JP5225344B2 - Living behavior storage device - Google Patents

Description

  Embodiments of the present invention relate to a living behavior storage device that converts and presents a living behavior stored for each user.

  Conventionally, when an observation target person such as an elderly person is monitored using a monitoring monitoring apparatus, the living behavior of the observation target person who wants to know is different for each type of user using the monitoring monitoring apparatus. For example, security personnel need to know abnormal conditions that require urgent action such as cardiac arrest, falls, suspicious person intrusions, etc., and doctors must keep patient (observer) biological information out of the normal range. It is necessary to know the abnormal values that are out of place, the day care manager needs to know the behavior that is different from the normal behavior of the elderly (observer), and the distant family members are elderly relatives (observer) ) It is necessary to know the state of daily life.

JP 2003-296855 A

  However, when it is desired to memorize the appropriate living behavior of the observation subject at any time according to the disease state and preventive medicine, the living behavior is slightly different depending on the terminology and field of the doctor or the like. However, the monitoring target in the conventional monitoring device is limited to any one of the living behaviors, and there is a problem that it is difficult to simultaneously monitor the living behaviors of the observation subjects having different types of users.

  Therefore, an object of the present invention is to provide a living behavior storage device that can present a living behavior of a person to be observed necessary for each type of user.

The present invention provides an observation unit that observes an observation target with a plurality of types of sensors and acquires observation information for each of the sensors, a primary storage unit that stores the acquired observation information for each of the sensors, and a user A knowledge storage unit that stores in advance observation knowledge composed of applications for obtaining necessary living behavior data for each type from observation information in the observation information for each sensor, and the observation knowledge An extraction unit for extracting necessary living behavior data for each type of the user using the necessary observation information from the observation information stored in the primary storage unit, and the extracted life a secondary storage unit for storing the behavior data, from the living behavior data, stored in advance to have rule storage unit conversion rule for converting the presentation content required for each type of the user, the conversion Based on the law, a conversion unit that converts the living action data stored in the secondary storage unit to the presentation content, depending on the type of the user, and the life activities data regarding the observation subject A presentation unit for presenting the presentation content , the observation unit includes an acceleration sensor and a sensor other than the acceleration sensor, and the observation unit uses the acceleration sensor to observe the person to be observed. Acceleration is detected as the observation information, and the knowledge storage unit stores in advance an application for obtaining a walking distance or the number of steps from the acceleration as observation knowledge, and the extraction unit stores the observation knowledge application and the acceleration. Based on the walking distance or the number of steps of the person to be observed as life activity data, the conversion unit presents the walking based on the walking distance or the number of steps. In addition to the living behavior data based on the acceleration sensor and the presentation content, the presentation unit also presents the lifestyle behavior data based on the other sensors and the presentation content. This is a daily behavior storage device characterized by that.

The block diagram of the living action memory | storage device of Example 1 of this invention. The block diagram of an observation part. An example of the conversion rule memorize | stored in the rule memory | storage part. Life activity data of the observation subject for one week stored in the secondary storage unit. The flowchart until it extracts life action data. The flowchart until it shows presentation content. An example of the presentation content. An example of a conversion rule when a new conversion rule is added. Living behavior data in the secondary storage unit when a conversion rule is added. The block diagram of the living action memory | storage device of Example 2. FIG. FIG. 10 is an example of a conversion rule corresponding to the second embodiment.

  Hereinafter, a living behavior storage device 1 according to an embodiment of the present invention will be described with reference to the drawings.

  First, terms used in this specification will be described.

  “Metabolic syndrome” refers to a condition in which visceral fat type obesity (visceral obesity / abdominal obesity) is combined with two or more of hyperglycemia, hypertension, and hyperlipidemia. Non-patent document 1 (Compendium of physical activities: an update of activity codes and METs (Metabolic equivalents) intensities METs are defined by ACSM (American College of Sports Medicine), Ainsworth BE, et.al., Sci Sports Exerc. 20000, 32 (9 Suppl): S498-504 Classification is based on the described Mets.

  “Mets” means a unit system in which the momentum at rest is 1 Mets.

  “ICF” means a classification for describing life function disorders in Non-Patent Document 2 (International Classification of Functioning, Disability and Health by WHO).

