CN110136832B - Cognitive ability assessment system and method based on daily behaviors of old people - Google Patents

Cognitive ability assessment system and method based on daily behaviors of old people Download PDF

Info

Publication number
CN110136832B
CN110136832B CN201910421469.5A CN201910421469A CN110136832B CN 110136832 B CN110136832 B CN 110136832B CN 201910421469 A CN201910421469 A CN 201910421469A CN 110136832 B CN110136832 B CN 110136832B
Authority
CN
China
Prior art keywords
user
server
behavior
cognitive ability
client
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910421469.5A
Other languages
Chinese (zh)
Other versions
CN110136832A (en
Inventor
安宁
贵芳
孙传能
王雯云
杨矫云
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hefei University of Technology
Original Assignee
Hefei University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hefei University of Technology filed Critical Hefei University of Technology
Priority to CN202210356940.9A priority Critical patent/CN114724714A/en
Priority to CN202210362582.2A priority patent/CN114974577A/en
Priority to CN201910421469.5A priority patent/CN110136832B/en
Publication of CN110136832A publication Critical patent/CN110136832A/en
Application granted granted Critical
Publication of CN110136832B publication Critical patent/CN110136832B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
    • G01B21/22—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring angles or tapers; for testing the alignment of axes
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
    • G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

The invention relates to a cognitive competence evaluation system and method based on the daily behaviors of the old, the system comprises a server and a plurality of behavior sensors, wherein the server acquires the behavior data of the daily behaviors related to a user in an implicit perception mode through the behavior sensors deployed in the living space of the user; the server evaluates the cognitive ability of the user at least according to the behavior data of the daily behaviors associated with the user to obtain the cognitive ability score of the user.

Description

Cognitive ability assessment system and method based on daily behaviors of old people
Technical Field
The invention relates to the field of database construction, relates to machine learning in the field of artificial intelligence, and particularly relates to a cognitive ability assessment system and method based on the daily behaviors of old people.
Background
At present, the cognitive ability of the old people is evaluated mainly by a cognitive disorder screening scale, and the old people are evaluated by workers, so that the cognitive ability of the old people is evaluated. However, the traditional scale evaluation has the following problems that the return visit time interval is long, and the regular return visit is difficult to realize; the evaluation time is long, the requirement on workers is high, and the wide-range popularization is difficult to realize; the old people with low education level are difficult to complete most of evaluation contents, and the like.
The invention patent CN106327049A designs a cognition evaluation system, which comprises an information module, a test module and an analysis module; the information module is used for acquiring medical information matched with the test module according to the data of the object and establishing a complete cognitive assessment database; the test module obtains cognitive test data of the object through testing, and comprises the following five sub-modules: the system comprises an attention and execution function testing module, a memory testing module, a mathematics and computing capability testing module, a language testing module and an action and behavior control and plan testing module; the analysis module determines the cognitive evaluation result of the object according to the medical information acquired by the information module and the cognitive test data acquired by the test module. However, the method still requires the staff to perform measurement to realize cognitive assessment, has high requirements on the staff and is difficult to realize automation. Furthermore, the cognitive decline of the elderly is a slow and inconspicuous process, the cognitive score obtained by evaluating the cognitive disorder screening scale for one or two times cannot well reflect the change of the cognitive ability of the elderly, and as mentioned above, the cognitive ability change obtained by frequently evaluating the cognitive disorder screening scale is difficult to operate because the cognitive disorder screening scale evaluating process has the problems of long time consumption, high requirement on workers, difficulty in completing most evaluation contents of the elderly with low education level and the like.
Therefore, there is a need for improvements in the prior art.
Moreover, on the one hand, since the skilled person in the art who is understood by the applicant is necessarily different from the examination department; on the other hand, since the inventor made the present invention while studying a large number of documents and patents, the disclosure should not be limited to the details and contents listed in the specification, but the present invention should not have the features of the prior art, but the present invention should have the features of the prior art, and the applicant reserves the right to increase the related art in the background art at any time according to the related specification of the examination guideline.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a cognitive ability assessment system and method based on the daily behaviors of the old. Because the cognitive ability of the user is reduced in the daily behaviors of the old, the invention realizes the long-term monitoring of the behavior data of the old through the sensing technology, and further realizes the evaluation of the cognitive ability of the old. The cognitive ability of the user is not required to be frequently evaluated through a cognitive disorder screening scale, the disturbance to the daily life of the user is reduced, and the labor intensity of medical staff is reduced.
According to a preferred embodiment, the cognitive ability assessment system based on the daily behaviors of the old comprises a server and a plurality of behavior sensors, wherein the server acquires behavior data of the daily behaviors related to a user in an implicit perception mode through the behavior sensors deployed in the living space of the user; the server evaluates the cognitive ability of the user according to at least the behavior data of the daily behaviors associated with the user to obtain the cognitive ability score of the user.
According to a preferred embodiment, the server periodically compares the cognitive ability score of the user corresponding to the current period with the cognitive ability score of the user corresponding to the previous period, and generates first early warning information when the cognitive ability score cycle ratio of the user decreases to exceed a preset trigger threshold value.
According to a preferred embodiment, the server evaluates the cognitive ability of the user according to the following scoring formula to obtain the cognitive ability score of the user:
Figure BDA0002065091780000021
before evaluating the cognitive ability of the user to obtain the cognitive ability score of the user, the server selects behavior data corresponding to at most n behaviors triggered daily as a standard task X for evaluating the cognitive ability of the user, wherein the standard task X is expressed as X ═ { X ═ X1,X2,…,XnT is a set of standard times for the average person to complete each of the n tasks, and is expressed as T ═ T1,T2,…,Tn}，TiIs the standard time for ordinary people to complete the ith task in the n tasks, and t is the set of the average time for the user to complete each task in the n tasks, and is expressed as t ═ t1,t2,…,tn}，tiIs the user completing the first of the n tasksAverage time of i tasks, alpha is the set of task completion degrees of each task in the n tasks completed by the user, and alpha isiRepresenting the task completion degree of the ith task in the n tasks, w is the set of weights of all the tasks in the n tasks, w isiAnd representing the weight of the ith task in the n tasks.
