas A on the basis of the normal state in which the care receiver G stops. First, a stable state in which the cared person G lies on the bed 6 is imaged by the imaging unit 1 and displayed on the display unit 15. The display unit 15 is a display such as a liquid crystal monitor or a CRT, and may also be used as an input device using a touch screen or a touch panel as necessary. The above stable state is registered as a reference image in the abnormality detection unit 2 and a desired part is designated as a detection region A as shown in FIG. The user uses a pointing device such as a mouse to designate a part corresponding to the characteristics of the cared person G as a detection area A having an arbitrary shape on the screen. For example, the detection area A may be designated as a rectangular area by clicking a vertex on a diagonal line with a mouse, or may be a circular area in which a center and a radius are designated, or a free area surrounded by a free curve. Or you may comprise so that the area | region of the preset pattern, the figure pattern of the detection area A already set, and the pattern set by the care receiver G may be copied and used.

  The detection area A is an area for detecting an abnormality when a change occurs in the image in this area. Typically, when the cared person's G body and limbs enter the detection area A, it is determined that the cared person G is about to get out of the bed 6 and an abnormality has occurred. The setting of such an area is different for each care recipient. For example, in the caregiver who has no sensation on the right side, the caregiver with a foot disability, or the caregiver with a hand disability, in any order, move the limbs, head, or body to get out of bed and get down The patterns, procedures, and methods are naturally different. There are also wandering and individual differences for each care recipient. Therefore, in the embodiment of the present invention, by making it possible to individually set the detection area for each care recipient, it is possible to adjust to an optimum setting according to the care recipient and realize a reliable danger detection.

  In the example of FIG. 7, the detection area A is set in an area that does not overlap the care receiver G. However, an area including the care receiver G can be set as the detection area A. For example, a detection area A can be set in an area including the care receiver G, and a change in the cell in the detection area A can be monitored to detect an abnormality. In this specification, it is assumed that a plurality of detection areas A are set, but it is needless to say that only one detection area A can be used. In this specification, a plurality of detection regions are used to include one detection region.

  Conventional detection methods often respond even when the caregiver's hand moves, and so many false detections occur, so the nurse eventually becomes untrusted by the system, or the system itself turns off and the inherent danger is necessary. Most of them fall into the pattern of missing actions. On the other hand, according to research, when a cared person moves his / her body, it has been reported that 60 to 70% is moved from the head. Thus, in order to exclude useless movement from detection and detect only actions that are likely to lead to dangerous actions, the present embodiment does not monitor the entire image of the care receiver G, By setting the detection area A and monitoring only such an area, the accuracy of abnormality detection is improved. Improve accuracy by monitoring from multiple points, not just one point, and adjust detection area according to each cared person to eliminate a considerable amount of false detection and high practicality with improved detection accuracy A motion detection device was realized.

  The detection area A is designated from the image displaying the stable state as shown in FIG. Preferably, it is set in the vicinity of a portion where movement is expected most first. In addition to the setting on the plane, when the imaging unit 1 is provided on the side of the care receiver G, it can be set on an image viewed from the side as shown in FIG. This makes it possible to accurately grasp the movement in the height direction. Furthermore, the images captured by the plurality of imaging units can be combined and displayed in a three-dimensional manner, and the detection area can also be specified in a three-dimensional manner. At this time, the detection area may be set not as a plane but as a solid.

  In addition, in any one of the detection areas A, in addition to a method of determining an abnormality when a change exceeding the threshold occurs, an abnormality occurs when a predetermined number of detection areas, for example, three detection areas, are detected. Or may be set to be determined as abnormal when changes are detected in a predetermined order in a plurality of detection areas. A plurality of detection areas can be set, and in the example of FIG. 7, a maximum of 10 can be set per cared person.