  “ADL” refers to the need for physical care described in Non-Patent Document 3 (Mahoney FI, Barthel D., Functional evaluation: The Barthel Index, Maryland State Medical Journal, 14, 56-61, 1965). (ADL: Activities of daily living Barthel index).

  “IADL” is a dementia described in Non-Patent Document 4 (Lawton, MP and Brody, EM, Assessment of older people: Self-Maintaining and instrumental activities of daily living, Gerontologist, -9, 179-186, 1969). It means a classification for mental care needs such as IADL (instrumental activities of daily living).

  “Observation target person” means a person whose living behavior is observed by the living behavior storage device 1 of the present embodiment, such as an elderly person, a patient, or a person who is instructed to eliminate metabolic syndrome. .

  The “user” means a person who observes, monitors, and watches the person to be observed by the living behavior storage device 1 of the present embodiment, such as a doctor, a care planner, a distant family, and the like.

  “Identification behavior” means behavior such as walking and life rhythm of the observation subject.

  “Life behavior data” is a further subdivision of the identification behavior. For example, regarding the walking of the observation subject, it means the start or stop of walking, the number of steps, the walking distance, and the number of meals regarding the life rhythm. Means sleep time, presence or absence of washing.

  A living behavior storage device 1 according to a first embodiment of the present invention will be described with reference to FIGS.

  The configuration of the living behavior storage device 1 of the present embodiment will be described based on the block diagram of FIG.

  The living behavior storage device 1 includes an observation unit 10, a primary storage unit 20, a knowledge storage unit 30, an extraction unit 40, a secondary storage unit 50, a presentation unit 60, a conversion unit 70, a rule storage unit 80, and a control unit 90. It has.

  The observation unit 10 has various sensors for observing observation information regarding the living behavior of the observation subject. That is, as shown in FIG. 2, the observation unit 10 includes a motion information input unit 11 that acquires the motion of the observation subject's body using an acceleration sensor, and a sound that acquires sound generated by the observation subject and surrounding environment using a microphone. The information input unit 12 includes an image information input unit 13 that acquires images of the observation target person and the surrounding environment from the camera, and acquires observation information (acceleration, sound, image, etc.) of the observation target person. The person to be observed wears the acceleration sensor. In addition, microphones are attached to bedrooms, living rooms, kitchens, and washrooms where the observers live, and detect sounds generated by the observers. Cameras are also used in bedrooms, living rooms, kitchens, It is attached to a washroom, etc., and takes pictures of the observation subjects themselves and the observation subjects. The observation information observed by the observation unit 10 is stored in the primary storage unit 20.

  The knowledge storage unit 30 stores, for each type of user such as a doctor and a care planner, observation knowledge describing life behavior data that needs to be obtained from the observation target and measurement intervals of the life behavior data. doing. This observation knowledge is stored for each identification action (for example, walking, life rhythm, etc.) required for each user such as ADL, IADL, and ICF.

  Based on the observation knowledge stored in the knowledge storage unit 30, the extraction unit 40 extracts necessary living behavior data for each type of user from the observation information of the observation target person stored in the primary storage unit 20. . That is, based on the observation information stored in the primary storage unit 20, the extraction unit 40 obtains the observation target person from a stationary state, an operating state such as a walking state (walking or running), the number of walks, the number of walks, and the stride. Extract life activity data such as walking distance. The extracted living behavior data is stored in the secondary storage unit 50 as the living behavior data of the person to be observed.

  The rule storage unit 80 stores conversion rules for converting daily behavior data into behavior expression terms and presentation formats for each type of user. The rule storage unit 80 will be described with reference to FIG. The table in FIG. 3 is an example of conversion rules stored in the rule storage unit 80.

  Each item arranged in the column direction of the table of FIG. 3 is divided for each identification action stored in the knowledge storage unit 30. The row direction is divided into user types.

  For each discriminating action in the column direction, for example, items (walking, life rhythm, etc.) necessary for each user such as ADL, IADL, and ICF are stored.

  As user types in the row direction, user types such as ADL, IADL, ICF, and metabolic syndrome are stored. Users of ADL, IADL, and ICF are in-care physicians and care planners, ICF users are ordinary home doctors, and users of metabolic syndrome are specialists such as the Center for Life Disease Management. is there. In this conversion rule, it is assumed that when there is a unique evaluation standard for care in each municipality or at a day care service center, the evaluation standard is described.