According to a preferred embodiment, the server collects behavior data of daily behaviors associated with the user and has a time attribute, the server acquires the behavior data of the daily behaviors associated with the user and stores the behavior data of the daily behaviors associated with the user into the time sequence database according to the time attribute, the server independently establishes a hidden markov model associated with each user for predicting the behavior of each user, the server trains the hidden markov model associated with the user by using the behavior data associated with the user in the time sequence database, the server predicts the predicted behavior of the user according to the hidden markov model associated with the user, and generates second warning information if the actual behavior of the user deviates from the predicted behavior and the deviation leads to a known risk and/or a known loss.
According to a preferred embodiment, the system further includes a first client, the first client is communicatively connected to the server, the first client is used as a relay device to acquire behavior data acquired by the behavior sensors arranged in the living space, the server generates second warning information and sends a risk verification request to the first client, the first client sends an alarm in response to the risk verification request, the alarm can be released only after the user verifies at least two biological feature identifications on the first client, the first client releases the alarm and sends risk false alarm feedback or risk release feedback to the server, and the server deletes the second warning information in response to the risk false alarm feedback or risk release feedback.
According to a preferred embodiment, the system further comprises a camera, the camera being communicatively connected to the first client, the camera device can be arranged outside an access door for accessing the living space and is used for collecting the first photo group and the second photo group, the first photo group is shot before the access door is opened, the second photo group is shot in a period which lasts for a preset time after the access door is opened until the access door is closed, the first client acquires a first photo group and a second photo group from the camera device, compares the first photo group with the second photo group to determine the personnel change state in the living space, and the first client determines an accommodation mode of persons accommodated in the residential space in response to a person change state in the residential space, the accommodation mode including one of: a user-independent mode that only one user exists in the living space; visitor patterns with other people except the user or only other people in the living space; and an empty mode in which no one is present in the living space; wherein the first client analyzes triggers of the behavior data collected by the behavior sensors during the guest mode, intercepts the corresponding behavior data of the undeterminable triggers if the triggers of the corresponding behavior data collected by the behavior sensors during the guest mode cannot be determined, and only transmits the behavior data collected by the behavior sensors during the guest mode, which can determine the triggers as users, to the server. Preferably, the length of the preset time may be set by a manufacturer and/or a user. For example, the preset time may be 1-5 s. Specifically, for example, the preset time may be 2s, 3s, or 4 s. That is, assuming that the preset time is 2s, the second photo group includes photos taken during a period from when the access door is opened to 2s after the access door is closed.
According to a preferred embodiment, the sensors at least comprise an angular displacement sensor which is installed on the access door and used for identifying the open-close state of the access door, the angular displacement sensor used for identifying the open-close state of the access door is in communication connection with the camera device, the angular displacement sensor used for identifying the open-close state of the access door is configured to immediately and actively send an electric signal indicating that the access door is opened to the camera device when the access door rotates 3-5 degrees towards the opening direction, the camera device receives the electric signal indicating that the access door is opened and then takes a plurality of photos within 1s as a first photo group, the angular displacement sensor used for identifying the open-close state of the access door is configured to not send the electric signal indicating that the access door is closed to the camera device when the access door rotates towards the closing direction to completely close the access door, and the camera device continuously takes a plurality of photos after the first photo group is finished until the access door is closed And the period lasting for the preset time after the electric signal is taken as a second picture group.
According to a preferred embodiment, the system further comprises a smart band and an identification module, the server assigns a unique identification to each user and stores the identification in the smart band, the intelligent hand ring sends electromagnetic waves carrying identity marks to a limited communication range, each behavior sensor is provided with an identity recognition module for recognizing the electromagnetic waves carrying the identity marks, to detect the smart band worn by the user and thereby identify the trigger of the behavior data collected by the behavior sensor, when the first client determines that the accommodation mode of the persons accommodated in the residential space is the user-solitary mode, the first client receives the behavior data acquired by the behavior sensor and defines a trigger acquired by the identity recognition module when the situation that the user does not wear the smart band as the user in a default mode; when the first client responds to the change state of the personnel in the living space to determine that the accommodation mode of the personnel accommodated in the living space is changed into a visitor mode from a user-independent mode, the first client instructs the user to wear the smart band, receives the behavior data collected by the behavior sensor and only determines the behavior data collected by the identity recognition module under the condition of detecting the smart band worn by the user as the behavior data of the user as the trigger.
According to a preferred embodiment, a predetermined behavior sequence and a predetermined time length of a specific behavior according with a user behavior habit are prestored in the server, after first early warning information is generated by the server, the server analyzes the accuracy of the comparison between the behavior sequence of the user in the current period and the predetermined behavior sequence according to received behavior data of daily behaviors associated with the user to obtain a first confidence degree, the server analyzes a second confidence degree of the comparison between the time length of the corresponding specific behavior of the user in the current period and the time length of the predetermined specific behavior according to the received behavior data of the daily behaviors associated with the user, a preset first confidence degree threshold value and a second confidence degree threshold value are stored in the server, the first confidence degree threshold value is larger than the second confidence degree threshold value, and the server judges that the first early warning information is invalid when the first confidence degree is larger than the first confidence degree threshold value and the second confidence degree is larger than the second confidence degree threshold value The server directly confirms that the first early warning information is valid when the first confidence degree is smaller than or equal to a first confidence degree threshold value and the second confidence degree is smaller than or equal to a second confidence degree threshold value, the server marks the first early warning information as the first early warning information to be verified when one condition of the first confidence degree smaller than or equal to the first confidence degree threshold value and the second confidence degree smaller than or equal to the second confidence degree threshold value is met, the server waits for the behavior data of the daily behavior associated with the user in the next period sent by the first client side to verify the first early warning information to be verified, the server confirms that the first early warning information is valid only when the cognitive ability score of the user corresponding to the next period is lower than the cognitive ability score of the corresponding user evaluated in the previous two periods in the next period and exceeds a preset trigger threshold value, and the server sends the first early warning information to other persons except the user when the server confirms that the first early warning information is valid There is a second client.
According to a preferred embodiment, a cognitive ability assessment method based on daily behaviors of the elderly uses the system to assess the cognitive ability of a user, wherein the system comprises a server and a plurality of behavior sensors, the server collects behavior data of the daily behaviors related to the user in an implicit perception mode through the behavior sensors deployed in the living space of the user; the server evaluates the cognitive ability of the user according to at least the behavior data of the daily behaviors associated with the user to obtain the cognitive ability score of the user.