[Threshold setting unit 4]
In the detection area A set by the area setting unit 3, the threshold setting unit 4 sets a threshold value that serves as a reference for whether or not an abnormality has occurred. The threshold setting unit 4 can set various parameters as thresholds. For example, the number of cells is set as a threshold value so that an abnormality is determined when a change occurs in a predetermined number of cells or more in the cells constituting the detection area A. A cell is a unit that constitutes an image picked up by the image pickup unit 1, and an area corresponding to one pixel of a CCD is defined as one cell, for example. The size of the cell can be arbitrarily set. For example, 10 cells × 10 pixels can be set as one cell, and can be arbitrarily set according to the sensitivity and resolution of the CCD and the display screen, the accuracy of abnormality detection, and the like. For cell change, chromaticity, luminance, and the like can be used. In these examples, the image picked up by the image pickup unit 1 is, for example, a color image of 256 gradations, and changes in chromaticity of an image picked up with 8-bit resolution are detected. Alternatively, when it is difficult to obtain a color image such as an infrared camera, a change in luminance is detected based on luminance information such as monochrome that excludes chromaticity information. Further, a plurality of parameters such as the number of cells as the threshold value and the change degree of chromaticity and luminance as the degree of change of each cell may be combined. In the example of the interface screen of the medical motion detection program shown in FIG. 9, the threshold setting integrated by combining the number of cells and the brightness can be continuously adjusted by the slider 26, and the selected threshold is changed to an actual image. An adapted preview can be displayed in the display area 27.

  The threshold value can be adjusted for each set detection area A. The user can adjust the default threshold value to an optimum value for each detection area A. In particular, the adjustment of the threshold value can be changed as appropriate based on the operation status not only before the system is operated but also after the system is actually operated. Combined with the fact that the area A can be individually set for each care recipient, it is possible to obtain optimum detection sensitivity in accordance with the use situation.

  The abnormality detection unit 2 detects an abnormality by monitoring only the detection region A set by the region setting unit 3, not the entire image captured by the imaging unit 1. Images captured in real time are accumulated in a temporary memory of the computer 22 at predetermined intervals, and a difference from the previously captured image is examined. If no change exceeding the threshold set by the threshold setting unit 4 is detected, the past image data is discarded and updated with newly captured image data. If a change in the newly captured image data that exceeds the threshold value set by the threshold value setting unit 4 as compared with the previous image data is detected, it is determined as abnormal. In the example shown in the figure, a signal imaged by the imaging unit 1 is A / D converted, and image data is captured in a frame memory of the computer 22 for processing.

  Alternatively, the normal reference image may be used instead of the image data obtained by previously capturing the image used as the reference for comparison. An abnormality may be detected by comparing an image captured in real time with a reference image. As described above, the abnormality detection unit 2 has a simple operation of monitoring only a part of the change of the image and detecting the abnormality, and the system can be configured at a low cost with a necessary processing as a light load. In particular, a complicated process of learning the movement of the care recipient is unnecessary, and the process can be performed lightly and at high speed. In addition, since it is sufficient for the image data to be able to detect the movement of the cared person, it is not necessary to have high accuracy. Since the number of cells is reduced from the viewpoint of privacy protection as described above, the size is small. It is small and can contribute to inexpensive and high-speed processing. This system keeps operating costs down and the size of the device itself is relatively easy.

[Recording unit 5]
As described above, when a change greater than or equal to the threshold set by the threshold setting unit 4 occurs, the abnormality detection unit 2 determines that an abnormality has occurred, notifies the base station 23 of the occurrence of the abnormality, and records it. Part 5 starts image recording. The recording unit 5 starts recording by using the abnormality detection in the abnormality detection unit 2 as a trigger, and continuously records the movement of the care receiver G. When recording is started, the recording unit 5 continuously records images or takes still images at predetermined time intervals. The recording is automatically stopped when a preset time has elapsed since the abnormality was detected. Alternatively, the recording may be manually stopped when the abnormality is released.