  Since ADL is an index for determining whether or not an observation subject is dementia, regarding walking, whether or not the observation subject can walk more than 45 m is used as one dementia criterion. Therefore, the item “value” in the rule storage unit 80 is “45”, the item “unit” is “m”, the item “large” is “≧”, the item “measurement method” is “cycle”, and the item “ “One week” is stored in the “measurement period”, and “not difficult to walk” is stored in the item “presentation content”. Then, the observation unit 10 observes whether or not one continuous walk (one cycle) has become 45 m or more during the measurement period of one week. The presentation unit 60 presents the presentation “not” to the ADL user (for example, a caregiver).

  In the ICF, even if it is the same walking, it is a description regarding the presence or absence of a loss of physical function, so whether or not it is a short distance walking (ICF code is d4500) is based on whether or not a walking of 1000 m is possible. Therefore, each item is a value different from ADL. That is, the “value” is a threshold value set for each user. When 45, it is possible to provide a meaningful presentation content for an ADL user, and when 1000, an ICF user, this threshold value. By changing, presentation contents necessary for each user can be obtained.

  In the metabolic syndrome, one reference is whether the total number of steps per day is 10,000 steps, not a one-cycle walk, even for the same walk. Therefore, the item “value” in the rule storage unit 80 is “10,000”, the item “unit” is “step”, the item “measurement method” is “integrated”, and the item “measurement period” is “1 day”. Is stored, and “presented” and “not achieved” are stored in the presented content.

  In IADL, each item is “NA (non-adaptive)” because IADL does not have a corresponding identification action item in a meal of walking or life rhythm.

  The conversion unit 70 converts the living behavior data stored in the secondary storage unit 50 into the presentation content based on the conversion rules stored in the rule storage unit 80.

  The presentation unit 60 presents living behavior data and presentation contents corresponding to each user.

  The control unit 90 controls the observation unit 10 and the observation storage unit 30 or adds a new conversion rule in order to extract living behavior data necessary for the conversion rule.

  A method in which the living behavior storage device 1 detects the living behavior data (information relating to walking, information relating to movement, living sounds, etc.) of the observation subject will be described.

  The first walking detection method will be described.

  Based on the control of the control unit 90, the acceleration sensor of the motion information input unit 11 in the observation unit 10 detects the XYZ triaxial acceleration raw waveform according to the stationary or movement of the observation target person.

  The observation unit 10 differentiates the detected raw waveform of the XYZ triaxial acceleration (difference between the front and rear accelerations) to obtain a differential acceleration. The primary storage unit 20 stores a raw waveform of triaxial acceleration or differential acceleration as observation information.

  The extraction unit 40 sets a threshold value for at least one differential acceleration stored in the primary storage unit 20 based on the observation knowledge stored in the knowledge storage unit 30, and the acceleration change exceeding the threshold value within a predetermined time. The presence or absence of occurrence identifies whether the observation subject is stationary or operating. Gravity acceleration (for example, it appears remarkably in the Z-axis direction) is observed in the raw waveform, but only the difference can be easily extracted by obtaining the differential waveform. Accordingly, the extraction unit 40 starts the walking of the observation target person when the acceleration change exceeding the first threshold value occurs, and ends the walking of the observation target person when the acceleration change value exceeding the first threshold value disappears. It can be detected. Furthermore, if the second threshold value that is larger than the first threshold value is exceeded, the extraction unit 40 can detect that the observation target person is in the traveling state.

  In this case, the extraction unit 40 calls the observation knowledge corresponding to the first walking detection method from the knowledge storage unit 30 in order to use the first walking detection method. In this first gait detection method, the observation knowledge stored in the knowledge storage unit 30 is that the identification behavior is related to “walking”, and the differential acceleration stored in the primary storage unit 20 is at least one axis or more. It is an application that sets a threshold value and detects whether or not an acceleration change exceeding the threshold value occurs within a predetermined time.

  Next, the second walking detection method will be described.

  Based on the control of the control unit 90, the acceleration sensor of the motion information input unit 11 in the observation unit 10 detects the XYZ triaxial acceleration raw waveform according to the stationary or movement of the observation target person.