Drawings
FIG. 1 is a block schematic diagram of a preferred embodiment of the system of the present invention.
List of reference numerals
100: the first client 200: the second client 300: server
400: the behavior sensor 500: the image pickup apparatus 600: intelligent hand ring
700: identity recognition module
Detailed Description
The following detailed description is made with reference to the accompanying drawings.
First, some terms used in the present invention are defined:
implicit perception may refer to behavior sensors coupled into or onto items of a living space to perceive a user's behavior as the user touches or manipulates the items to obtain behavioral data. In other words, implicit perception is the perception process of implicitly acquiring behavioral data of a user through a behavioral sensor coupled to an item in a residential space that may be triggered by the daily behavior of the user during the user's daily behavior. This makes the implicit sensing different from the user actively testing, and the user does not need to deliberately trigger the behavior sensor during the implicit sensing, but the behavior sensor is triggered by the way in the process of completing the daily behavior to obtain the corresponding behavior data. The setting mode and the measuring mode of the behavior sensor do not disturb the daily behaviors of the user. That is, a number of behavior sensors are disposed in the user's living space in a manner that does not disturb the user, collecting behavior data for daily behavior associated with the user. The items may include, for example, at least one of various types of doors, various types of switches, and various types of appliances. Specifically, the article may include at least one of an interior door, a window door, an access door, a refrigerator door, a microwave oven door, a wardrobe door, a faucet switch, a light switch, a cooktop switch, a power switch, a wash cup, and a wash basin.
The time series database may refer to a time series database. The time sequence database is mainly used for processing data with time tags (changed according to time sequence, namely time sequence) or time attributes. Data with time tags or time attributes are also referred to as time series data.
Example 1
The embodiment discloses a cognitive ability assessment system based on the daily behaviors of the old, or a cognitive ability assessment system based on the daily behaviors of users, or a system for assessing cognitive ability based on a time sequence database, or a system for assessing cognitive ability. The system is adapted to perform the various method steps recited in the present invention to achieve the desired technical effect. The preferred embodiments of the present invention are described in whole and/or in part in the context of other embodiments, which can supplement the present embodiment, without resulting in conflict or inconsistency.
According to a preferred embodiment, referring to fig. 1, the system may comprise a server 300 and/or several behavior sensors 400. The server 300 may collect behavior data for daily behavior associated with a user via a number of behavior sensors 400 disposed or deployed in the user's living space. Alternatively, the server 300 may collect behavior data of daily behaviors associated with the user in an implicit manner through several behavior sensors 400 deployed within the user's living space. Preferably, the behavior sensor 400 may include at least one of an angle sensor, a door magnetic switch sensor, a laser sensor, a water flow sensor, a micro switch sensor, a pressure sensor, a temperature sensor, a humidity sensor, a smart switch, and an angle sensor. The server 300 may evaluate the cognitive abilities of the user based on the behavioral data of the daily behaviors associated with the user to obtain a cognitive ability score of the user. Preferably, the server 300 collects behavior data of daily behaviors associated with the user in an implicit sensing manner through several behavior sensors 400 disposed in the user's living space. Preferably, the living space refers to a living environment of the user. For example, the living space may include at least one of a bedroom, a living room, a kitchen, and a toilet. The invention can at least realize the following beneficial technical effects by adopting the mode: firstly, the disturbance of frequent assessment of cognitive abilities on the daily behavior of the user can be reduced; secondly, the difficulty of continuously acquiring the cognitive ability scores of the users is reduced, so that the cognitive ability changes of the users can be observed for a long time.
According to a preferred embodiment, the server 300 may periodically compare the cognitive performance score of the user corresponding to the current period with the cognitive performance score of the user corresponding to the previous period. The server 300 may generate the warning information when the cognitive ability score ratio of the user decreases beyond a preset trigger threshold. Preferably, the server 300 may transmit the first warning information to the second client. Preferably, the comparison period for the server 300 to periodically compare the cognitive performance score of the user corresponding to the current period with the cognitive performance score of the user corresponding to the previous period may be set manually. Preferably, the preset trigger threshold may be set manually. For example, the preset trigger threshold may be 5% to 60%, for example. That is, assuming that the server 300 compares the cognitive ability scores of the users in a period of days, the preset trigger threshold is set to 10%. The cognitive ability score of the user at the current day is 50 points, the cognitive ability score of the user at the current day is 38 points, the cognitive ability score ring ratio of the user is reduced by 24%, and the cognitive ability score ring ratio exceeds 10%, and the server 300 generates early warning information. Preferably, the comparison period in which the server 300 periodically compares the cognitive performance score of the user corresponding to the current period with the cognitive performance score of the user corresponding to the previous period may have a plurality of periods having different period lengths from each other. For example, the server 300 may be configured to cycle by at least one of day, week, month, quarter, and year. For example, the server 300 may be configured to compare cognitive performance scores of corresponding users in a day period. As another example, the server 300 may be configured to compare cognitive performance scores of corresponding users on a daily and weekly basis. That is, the server 300 compares the cognitive ability scores of the users in the two adjacent days and compares the cognitive ability scores of the users in the two adjacent weeks. Preferably, the server 300 may periodically compare the cognitive performance score of the user corresponding to the current period with the cognitive performance score of the user corresponding to the previous period in at least two comparison periods different from each other. The server 300 may generate the warning information when the cognitive ability score ratio of the user decreases beyond a preset trigger threshold. The different comparison periods may correspond to different preset trigger thresholds from each other. The preset trigger threshold may be smaller for comparison periods having longer period lengths. For example, when the server 300 compares the cognitive performance scores of the corresponding users in two adjacent days, the corresponding preset trigger threshold may be 15%. When the server 300 compares the cognitive ability scores of the corresponding users in two adjacent weeks, the corresponding preset trigger threshold may be 10%. Preferably, the period length may refer to a time duration included in one period. The invention can at least realize the following beneficial technical effects by adopting the mode: firstly, the server 300 regularly processes and compares the behavior data of the user collected by the sensor to feed back the change of the cognitive ability of the user, and the server 300 generates early warning information when the cognitive ability score ratio of the user decreases to exceed a preset trigger threshold value so as to know that the cognitive ability level of the user may rapidly decrease; secondly, if the cognitive ability of the user is continuously evaluated every day by adopting the existing cognitive disorder screening scale, the cognitive ability is almost impossible to realize, but the cognitive ability of the user is evaluated by utilizing rich user behavior data acquired every day by a sensor, and the change of the cognitive ability of the user in periods with different period lengths can be obtained and early warning is timely given out in a mode of periodic comparison in a plurality of periods; third, in general, assuming that the cognitive ability of the user slowly decreases with time, the larger the time span is, the more the cognitive ability should be decreased, so the conventional manner is that the preset trigger threshold corresponding to the comparison period with the longer period length is larger, but the inventor of the present invention thinks that, as the larger the period length is, the larger the data amount is, the more real a cognitive ability score is reflected, and the smaller the error is, therefore, the present invention breaks the convention, and sets the preset trigger threshold corresponding to the comparison period with the longer period length in the server 300 to be smaller, for example, when the server 300 compares the cognitive ability scores of the corresponding users in two adjacent days, the corresponding preset trigger threshold is 19%, when the server 300 compares the cognitive ability scores of the corresponding users in two adjacent weeks, the corresponding preset trigger threshold is 16%, when the server 300 compares the cognitive ability scores of the corresponding users in two adjacent seasons, the corresponding preset trigger threshold is 13%, and when the server 300 compares the cognitive ability scores of the corresponding users in two adjacent seasons, the corresponding preset trigger threshold is 10%, so that the system false alarm is reduced, and the early warning information is more accurate.