  Recording may be started before the occurrence of an abnormality, or may be recorded before the occurrence of the abnormality. For example, data captured in real time by the imaging unit 1 is temporarily stored in a memory of a computer or a cache of a fixed disk, and an image for several seconds is held for a certain period of time. After a certain period of time, the data is discarded by updating the cache. When the occurrence of an abnormality is detected, image data for several seconds before the abnormality detection is extracted from the data stored in the cache, and the recording unit 5 is started from this data. With this configuration, it is possible to record from an image before abnormality detection. In this way, by recording an image from several seconds before the occurrence of an abnormality, for example, 5 seconds before, it is possible to record how the abnormality occurs and to help elucidate the process of the occurrence of the abnormality.

  As the recording unit 5, for example, a fixed storage device such as a hard disk, a semiconductor such as a DVR (digital video recorder), a CD-R, or an MO, a tape, magnetism, light, or a magneto-optical recording medium can be used. It is preferable to use a DVR that has a large capacity, is excellent in random accessibility, is inexpensive, and can be easily exchanged. In this way, by recording the care receiver's behavior at the time of occurrence of abnormality in the recording unit 5, a log can be left, which can be used for analysis of the problem behavior of the care receiver. In particular, in the DVR, since recorded images and videos can be called instantly, it can be used conveniently in later analysis. In addition, recording does not continue for 24 hours, but recording starts from the occurrence of problematic behavior, so that necessary data can be efficiently stored.

  The data recorded in the recording unit 5 can be automatically backed up. For example, data is automatically transferred from the recording unit 5 to a recording medium such as a hard disk of a server connected to a network, a backup DVD-R or a tape at regular intervals, and backup is executed. Prepare for the situation.

  The functions of the abnormality detection unit 2, the region setting unit 3 for setting the detection region A, the threshold setting unit 4 for setting a threshold indicating image change, and the communication unit 16 can be realized by a computer 22 as shown in FIG. . A monitor connected to the computer 22 is used as the display unit 16, and the area setting unit 3 and the threshold setting unit 4 are operated with an input device such as a mouse or a keyboard. A computer 22 shown in FIG. 10 functions as a server, and controls and monitors information exchange. Since the computer 22 is network-connected to the base station 23 and the cameras of each hospital bed, the computer 22 may be installed anywhere, and is installed in the vicinity of the base station 23 or a dedicated server room. Further, the computer 22 can also function as the base station 23. In this case, the monitor installed in the base station 23 can be shared with the monitor of the computer 22.

[Base station 23]
When an abnormality is detected by the abnormality detection unit 2, information is transmitted to the base station 23. The base station 23 is preferably shared with a room or the like packed by a caregiver such as a nurse station. When the base station 23 is notified of the occurrence of an abnormality, the person in charge at the base station 23 is notified visually by an alarm, voice, light, or the like. For example, it sounds an alarm sound and lights / flashes a warning light. Further, the monitor 24 provided in the base station 23 is turned on, and the state of the bed 6 in the corresponding hospital bed is monitored. The state of the care recipient G can be confirmed in detail by turning on the lighting of the hospital bed or increasing the illuminance as necessary (not necessary in the case of an infrared camera). In addition, by changing the illuminance, it is possible to expect an effect of attracting the care recipient G as described later and discouraging continuing dangerous actions.

  Since the state of the care recipient G can be confirmed on the monitor 24 by the video, it can be surely confirmed by human eyes whether the abnormality is true. In the case of a false detection such as just turning over, the abnormality occurrence signal is manually canceled. If it is recognized that an abnormality has occurred, the caregiver immediately goes to the appropriate bed. In order to confirm the state of the hospital bed with the video at the base station 23, the video captured by the imaging unit 1 provided in each hospital bed is transmitted to the base station 23 in real time. Since this system can be used as both the imaging unit 1 that monitors abnormalities and the imaging unit 1 that checks whether there is an abnormality at the time of abnormality detection, it is possible to eliminate the need for a sensor or the like that is related to the movement of the cared person and to be inexpensively configured There is. Also, by not using such a sensor, the problem of sensor misdetection can be eliminated, and the sensor can be removed from the vicinity of the cared person's bed, and the sensor can be cleaned up. A simple medical motion detection apparatus is realized by eliminating the problem of handling.