  The observation unit 10 differentiates the detected raw waveform of the XYZ triaxial acceleration (difference between the front and rear accelerations) to obtain a differential acceleration. The primary storage unit 20 stores a raw waveform of triaxial acceleration or differential acceleration as observation information.

  Based on the observation knowledge stored in the knowledge storage unit 30, the extraction unit 40 obtains the acceleration scalar amount A from the raw waveform of the XYZ triaxial acceleration stored in the primary storage unit 20 using the following equation (1). .

A = √ {(X _n −X _n−1 ) ² + (Y _n −Y _n−1 ) ² + (Z _n −Z _n−1 ) ² } (1)

Here, X, Y, and Z represent the X-axis, Y-axis, and Z-axis accelerations, respectively. That is, the acceleration scalar quantity A indicates the total scalar quantity of the acceleration of each axis generated between the previous sampling (n−1) and the current sampling n.

  The extraction unit 40 identifies whether the observation target person is in a stationary state or a walking state based on whether or not an acceleration change exceeding the threshold value occurs within a predetermined time with respect to the acceleration scalar quantity A. As a result, the extraction unit 40 detects the start of the walking of the observation target if a change in acceleration exceeding the first threshold occurs, and the end of the walking of the observation target if there is no change in acceleration exceeding the first threshold. it can. Furthermore, if the second threshold value that is larger than the first threshold value is exceeded, the extraction unit 40 can detect that the observation target person is in the traveling state.

  In this case, the extraction unit 40 calls the observation knowledge corresponding to the second walking detection method from the knowledge storage unit 30 in order to use the second walking detection method. In this second walking detection method, the observation knowledge stored in the knowledge storage unit 30 is that the identification behavior is related to “walking”, the acceleration scalar quantity A is calculated, and for this acceleration scalar quantity A, This is an application that detects whether or not an acceleration change exceeding the threshold value occurs within a predetermined time.

  Next, a method for recognizing the motion of the observation subject will be described.

  Based on the control of the control unit 90, the acceleration sensor of the motion information input unit 11 in the observation unit 10 generates the three-axis acceleration of the XYZ three-axis acceleration according to the stationary or movement of the observation subject. Detect the waveform.

  The observation unit 10 differentiates the detected raw waveform of the XYZ triaxial acceleration (difference between the front and rear accelerations) to obtain a differential acceleration. The primary storage unit 20 stores a raw waveform of triaxial acceleration or differential acceleration as observation information.

  Based on the observation knowledge stored in the knowledge storage unit 30, the extraction unit 40 uses the XYZ three-axis acceleration raw waveform stored in the primary storage unit 20 to move the observation subject's motion state (walking, running, vehicle, stationary Etc.), and the resting state is detected.

  Non-Patent Document 5 (K. Ouchi, et al. “LifeMinder: A wearable healthcare support system with timely instruction based on the user's context,” IEICE Transactions on Information and Systems, Vol. E87-D, No. .6, pp. 1361-1369, 2004., Non-Patent Document 6 (T. Iso, et al., Gait Analyzer based on a Cell Phone with a Single Threeaxis Accelerometer, Proceedings of MobileHCI'06, pp. 141-144 , 2006.), Non-Patent Document 7 (Ikeya et al., Moving State Estimation Method Using 3-Axis Accelerometer, IPSJ Research Report 2008-UBI-19 (14), pp.78-80,2008.) Use. Of course, the method is not limited to these.

  Thereby, the extraction unit 40 can recognize the operation state of the observation target person.

  In this case, the extraction unit 40 calls the observation knowledge corresponding to the motion recognition method from the knowledge storage unit 30 in order to use the motion recognition method. In this motion recognition method, the observation knowledge stored in the knowledge storage unit 30 is an application in which the identification behavior relates to, for example, “walking”, and represents the methods described in the non-patent documents 5 to 7. is there.

  A method for detecting the living sound of the observation target will be described.

  The living sound detection method performs sound scene clustering on the sound information input by the sound information input unit 12 of the observation unit 10 based on the observation knowledge stored in the knowledge storage unit 30. The “living sound” is a sound generated in the daily life of the observation subject, for example, a sound of a meal, a sound of a wash surface, a sound of a vacuum cleaner, a sound of cooking, and the like.