According to a preferred embodiment, the system may include at least one of a first client, a smart band 600, and an identification module 700. The server 300 may assign a unique identification to each user and store the identification in the smart band 600. The smart band 600 may be carried around by a user. The smart band 600 may transmit electromagnetic waves carrying the identification to a limited communication range. Each behavior sensor 400 may be equipped with an identification module 700 to identify the associated user of the behavior data collected by the behavior sensor 400. When the smart band 600 moves along with the user so that the communication range thereof covers the corresponding identification module 700, the corresponding identification module 700 identifies the identity of the user related to the behavior data collected by the behavior sensor 400 according to the obtained electromagnetic wave with the identity and generates information with the time attribute. The respective first client is associated with a user having a respective identity.
According to a preferred embodiment, the system may comprise several cameras. Several cameras may be provided around the perimeter of the doors that access the user's living space. The system can identify the condition of people in the living space according to the images collected by the cameras. The system may ignore the affected behavioral data in the event that the behavioral data has an impact by someone other than the user.
According to a preferred embodiment, the server 300 may evaluate the cognitive ability of the user according to the following scoring formula to obtain the cognitive ability score of the user:
Figure BDA0002065091780000091
preferably, the server 300 may select behavior data corresponding to at most n behaviors triggered daily as a standard task X for evaluating the cognitive ability of the user, denoted as X, before evaluating the cognitive ability of the user to obtain a cognitive ability score of the user＝{X1,X2,…,Xn}. T may be a set of standard times for an average person to complete each of the n tasks, denoted as T ═ T1,T2,…,Tn}。TiIt may be a standard time for an ordinary person to complete the ith task among the n tasks. t may be a set of average times for the user to complete each of the n tasks, denoted as t ═ t1,t2,…,tn}。tiMay be the average time that the user completed the ith task of the n tasks. Alpha may be a set of task completions for the user to complete each of the n tasks. Alpha is alphaiThe task completion degree of the ith task in the n tasks completed by the user can be represented. w may be a set of weights for each of the n tasks. w is aiThe weight of the ith task of the n tasks may be represented. For example, after the user opens the door, the door is not closed for a certain time, and at this time, it is very likely that the user forgets to close the door, the task completion degree of the door opening and closing event is reduced, and the cognitive ability score calculated by the cognitive ability scoring formula will be reduced accordingly. For another example, if the user turns off the water for a long time in the process of washing the dishes, the event of completing the dish washing event is far longer than the normal time for completing the dish washing, and it is highly likely that the user forgets to turn off the water, if the event of turning on and off the water faucet is one of the at most n behaviors triggered by the user daily, the event is adopted to evaluate the cognitive ability of the user, and the calculated cognitive ability score is also reduced. According to the method and the device, the cognitive ability of the user is evaluated by adopting the at most n behaviors which are daily triggered by the user, so that excessive interference caused by a few and accidental data on the evaluation of the cognitive ability of the user due to the fact that the source data used as the evaluation is too little can be reduced. Preferably, different users have different habits, for example, some users may be accustomed to opening an access door for ventilation. For another example, some users may be accustomed to opening the door of a microwave oven for long periods of time without using the microwave oven to dispense a smell. When the invention is used, the user can be explained firstlyThe cognitive ability scoring rule provided by the invention enables a user to avoid the event that the scoring is greatly deviated due to personal habits caused by not knowing the scoring rule. After the user knows the rules, the user can not perform some actions at some behavior moments to obtain higher cognitive ability scores and can also embody the cognitive ability of the user. Preferably, n may be an integer of 1 or more. n may be set by human, for example, to 1,2, 3, or 4. When setting, the number of the behavior sensors arranged in the living space of the user can be determined, and the more the behavior sensors are, the larger the value of n can be set.
According to a preferred embodiment, the behavior data collected by the server 300 for the daily behavior associated with the user may be accompanied by a time attribute. After obtaining the behavior data of the daily behavior associated with the user, the server 300 may store the behavior data of the daily behavior associated with the user in the time-series database according to the time attribute. The server 300 may build a hidden markov model associated with each user independently for predicting their behavior. The server 300 may train a hidden markov model associated with the user using behavioral data relating to the user in a time series database. The server 300 may predict the predicted behavior of the user based on a hidden markov model associated with the user. The server may generate the second warning information in the event that the actual behavior of the user deviates from the predicted behavior and the deviation would result in a known risk and/or a known loss.