  The monitor 24 may be set so that it is turned off at normal times and automatically turned on when an abnormality occurs. This can extend the life of the monitor 24. Alternatively, if necessary, the monitor 24 may be turned on even during normal times so that the state of each bed is appropriately switched and displayed so that the image of the corresponding bed is switched when an abnormality occurs. At this time, it is possible to add a highlight process that draws the caregiver's attention, such as blinking red on the screen. Thus, by reporting the occurrence of abnormality in both voice and video, the caregiver can reliably recognize the occurrence of the abnormality without overlooking it.

  The monitor 24 can also display the screen divided into a plurality of screens, and can simultaneously display images captured by a plurality of cameras on the same bed on a single screen, or display cameras on other beds at the same time. . Thus, even when a plurality of abnormalities occur at the same time, the situation can be confirmed by displaying them on the monitor at the same time. In the illustrated example, 16 cameras can be managed by one monitor 24. Of course, a plurality of monitors may be prepared in advance such as preparing a spare monitor, and individual images may be displayed using these monitors. Alternatively, it goes without saying that a plurality of images may be switched and displayed on a single monitor.

[Terminal 25]
The caregiver carries a terminal 25 having a communication function such as PDA or PHS, and the base station 23 transmits an abnormality occurrence signal to each terminal 25. For example, when the PDA receives an abnormality occurrence signal, the monitor of the PDA is automatically turned on, the image of the corresponding hospital bed transmitted from the base station 23 is displayed, and the caregiver displays the actual situation of the cared person G in video. I can confirm. If an error is detected in the same manner as described above, the abnormality occurrence signal is manually canceled, and if an abnormality is confirmed, the person is rushed to the site to assist the care recipient.

  Furthermore, by providing the PDA with the same function as that of the base station 23, the mobility of the caregiver can be utilized. Nurses are often not only packed in nurse stations but also on patrols. For this reason, the nurse station may be unattended. Even in such a case, when the abnormality detection unit 2 detects an abnormality and transmits an abnormality detection signal to the nurse station, the nurse station transmits an abnormality detection signal to each nurse's PDA, and the nurse who is traveling is also abnormal. An outbreak can be announced. And since the nurse can confirm the state of the actual bed from the PDA, it can also confirm whether it is a false detection or not, and if an abnormality occurs, it can promptly take action for accident prevention. In particular, among a plurality of nurses, a nurse closest to the bed where an abnormality is detected can rush to the scene. As a result, it is possible to expedite the site in a short time and efficiently, and to increase the possibility of preventing accidents. In addition, even if an accident occurs, early detection and appropriate measures are possible, minimizing damage. It becomes possible to suppress.

  In addition, when a nurse in a nearby location is in the process of assisting another care recipient, the nurse in the second closest location can rush to the site instead. These determinations and work sharing are determined by nurses calling directly on terminals 25 such as PHS and PDA, or the base station 23 can determine the positions of each nurse and automatically assign them. Good. In this case, it is preferable that the PDA or PHS is provided with a current position detection function such as GPS so that the base station 23 can grasp the position of each nurse. Alternatively, the positions of the nurses detected by the current position detection function may be transmitted to each nurse's terminal 25 so that the positions of the nurses can be confirmed on the monitor. Thereby, since each nurse's position can visually confirm each other, it becomes easy to allocate each other's work share efficiently. Preferably, the terminal 25 carried by a caregiver such as a nurse is a terminal in which PDA and PHS functions are integrated, and can perform high-speed data communication with the base station 23 by a communication method such as a wireless LAN. Necessary functions such as a call function capable of voice communication with a person and a GPS function capable of detecting the position of the terminal 25 are incorporated. It also includes hardware such as a liquid crystal monitor, communication unit, and charging unit. Of course, it is needless to say that a plurality of devices can be carried in combination and used like existing PDAs and PHS.