  First, based on the control of the control unit 90, the microphone of the sound information input unit 12 in the observation unit 10 inputs sound information of the observation target person.

  Next, the observation unit 10 samples the sound waveform at 22 kHz with respect to this sound information.

  Next, the observation unit 10 extracts a feature amount necessary for sound recognition from the sampled sound waveform. The primary storage unit 20 compresses the raw sound waveform, the extracted feature amount, or the raw waveform and stores it as observation information.

  Next, the extraction unit 40 divides the feature amount stored in the primary storage unit 20 into units of one second for matching with the sound source dictionary based on the observation knowledge stored in the knowledge storage unit 30. The sound source dictionary is generated for each living sound.

  Next, the extraction unit 40 performs matching with the sound source dictionary for each division unit of the feature amount, and selects one having a high likelihood.

  Next, the extraction unit 40 detects the start and end of a portion where the same sound source dictionary is arranged.

  Next, the observation unit 10 is cut at the start and end of the detection part, and the cut part becomes a scene in which a living sound to be acquired is generated.

  In this case, the extraction unit 40 calls the observation knowledge corresponding to the life sound detection method from the knowledge storage unit 30 in order to use the life sound detection method. In this life sound detection method, the observation knowledge stored in the knowledge storage unit 30 is that the identification behavior relates to, for example, “meal of life rhythm”, and is based on a sound source dictionary storing sound information at the time of meal. An application for scene clustering.

  The living behavior data extraction method described above will be described again based on the flowchart of FIG.

  In step S100, the observation unit 10 measures the acceleration of the observation subject using the acceleration sensor of the motion information input unit 11, and proceeds to step S101. It is assumed that the acceleration sensor continues to measure the acceleration of the observation subject in real time.

  In step S101, if the extraction unit 40 determines that the observation target person has started walking or running (hereinafter simply referred to as “walking”) based on the acceleration, the process proceeds to step S102 (in the case of Yes). If the start of walking cannot be detected, the process proceeds to step S110 (in the case of No).

  In step S102, the extraction unit 40 measures the number of steps and the walking distance of the observation target person, and proceeds to step S103.

  In step S103, the extraction unit 40 determines whether or not the observation target person has finished walking based on the current acceleration that is being measured, and if completed, the process proceeds to step S104 (in the case of Yes). If not completed, the process returns to step S102 (in the case of No).

  In step S <b> 104, the extraction unit 40 stores the walking start time, end time, number of steps, and walking distance in the secondary storage unit 50. Then, the process returns to step S101.

  In step S111, since the start of walking cannot be detected, the observation unit 10 inputs sound information using the microphone of the sound information input unit 12, and the extraction unit 40 performs acoustic scene clustering as described above to perform observation. It detects what kind of life sound the subject is currently making. If detection ends, the process proceeds to step S111.

  In step S111, the observation unit 10 uses an acceleration sensor to measure whether or not the observation target person has started walking from a stationary state. If the operation (for example, washing, eating, grooming, etc.) is completed, the process proceeds to step S112 (in the case of Yes), and if it is still stopped, the operation of the observation subject is not completed and the process returns to step S110 (No If).

  In step S111, since the observation subject has started walking, the extraction unit 40 stores, in the secondary storage unit 50, the identification behavior (for example, a wash surface, a meal, a dressing, etc.) corresponding to the life sound obtained by the acoustic scene clustering. Then, the process returns to step S102.

  As described above, the living behavior storage device 1 can detect and store the living behavior data by combining the start time and the end time with respect to the number of steps, the walking distance, and the identification behavior when the observation target starts walking.

  As described above, how the conversion unit 70 presents to the user the living behavior data of the observation target detected by the living behavior storage device 1 using the conversion rules stored in the rule storage unit 80. Will be described with reference to FIG. FIG. 4 shows the daily behavior data of the observation target for one week from February 20 to February 26 stored in the secondary storage unit 50 in time units. Total data is stored.