According to a preferred embodiment, the system may include a first client 100, the first client 100 being communicatively connected to a server 300. The first client 100 may acquire behavior data collected by a number of behavior sensors 400 disposed within the residential space as a relay device. The server 300 may transmit a risk verification request to the first client 100 after generating the second warning information. The first client 100 may issue an alert in response to the risk verification request. The alert can only be dismissed by the user after verification of at least two biometrics on the first client 100. The first client 100 may send risk false positive feedback or risk removal feedback to the server 300 after de-alerting. The server 300 may delete the second warning information in response to the risk false positive feedback or the risk cancellation feedback.
According to a preferred embodiment, the system may include a camera 500. The camera 500 may be communicatively connected to the first client 100. The image pickup device 500 may be installed outside an entrance door that enters and exits the living space. The camera device 500 may be used to capture a first group of pictures and a second group of pictures. The first photo group may be taken before the access door is opened. The second photo group may be taken during a preset time after the access door is opened until the access door is closed. The first client 100 may acquire the first and second picture groups from the camera 500. The first client 100 may compare the first and second groups of pictures to determine a person change status within the residential space. The first client 100 may determine an accommodation mode of the persons accommodated in the residential space in response to the person change status in the residential space. The accommodation mode may include one of the following modes: a user-independent mode that only one user exists in the living space; visitor patterns with other people except the user or only other people in the living space; and a vacant mode in which no one is present in the living space. The first client 100 may analyze triggers for behavior data collected by several behavior sensors 400 during guest mode. The first client 100 may intercept the trigger-indeterminable corresponding behavior data collected during the guest mode if the trigger of the trigger-indeterminable corresponding behavior data is indeterminate, and only send the trigger-determinable behavior data collected during the guest mode by the number of behavior sensors 400 to the server 300.
Preferably, the length of the preset time may be set by a manufacturer and/or a user. For example, the preset time may be 1-5 s. Specifically, for example, the preset time may be 2s, 3s, or 4 s. That is, assuming that the preset time is 2s, the second photo group includes photos taken during a period from when the access door is opened to 2s after the access door is closed. The invention can at least realize the following beneficial technical effects by adopting the mode: first, in the prior art, the camera device 500 is generally arranged outside the door to collect the characteristics of the person through the camera device 500 to perform characteristic recognition so as to determine whether the corresponding person has the right to enter the door, but the invention does not relate to the invention for determining the accommodation mode of the person in the living space. Although the old people can operate on the first client 100 to remove unreal behavior data, the old people can operate the intelligent device unfailingly and difficultly, so that the system can be simplified and the operation steps of the old people can be reduced to the greatest extent in order to simplify the disturbance of the system to the old people; second, this camera device 500 sets up outdoors, can not gather the relevant photo of old man's life privacy, avoids the old man to appear conflicting mood.
According to a preferred embodiment, the plurality of behavior sensors may include an angular displacement sensor mounted on the access door for recognizing an open and closed state of the access door. An angular displacement sensor for recognizing the state of the door opening and closing may be communicatively connected to the camera 500. The angular displacement sensor for recognizing the open and close state of the access door may be configured to actively transmit an electric signal indicating that the access door is opened to the image pickup device 500 immediately when the access door is rotated in the opening direction by 3 ° to 5 °. The image capturing device 500 can capture a plurality of pictures as a first group of pictures within 1s after receiving the electric signal that the access door is opened. The angular displacement sensor for recognizing the open and close state of the access door may be configured to transmit an electrical signal indicating that the access door is closed to the image pickup device 500 only when the access door is rotated in a closing direction until the access door is completely closed. The image pickup device 500 may continuously take a plurality of photographs after the first photograph group is taken until a preset time period elapses after an electric signal for closing the door is received as the second photograph group. The invention can at least realize the following beneficial technical effects by adopting the mode: the angular displacement sensor for recognizing the opening and closing state of the access door is in communication connection with the camera device 500, and the time of the first photo group and the time of the second photo group acquired by the camera device 500 are related to the opening and closing state of the access door, so that the first photo group and the second photo group shot by the invention are related to the change state of people in the living space with high probability, and excessive useless photos cannot be shot, for example, photos before the access door is passed by passersby when the access door is not opened, the calculation amount of the first client 100 can be reduced, and the first client 100 can efficiently and accurately analyze the accommodating mode of the people accommodated in the living space.
According to a preferred embodiment, the system may comprise a smart band 600 and/or an identification module 700. The server 300 may assign a unique identification to each user and store the identification in the smart band 600. The smart band 600 may transmit electromagnetic waves carrying the identification to a limited communication range. Each behavior sensor 400 may be equipped with an identification module 700 for recognizing electromagnetic waves carrying identification to detect the smart band 600 worn by the user, thereby recognizing a trigger of behavior data collected by the behavior sensor 400. When the first client 100 determines that the accommodation mode of the people accommodated in the living space is the user-independent mode, the first client 100 may receive the behavior data collected by the behavior sensor 400 and define, as the user, a trigger that the identification module 700 collects the behavior data without detecting that the user does not wear the smart band 600 by default.
When the first client 100 determines that the accommodation mode of the persons accommodated in the residential space is changed from the user-exclusive mode to the guest mode in response to the state of change of the persons in the residential space, the first client 100 may instruct the user to wear the smart band 600. When the first client 100 determines that the accommodation mode of the persons accommodated in the residential space is changed from the user-independent mode to the guest mode in response to the state of change of the persons in the residential space, the first client 100 may receive the behavior data collected by the behavior sensor 400 and determine only the behavior data collected by the identification module 700 in case of detecting the smart band 600 worn by the user as the behavior data of which the trigger is the user. The invention can at least realize the following beneficial technical effects by adopting the mode: firstly, in the user's solitary mode, the user may not carry the smart band 600, making the user more free; secondly, in the life process of the user, due to various reasons such as the door crossing of relatives and friends, maintenance of equipment by maintenance personnel or the table lookup of water, electricity and gas companies and the like, it is inevitable that other people enter the living space, and in the visitor mode, the user can carry the smart bracelet 600 to still only acquire behavior data triggered by the user to evaluate the cognitive ability of the user under the condition that other people are in the living space, so that the situation that the acquired behavior data triggered by other people falsely evaluate the cognitive ability of the user is avoided, and the accuracy of the cognitive ability evaluation of the invention is improved; third, in the case of the guest mode, false alarms of the first and/or second pre-warning information due to the user and the guest triggering different behavior sensors 400 at the same time are reduced.