  Moreover, the structural example of the medical motion detection apparatus provided with the alerting part 28 is shown in FIG. 11 as other embodiment. When the anomaly detection unit 2 detects an abnormality, the alerting unit 28 issues a message that alerts the care recipient G to stop the care receiver G from continuing a dangerous action, By taking the time to move to, delay the occurrence of the accident until the nurse arrives. The alerting unit 28 includes a monitor 28A and a speaker 28B for reproducing a message. The message includes voice guidance, a warning sound, an image and a moving image, lighting and flashing of lighting, and the like, and as appropriate, these are appropriately combined to alert the care recipient G. It is preferable that the message has a content that does not give anxiety or discomfort to the cared person so that the cared person is surprised by issuing a warning message and does not induce an accident. For example, it sounds like "How did you do it?", "Please wait a moment", "I'll continue now". For voice guidance, recording by a nurse in charge or the like is more preferable than synthetic voice of a computer because it gives a caregiver a sense of security. For example, an appropriate message is recorded in advance according to the personality and preference of the cared person, the dialect of the place of birth, etc., and is reproduced when an abnormality occurs. At this time, not only audio but also moving images can be added to the message. When the nurse's face is displayed on the monitor 28A provided in the vicinity of the bed, the cared person can confirm not only the voice but also the face, so that a greater sense of security can be obtained. The video message is recorded in advance with the nurse's face and voice simultaneously with a video camera or the like. The message is not limited to this, but can also be music or words that the care recipient likes, voices and appearances of close relatives, images of pets and plants, messages of entertainers such as singers and actors, and video clips. Or it is not restricted to an audio | voice and an image | video, The generation | occurrence | production of an odor can also be utilized. In this way, by reproducing the message according to the care recipient's preference, it is possible to prevent the occurrence of an accident by drawing the care recipient's attention and stopping the operation, earning time until the nurse rushes I can expect. Furthermore, these messages can be common to all care recipients, and can also be dedicated messages for each care recipient. The medical motion detection device automatically selects and plays back a dedicated message corresponding to the cared person at the time of occurrence of an abnormality, so that the cared person's attention can be more reliably prevented and the occurrence of an accident can be prevented. The possibility increases.

  Further, when an abnormality occurs, communication may be performed between the base station and the terminal and the monitor 28A provided in the hospital room so that the nurse can directly talk with the care receiver G in real time. For conversation, the nurse call communication function provided in the bed can be used. Alternatively, the alerting unit 28 may have a communication function. As a result, it is possible to reliably receive and respond to the care recipient G, to respond appropriately to the abnormal situation, and to check whether or not it is malfunctioning.

  Furthermore, by recording the log in the recording unit 5, it is possible to analyze the care recipient's behavior pattern, and to aggregate data such as the time from the occurrence of an abnormality to arrival at the site, the incidence of accidents, and the avoidance rate. It can be used for future medical policy planning, nursing care system, patrol system improvement, clinical medical data, etc. There has been no recording system for behavior patterns that can cause such bed fall accidents, and the elucidation and research of disease states and patterns have not been sufficiently advanced so far. In this example, by recording data in the recording unit 5 from the time of occurrence of abnormality, not only the above-mentioned accident prevention but also such clinical research data can be used as clinical data, It can contribute to the establishment of treatment methods. In particular, if the DVR is used for the recording unit 5, the recorded video can be called instantly, so that the problem behavior can be easily analyzed, and the recording medium is a large-capacity medium such as a DVD-R disk, so that it is relatively inexpensive. The advantage is that it is easy to handle, saves space, and does not degrade data. In addition, as the investigation of the medical condition progresses, it becomes possible to make predictions about the pattern in which a fall accident is likely to occur, and more efficient measures can be taken, such as planning a tour in a time corresponding to this. This can reduce the burden on the caregiver. At the same time, feedback for more effective use of the system can be obtained by analyzing the log. For example, the imaging unit 1 can be adjusted to a more optimal position to improve accuracy. In addition, the safety of care recipients is protected by reducing the number of bed fall accidents, and the recovery of the medical condition is promoted. Can also be expected. In addition, the post-processing burden on the hospital side due to the reduction of fall accidents is reduced, and it is expected to greatly contribute to the construction of an environment that can literally care for the care recipient while maintaining medical dignity.