  Regarding ADL, the conversion unit 70 analyzes whether or not the observation target person has walked more than 45 m from the one-week life behavior data in FIG. 4 based on the ADL conversion rule in FIG. 3. On February 25, the observation subject walks 600 m in one cycle and exceeds 45 m in one cycle. Therefore, the conversion unit 70 calls the item “A walking is not difficult” in the presentation content of the ADL conversion rule in FIG. 3, and the presentation unit 60 gives the ADL user (for example, a caregiver) As shown in FIG. 7, a walking distance of 600 m and “not difficult to walk” are presented. Conversely, if the observation subject has not walked more than 45 m once a week, this “not difficult to walk” is not presented, so the ADL user can determine that the observation subject is difficult to walk. .

  The presentation of ADL described above will be further described based on the flowchart of FIG.

  In step S200, the living behavior storage unit 1 starts determination regarding ADL, and proceeds to step S201.

  In step S201, the conversion unit 70 detects whether or not the observation target person has a walking distance of 45 m or more in one cycle. If there is, the process proceeds to step S202 (in the case of Yes), and if not, the process proceeds to step S201. Proceed (if no).

  In step S202, since there was a walking distance of 45 m or more in one cycle, the presentation unit 60 presents that the observation target is not difficult to walk. Then, the process proceeds to step S203.

  In step S210, since the walking distance in one cycle is less than 45 m, the presentation unit 60 presents that the observation target is difficult to walk, and the process proceeds to step S203.

  In step S203, it is detected whether or not the observation subject has taken a meal three times a day. If it is three times or more, the process proceeds to step S204 (in the case of Yes). Proceed (if no).

  In step S211, the presentation unit 60 presents the date and the number of meals on the day when the number of meals of the observation subject is not three, and the process proceeds to step S204.

  In step S204, the conversion unit 70 proceeds to another ADL determination.

  By the above, the determination regarding ADL can be performed and the living behavior of the observation target can be surely presented to the user of ADL.

  With respect to the metabolic syndrome, the conversion unit 70 analyzes whether or not 10,000 steps per day are exceeded from the one-week life activity data of FIG. 4 based on the conversion rule of the metabolic syndrome in FIG. Since the total number of steps of the observation subject on February 20 is 5312 steps, it corresponds to less than 10,000 steps in the metabolic syndrome conversion rule of FIG. Therefore, the conversion unit 70 calls the item “target not achieved” in the metabolic syndrome conversion rule in FIG. 3, and the presentation unit 60 presents 5312 steps and “target not achieved”. On the other hand, since February 25 has exceeded 10,000 steps, 1004 steps are indicated as “achievement of target”.

  As described above, the living behavior storage device 1 according to the present embodiment can present necessary living behavior data and presentation contents for each different user regarding the living behavior data of the observation target person.

  A case where a new conversion rule is added to the conversion rule stored in the rule storage unit 80 shown in FIG. 3 will be described with reference to FIGS.

  In FIG. 8, a conversion rule relating to the watching user is added to the conversion rule of FIG. In this specification, the “watching user” is assumed to be a family living remotely. The added conversion rule is to take a photograph of the observation subject's meal and a photograph of the observation subject's own meal scenery in the item “presentation content”. The control unit 90 is used.

  By adding a new conversion rule to the rule storage unit 80, the control unit 90 uses the camera of the image acquisition unit 13 of the observation unit 10 to detect an application for taking a photograph of the observation subject at the time of meal. Add to part 30.

  As a result, the contents of the living behavior data stored in the secondary storage unit 50 by the extraction unit 40 are changed as shown in FIGS. That is, when the observation subject's meal is detected, not only the meal photograph is taken, but also the observation subject's own photograph is taken. Then, the presentation unit 60 presents the observation user with a photograph of the observation subject in addition to the meal photograph. Thereby, the watching user can know whether or not the observation target person is fine by this presentation.

  As described above, the living behavior storage device 1 according to the present embodiment can cope with an increase in the number of new users by adding conversion rules to the rule storage unit 80.

  Next, the living behavior storage device 1 according to the second embodiment of the present invention will be described with reference to FIGS.

  FIG. 10 is a block diagram of the living behavior storage device 1 of the present embodiment. The difference between the living behavior storage device 1 of the present embodiment and the living behavior storage device 1 of the first embodiment is that a communication unit 70 is added and a plurality of presentation units 61, 62, and 63 are provided.

  In the first embodiment, conversion rules and display contents differ for each user, but in this embodiment, the control unit 90 controls in real time whether or not each presentation unit 61, 62, 63 is contacted to each user. it can.