According to a preferred embodiment, the server 300 may have a predetermined sequence of behaviors and a predetermined length of time for a specific behavior according to the behavior habit of the user pre-stored therein. After the server 300 generates the first warning information, the server 300 may analyze the accuracy of the behavior sequence of the user in the current period compared with the predetermined behavior sequence according to the received behavior data of the daily behavior associated with the user to obtain a first confidence level. The server 300 may analyze a second confidence level of the time length of the corresponding specific behavior of the user in the current period compared with the predetermined time length of the specific behavior according to the received behavior data of the daily behavior associated with the user. The server 300 may have a first confidence threshold and a second confidence threshold stored therein. The first confidence threshold may be greater than the second confidence threshold. The server 300 may determine that the first warning information is invalid when the first confidence is greater than the first confidence threshold and the second confidence is greater than the second confidence threshold. The server 300 may directly confirm that the first warning information is valid when the first confidence degree is less than or equal to the first confidence degree threshold and the second confidence degree is less than or equal to the second confidence degree threshold. The server 300 may mark the first warning information as the first warning information to be verified when one of the first confidence degree is less than or equal to the first confidence degree threshold and the second confidence degree is less than or equal to the second confidence degree threshold is satisfied. The server 300 may wait for the next period of the daily behavior data associated with the user from the first client 100 to verify the first warning information to be verified. And only when the cognitive ability score of the user corresponding to the next period is lower than the cognitive ability scores of the users corresponding to the previous two periods of the next period by more than a preset trigger threshold value according to the cognitive ability score of the user corresponding to the next period, the server 300 confirms that the first early warning information is valid. The server 300 may transmit the first warning information to the second client 200 held by a person other than the user when confirming that the first warning information is valid. The invention can at least realize the following beneficial technical effects by adopting the mode: firstly, the method reduces the false alarm of the first early warning information by setting the comparison between the first confidence coefficient and the first confidence coefficient threshold value and the comparison between the second confidence coefficient and the second confidence coefficient threshold value, and improves the accuracy of the first early warning information; second, because the time length of the behavior of the user is more likely to fluctuate due to the latent factors compared to the behavior sequence, the first confidence threshold preset in the server 300 is greater than the second confidence threshold, thereby further reducing false alarms of the first warning information.
According to a preferred embodiment, the holder of the second client 200 may be, for example, a medical staff member or an operator of the system of the invention. The server 300 may transmit the first warning information to the second client 200 so as to let relevant persons know the change of the cognitive ability of the user. Preferably, after the second client 200 receives the first warning information, the second client 200 may instruct the holder to go to the living space of the user and evaluate the cognitive ability of the user by using a MoCA scale and/or an MMSE scale to obtain a field score reflecting the cognitive ability of the user, and when the field score is lower than a corresponding preset score, the user is determined to be no longer suitable for solitary residence. Therefore, the cognitive ability of the users whose cognitive ability is reduced to be not suitable for solitary residence which can generate corresponding risks can be accurately judged on site, and the solitary behaviors of the users can be interfered in time. For example, the user is sent to a nursing home where nursing staff intensively attends to the user, or family members are notified to take care of the user, so as to prevent the risk of the user whose cognitive ability is seriously deteriorated who lives alone.
Example 2
The embodiment discloses a cognitive ability assessment method based on the daily behaviors of the old, or a cognitive ability assessment method based on the daily behaviors of a user, or a method for assessing cognitive ability based on a time sequence database, or a method for assessing cognitive ability. The method may be implemented by the system of the present invention and/or other alternative components. For example, the method of the present invention may be implemented using various components of the system of the present invention. The preferred embodiments of the present invention are described in whole and/or in part in the context of other embodiments, which can supplement the present embodiment, without resulting in conflict or inconsistency.
According to a preferred embodiment, the method may use the system of the invention to assess the cognitive abilities of a user.
According to a preferred embodiment, the method may comprise at least one of the following steps 1,2, 3 and 4. The method can comprise an information acquisition step, an expert knowledge base construction step, a data processing step and a cognitive ability assessment and early warning step. The data acquisition step consists of a sensor and a server, wherein the sensor collects the action information of the old people at each moment, and the server collects and integrates the data of the sensor; the information knowledge base construction step comprises classification of original characteristics to a higher level and storage of classification data; the data processing step includes the construction of a behavior prediction classifier based on the classified stored information, in which a hidden markov model is used as a classification model; and the cognitive ability evaluation step comprises scoring the obtained behaviors and early warning the result that the score is reduced and exceeds a threshold value.
Step 1: and acquiring basic information, and constructing a sensor information data set comprising a sensor equipment information table and a sensor data information table. The sensor device information table describes basic information of various sensors, and the attributes of the basic information include: sensor name, MAC address, role description, sensor type, data unit, installation location. The MAC address serves as a unique identifier of the sensor device information. The sensor data information table describes the information content collected by the sensor, and the attributes of the sensor data information table comprise: sensor device, data content, upload time.
By arranging the sensors in the room, data are uploaded to the cloud server when the sensors are triggered, the server stores the data, and the data are periodically filed, stored and processed according to the uploading time of the data.
Step 2: and constructing an expert knowledge base, and providing priori knowledge of the relationship between the sensor and the behavior through expert knowledge to increase the accuracy of behavior prediction and further increase the accuracy of cognitive early warning. By constructing a table of relationship information between behavior, sensors and attributes, behavior prediction is not based solely on rules. The expert knowledge base mainly comprises a sensor behavior table, and the sensor behavior table comprises attributes: the method comprises the following steps of defining basic behaviors of daily life through artificial common sense, decomposing the behaviors into specific actions, associating the objects triggered by the actions with sensors, namely limiting the behaviors among the sensors possibly triggered by the behaviors, and improving the accuracy of behavior prediction through the rules. And classifying the bottom-layer sensors into all behaviors according to classification through a sensor behavior table, and storing the many-to-many relationship of the behaviors to the sensors in an expert knowledge base.