  Moreover, although the above embodiment has described an example of preventing a fall accident when the number of caregivers such as nurses is reduced at night, it goes without saying that this embodiment can also be used during the day. For example, it can be suitably used when the number of caregivers is smaller than the number of caregivers for some reason such as weekends and holidays.

  Further, the above embodiment can be applied to a so-called healthy person as well as a cared person having some kind of disease. Even a healthy person may be unaware that he / she has some kind of disability, so he may be able to get out of the bed without assistance even though he / she needs assistance. For example, if you have a physical injury due to an injury or illness and you have not received a doctor's examination, or you have not heard of a diagnosis, you missed it, or you misunderstood. I sometimes think that I can get off. Even in such a case, as described above, an accident may occur in which the user falls from the bed and breaks without assistance. The embodiment of the present invention can also be applied to such a case, and an accident can be prevented beforehand. In the present specification, the term “cognitive impairment” is used in the sense including the above-mentioned misunderstandings of healthy individuals, and the present invention can be applied to this case.

  The medical motion detection device, the medical motion detection method, the medical motion detection program, and the computer-readable recording medium of the present invention include a special nursing home for the elderly, a bedridden elderly person, etc., a geriatric health facility, a hospital, etc. In this facility, it can be suitably used for the purpose of preventing a fall / fall accident from the bed.

It is explanatory drawing explaining the vicious circle where a cared person gets worse by repeated fall / fall accidents. It is explanatory drawing which shows operation | movement of the conventional abnormality monitoring apparatus. It is explanatory drawing which shows the outline of the other conventional motion detection apparatus. It is a perspective view which shows the bed to which the other conventional safety management system is applied. It is a perspective view explaining the dynamic detection operation | movement detected by the other conventional dynamic detection system. It is a block diagram which shows the structural example of the medical motion detection apparatus which concerns on one embodiment of this invention. It is the schematic which shows a mode that a detection area | region is designated from the top view which imaged the care receiver. It is the schematic which shows a mode that a detection area | region is designated from the side view which imaged the care receiver. It is the schematic which shows the example which adjusts a threshold value with a threshold value setting part. It is a block diagram which shows the structural example of the medical motion detection apparatus which concerns on other embodiment of this invention. It is a block diagram which shows the structural example of the medical motion detection apparatus which concerns on other embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Imaging part 2 ... Abnormality detection part 3 ... Area setting part 4 ... Threshold setting part 5 ... Recording part 6 ... Bed 6a ... Bed head side 6b ... Bed foot side 7, 8 ... Weight sensor 9a, 9b ... Infrared photoelectric sensor DESCRIPTION OF SYMBOLS 10 ... Infrared pyroelectric sensor 11 ... Touch sensor 12 ... String member 13 ... Bed sensor 14 ... Infrared CCD camera 15 ... Display part 16 ... Communication part 17 ... Mattress 18 ... Bed shelf 19 ... Upper light sensor 20 ... Pressure sensitive sheet 21 ... Lower light sensor 22 ... Computer 23 ... Base station 24 ... Monitor 25 ... Terminal 26 ... Slider 27 ... Display area 28 ... Warning part 28A ... Monitor 28B ... Speaker A ... Detection area G ... Care recipient

Claims (11)