  FIG. 8 shows an example of conversion rules corresponding to this embodiment.

  “Fall” is added as the identification action item, and the fall user is added as the user type. In this case, the fallen user is assumed to be a home security center monitor.

  The control unit 90 controls the observation unit 10 and the knowledge storage unit 30 so as to detect the fall of the observation target person as life activity data. As this detection method, for example, the detection is performed from the value of the acceleration sensor and the image of the observation subject photographed by the camera.

  When the fall of the observation subject is detected as the living behavior data, the watching user notifies that the fall has occurred in real time regardless of whether the observation subject has reacted after the fall. That is, when the fall of the observation subject is detected, the control unit 90 transmits a notification that the observation subject has fallen to the presentation unit 61 of the watching user via the communication unit 70.

  On the other hand, when the fall user detects the fall of the observation target, the user of the fall makes an inquiry to the observation target via the presentation unit 62 to confirm whether the observation target is conscious. If there is no response from the fallen observation subject, the fact is notified in real time to the presentation unit 63 of the fallen user. The user of the fall who received the communication arranges an ambulance. On the other hand, if there is a response from the observation subject and it can be confirmed that the observation subject can move, the number of falls is stored in the secondary storage unit 50, and the presentation unit 61 presents the number of falls to the user.

  According to the present embodiment, the living behavior of the person to be observed can be acquired in more detail by presenting the necessary living behavior data and the presentation content to the presentation unit for each user.

Example of change

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

  DESCRIPTION OF SYMBOLS 1 ... Living action storage device, 10 ... Observation part, 20 ... Primary storage part, 30 ... Knowledge storage part, 40 ... Extraction part, 50 ... Secondary storage part, 60 ... Presentation unit, 70 ... Conversion unit, 80 ... Rule storage unit, 90 ... Control unit

Claims (5)

An observation unit for observing the observation target with a plurality of types of sensors and obtaining observation information for each sensor ;
A primary storage unit that stores the obtained observation information for each of the sensors ;
A knowledge storage unit that stores in advance observation knowledge configured from an application for obtaining necessary living behavior data for each type of user from necessary observation information in the observation information for each sensor ;
An application for the observation knowledge , and an extraction unit for extracting necessary living behavior data for each type of the user using the necessary observation information from the observation information stored in the primary storage unit ;
A secondary storage unit for storing the extracted life behavior data;
A rule storage unit that stores in advance conversion rules for converting from the living behavior data to the required presentation content for each type of the user;
Based on the conversion rule, a conversion unit that converts the living behavior data stored in the secondary storage unit into the presentation content;
In accordance with the type of the user, a presentation unit that presents the living behavior data regarding the observation target and the presentation content ;
Equipped with,
The observation unit includes an acceleration sensor and another sensor other than the acceleration sensor, and the observation unit detects the observation person's acceleration as the observation information using the acceleration sensor,
The knowledge storage unit previously stores an application for obtaining a walking distance or the number of steps from the acceleration as observation knowledge,
The extraction unit extracts the walking distance of the person to be observed or the number of steps as living behavior data based on the application of the observation knowledge and the acceleration,
The conversion unit stores in advance a conversion rule for converting into presentation content related to walking based on walking distance or number of steps.
In addition to the living behavior data and the presentation content based on the acceleration sensor, the presentation unit also presents the living behavior data and the presentation content based on the other sensors.
A living behavior storage device characterized by that. In addition to the acceleration sensor, the sensor has a microphone or a camera.
  The living activity storage device according to claim 1, wherein: The type of the user is ADL, IADL, ICF, or metabolic syndrome.
  The living behavior storage device according to claim 1 or 2, characterized by the above. For each different user, further comprising a communication unit that transmits the living behavior data and the presentation content,
The living behavior storage device according to any one of claims 1 to 3, wherein: The conversion rule stored in the rule conversion unit is presented for each type of user when the walking distance or the number of steps exceeds a threshold set for each type of user. Each of the presentation contents is described,
The conversion unit compares the walking distance or the number of steps with the threshold value for each type of user, and the walking distance or the number of steps exceeds the threshold value for each type of user. If so, call the presentation content corresponding to the type of user exceeding the threshold from the conversion rule,
The living behavior storage device according to any one of claims 1 to 4, wherein:

Images