And step 3: behavior prediction analysis uses hidden markov models to predict behavior from sensor data for temporal sequencing implied in the sensor and behavior data. The hidden Markov model introduces a hidden variable, and considers that the change of the state in the data is caused by the hidden variable of the previous time point and the hidden variable, but the hidden variable cannot be directly observed by an observer. In the hidden Markov model, the behavior of the time point to be predicted is solved by a transition matrix among hidden variables, an initial state probability matrix and an emission matrix of the behavior corresponding to the hidden variables. In behavior recognition using hidden markov models, behaviors are considered hidden variables of sensors, sensor data is observable and behavior data is hidden.
According to a preferred embodiment, the present invention trains hidden Markov models using pre-collected data. And performing behavior prediction on the sensor data collected on the day by using a hidden Markov model in a 24-hour period. The steps of the process of markov model training and prediction will be given below:
variables are defined for hidden markov models:
name of variable Interpretation of variables
T Data sequence length
A Transition matrix between state hidden variables
B Transmission matrix
π Initial state probability matrix
N Number of hidden variable values
Q＝{q0,q1,…,qN-1} Set of possible values for hidden variables
I＝(i1,i2,…iT-1) Hidden variable sequence
O＝(O0,O1,…,OT-1) Observed data sequence
The value of the element defining the ith row and jth column of matrix A is also defined by the ith state qiTransition to jth state qjHas a probability ofij：
aij＝P(qj|qi)
The value defining the kth row and jth column of matrix B is also in state qjThe probability of observing data k:
bj(k)＝P(k|qj)
so far, the construction of the markov model λ ═ (a, B, pi) is completed, but both a and B are unknown, and the process and formula for solving the hidden markov model using the viterbi algorithm will be given below.
What the viterbi algorithm wants to maximize is the probability of a sequence of hidden variables after given observation data, i.e.:
max P(I|O)
viterbi algorithm definition
Figure BDA0002065091780000161
The recurrence formula can be derived from the above formula:
Figure BDA0002065091780000162
definition of Ψt(i)＝argmax1≤j≤Nδt-1(j)ajiThen, the flow of the viterbi algorithm is as follows:
inputting: model λ ═ (a, B, pi), and the observation data sequence O ═ O (O)0,o1,…,oT-1)
And (3) outputting: hidden variable sequence I ═ I (I)0,i1,…,iT-1)
Initializing a local state:
δ0＝πibi(o0),i＝0,1,…,N-1
Ψ0(i)＝0,i＝0,1,…,N-1
dynamic programming recursion T is performed as 1,2, …, local state at time T-1:
Figure BDA0002065091780000171
Figure BDA0002065091780000172
calculating Ψ with the maximum time TT(i) At this time ΨT(i) Namely the most probable hidden variable state at the moment T ═ T
Figure BDA0002065091780000173
Obtaining I ═ I by backtracking with Ψ (I)0,i1,…,iT-1)：
it＝Ψt+1(it+1)
In the prediction using the hidden markov model, the sensor data is regarded as an observation sequence, i.e., O ═ O (O)0,o1,…,oT-1) Behavior is considered to be a hidden variable, I ═ I (I)0,i1,…,iT-1). The behavior corresponding to each time instant can be obtained by using the viterbi algorithm.
And 4, step 4: and evaluating the cognitive ability and realizing early warning by identifying the obtained behavior information. The decline in cognitive ability manifests itself as behavioral errors in daily life, skipping or performing incorrect critical steps when completing an activity, rendering the activity meaningless to implement. Errors in these actions may include forgetting to turn off the gas, keeping the refrigerator door open for a long time, or taking a long time to complete a simple job.
The invention selects n behaviors which are daily triggered at most as a standard task X, and the standard task X is expressed as X ═ { X }1,X2,…,XnAnd testing to obtain the time of the general person for completing the n tasks as the standard time T of the behavior, wherein the standard time T is expressed as T ═ T { (T)1,T2,…,TnLet us assume that each person's task completion time is t ═ t1,t2,…,tnThe task completion degree is alpha ═ alpha1,α2,…,αnDetermining the weight of each task as w ═ w according to the importance of each task1,w2,…,wnThe scoring formula for cognitive ability is as follows:
Figure BDA0002065091780000174
the quality of the performance of the activity is quantified by normalizing the cognitive ability by a score. According to the daily scoring information, alarm information is generated when the cognitive ability score of the user is reduced by more than 10% of a threshold value.
According to a preferred embodiment, the behavior sensor device information table describes basic information of various types of behavior sensors, and attributes thereof are shown in the following table:
serial number (symbol) Attribute information
1 Device Name Sensor name
2 MAC address MAC address
3 Device Info Description of the action
4 Device Type Sensor type
5 Data Unit Data unit
6 Location Mounting location
According to a preferred embodiment, the sensor device collects data and transmits the data to a server. The server stores the sensor data information, and the attributes of the sensor data information are shown in the following table:
serial number (symbol) Attribute information
1 Device ID Sensor device
2 Data Data content
3 Time Upload time
According to a preferred embodiment, to detect a human basic action, the following deployments are made for a general family house:
Figure BDA0002065091780000181
Figure BDA0002065091780000191
according to a preferred embodiment, the behavior sensor device internally comprises a communication gateway and a communication module. The different sensors transmit information to the first client through the communication module. The communication module may be a wired communication module or a wireless communication module. The wireless communication module may be, for example, a bluetooth module or a ZigBee module. And the first client forwards all data to the server by connecting WIFI. The system is compatible with various behavior sensors, and when a new behavior sensor is accessed, the data of the sensor only needs to be set as a communication protocol determined by the current system. The behavior sensor transmits data only when a behavior generates a trigger, so that the transmission of data volume can be reduced.
According to a preferred embodiment, the communication protocol format definition is as follows:
Figure BDA0002065091780000192
Figure BDA0002065091780000201
according to a preferred embodiment, after the data is stored in the server side, due to the large data volume, the data is archived at twelve fixed time points in the evening every day, and the collected data on the day is sorted, packaged and stored.
According to a preferred embodiment, the behavior sensor-behavior correspondence table constructed from expert knowledge is as follows:
Figure BDA0002065091780000202
Figure BDA0002065091780000211
Figure BDA0002065091780000221
it should be noted that the above-mentioned embodiments are exemplary, and that those skilled in the art, having benefit of the present disclosure, may devise various arrangements that are within the scope of the present disclosure and that fall within the scope of the invention. It should be understood by those skilled in the art that the present specification and figures are illustrative only and are not limiting upon the claims. The scope of the invention is defined by the claims and their equivalents.