One or more imaging units (1) arranged at one or more different positions to image the movement of the care recipient G from one or more angles;
An area setting section (3) capable of setting a plurality of arbitrary detection areas (A) for the image of the cared person G captured by the imaging section (1);
In order to set the threshold value of the change that occurs in a part of the image as an adjustment as a reference for determining the occurrence of abnormality of the care recipient G in the plurality of detection areas (A) set by the area setting unit (3) Threshold setting section (4),
Based on the threshold set by the threshold setting unit (4), an abnormality detection unit for detecting whether an abnormality has occurred in the plurality of detection regions (A) set by the region setting unit (3) ( 2) and
Based on the abnormality detection in the abnormality detection unit (2), a communication unit (16) for notifying the base station (23) of the occurrence of the abnormality,
A medical motion detection device comprising:   The medical motion detection device according to claim 1, further comprising: a plurality of detection regions (A) set by the region setting unit (3) based on a threshold set by the threshold setting unit (4). A medical motion detection apparatus comprising: a recording unit (5) that starts recording an image of a cared person G when it is determined that an abnormality has occurred.   The medical motion detection device according to claim 1 or 2, wherein when the abnormality detection unit (2) detects an abnormality, the communication unit (16) notifies the base station (23) of the occurrence of the abnormality. A medical motion detection device that notifies the occurrence of an abnormality to a terminal (25) carried by a caregiver.   It is a medical motion detection apparatus in any one of Claim 1 to 3, Comprising: When the said abnormality detection part (2) detects abnormality, the alerting part which emits the message which alerts the care receiver G ( 28) A medical motion detection apparatus further comprising:   The medical motion detection device according to any one of claims 1 to 4, wherein a threshold value set by the threshold value setting unit (4) is a detection region (A) set by the region setting unit (3). The medical motion detection device according to claim 1, wherein the number of cells is set so as to determine that an abnormality has occurred when the number of cells constituting the detection area (A) changes by a predetermined number or more.   The medical motion detection device according to any one of claims 1 to 5, wherein the abnormality detection unit (2) includes a plurality of detection regions (A) set by the region setting unit (3), A medical motion detection apparatus configured to determine that an abnormality has occurred when an abnormality is detected in a predetermined number of detection regions (A).   The medical motion detection apparatus according to any one of claims 1 to 6, wherein the detection area (A) set by the area setting unit (3) is at least one of a rectangular area, a circular area, and a free area. A medical motion detection device characterized by the above. Imaging the cared person G on the bed place continuously from one or more positions and / or angles so that the movement of the cared person G can be detected;
A plurality of detection areas (A) are set in the vicinity of a site that is frequently moved when the care receiver G performs a dangerous action on the captured image of the care receiver G, and a plurality of detection areas ( As a reference for determining whether or not an abnormality has occurred in (A), a step of setting a threshold value of a change occurring in a part of the image;
A step of comparing with each of the previous images in order to detect whether or not a change exceeding a threshold value has occurred in at least one of the plurality of detection regions (A) among the captured images;
When a change exceeding the threshold is detected, notifying the base station (23) of the occurrence of an abnormality and displaying an image of the care receiver G;
A medical motion detection method comprising:   The medical motion detection method according to claim 7, wherein when a change exceeding a threshold value is further detected, an abnormality is notified to a terminal (25) carried by a caregiver and an image of the care receiver G is obtained. Is displayed on the monitor of the terminal (25). A function of continuously imaging the cared person G on the bed place from one or more positions and / or angles so that the movement of the cared person G can be detected;
A plurality of detection areas (A) are set in the vicinity of a site that is frequently moved when the care receiver G performs a dangerous action on the captured image of the care receiver G, and a plurality of detection areas ( A function for setting a threshold value of a change occurring in a part of an image as a reference for determining whether or not an abnormality has occurred in (A),
A function of comparing with each of the previous images in order to detect whether or not a change exceeding the threshold value has occurred in at least one of the plurality of detection regions (A) among the captured images,
When a change exceeding the threshold is detected, the base station (23) is notified of the occurrence of an abnormality, and the function of displaying the image of the care recipient G;
When a change exceeding the threshold is detected, a function of notifying the terminal (25) carried by the caregiver of the occurrence of an abnormality and displaying an image of the care receiver G on the monitor of the terminal (25);
A motion detection program for medical use to realize a computer.   A computer-readable recording medium storing the medical motion detection program according to claim 10.

Images