Claims (7)

1. A cognitive ability assessment system based on the daily behavior of the elderly, characterized in that the system comprises a server (300) and a plurality of behavior sensors (400),
wherein the server (300) collects behavior data of daily behaviors associated with the user in an implicit perception manner through the number of behavior sensors (400) deployed within the user's living space;
the server (300) evaluates the cognitive ability of the user according to at least the behavior data of the daily behaviors associated with the user to obtain a cognitive ability score of the user;
the server (300) compares the cognitive ability score of the user corresponding to the current period with the cognitive ability score of the user corresponding to the previous period periodically, and the server (300) generates first early warning information when the cognitive ability score cycle ratio of the user is reduced and exceeds a preset trigger threshold value;
the server (300) compares the cognitive ability score of the user corresponding to the current period with the cognitive ability score of the user corresponding to the previous period periodically in at least two different comparison periods, wherein the different comparison periods correspond to different preset trigger thresholds, and the longer the comparison period is, the smaller the preset trigger threshold corresponding to the comparison period with the longer period length is;
the server (300) is pre-stored with a preset behavior sequence and a preset time length of specific behaviors according with user behavior habits, after the server (300) generates first early warning information, the server (300) analyzes the accuracy of the comparison between the behavior sequence of the user in the current period and the preset behavior sequence according to received behavior data of daily behaviors associated with the user to obtain a first confidence coefficient, the server (300) analyzes a second confidence coefficient of the comparison between the time length of corresponding specific behaviors of the user in the current period and the preset time length of the specific behaviors according to the received behavior data of the daily behaviors associated with the user, a preset first confidence coefficient threshold and a preset second confidence coefficient threshold are stored in the server (300), the first confidence coefficient threshold is larger than the second confidence coefficient threshold, and the server (300) judges the second confidence coefficient when the first confidence coefficient is larger than the first confidence coefficient threshold and the second confidence coefficient is larger than the second confidence coefficient threshold The warning information is invalid when the warning information is invalid,
the server (300) directly confirms that the first early warning information is valid when the first confidence degree is less than or equal to a first confidence degree threshold value and the second confidence degree is less than or equal to a second confidence degree threshold value,
the server (300) marks the first early warning information as to-be-verified first early warning information when one condition of the first confidence coefficient less than or equal to a first confidence coefficient threshold value and the second confidence coefficient less than or equal to a second confidence coefficient threshold value is met, the server (300) waits for the behavior data of the daily behavior associated with the user in the next period sent by the first client (100) to verify the to-be-verified first early warning information, and only when the cognitive performance score of the user corresponding to the next period is lower than the cognitive performance score of the corresponding user evaluated in two previous periods of the next period by more than a preset trigger threshold value according to the server (300),
when the server (300) confirms that the first early warning information is valid, the first early warning information is sent to a second client (200) held by other people except the user; the server (300) evaluates the cognitive ability of the user according to the following scoring formula to obtain a cognitive ability score of the user:
Figure FDA0003310665070000021
before evaluating the cognitive ability of the user to obtain the cognitive ability score of the user, the server (300) selects behavior data corresponding to at most n behaviors triggered daily as a standard task X for evaluating the cognitive ability of the user, wherein the standard task X is expressed as X ═ { X ═ X1,X2,…,XnT is a set of standard times for the average person to complete each of the n tasks, and is expressed as T ═ T1,T2,…,Tn}，TiIs the standard time for ordinary people to complete the ith task in the n tasks, and t is the set of the average time for the user to complete each task in the n tasks, and is expressed as t ═ t1,t2,…,tn}，tiIs the average time of the user to complete the ith task in the n tasks, alpha is the set of task completion degrees of the user to complete each task in the n tasks, and alpha is the average time of the user to complete the ith task in the n tasksiRepresenting the task completion degree of the ith task in the n tasks, w is the set of weights of all the tasks in the n tasks, w isiAnd representing the weight of the ith task in the n tasks.
2. The system of claim 1, wherein the server (300) collects behavior data for daily behavior associated with the user with a time attribute, the server (300) acquires the behavior data of the daily behavior associated with the user and stores the behavior data of the daily behavior associated with the user into a time sequence database according to the time attribute, the server (300) independently builds for each user a hidden Markov model associated therewith for predicting its behaviour, the server (300) training a hidden Markov model associated with the user using the user related behavior data in the time series database, the server (300) predicting the predicted behavior of the user from the hidden Markov model associated with the user, and generating second warning information in the event that the actual behavior of the user deviates from the predicted behavior and the deviation would result in a known risk and/or a known loss.
3. The system of claim 2, further comprising a first client (100), wherein the first client (100) is communicatively connected to the server (300), wherein the first client (100) acts as a relay device to acquire the behavior data collected by the behavior sensors (400) disposed in the residential space, wherein the server (300) generates second warning information and sends a risk verification request to the first client (100), wherein the first client (100) sends an alarm in response to the risk verification request, wherein the alarm can be released only after the user passes the verification of at least two biometrics on the first client (100), wherein the first client (100) sends a risk false positive feedback or a risk release feedback to the server (300), wherein the server (300) deletes the second warning information in response to the risk false positive feedback or the risk release feedback And (4) information.
4. The system of claim 3, further comprising a camera device (500), the camera device (500) communicatively connected to a first client (100), the camera device (500) capable of being located outside an access door to access the living space, the camera device (500) configured to capture a first group of photographs taken before the access door is opened and a second group of photographs taken during a predetermined time after the access door is opened until the access door is closed, the first client (100) obtaining the first group of photographs and the second group of photographs from the camera device (500), the first client (100) comparing the first group of photographs and the second group of photographs to determine a person change status within the living space, and the first client (100) determining a person accommodation mode for persons accommodated within the living space in response to the person change status within the living space, the accommodation mode includes one of the following modes:
a user-independent mode that only one user exists in the living space;
visitor patterns with other people except the user or only other people in the living space; and
an empty mode in which no person is present in the living space;
wherein the first client (100) analyzes triggers of the behavior data collected by the behavior sensors (400) during the guest mode, the first client (100) intercepts the corresponding behavior data of the undeterminable triggers if the triggers of the corresponding behavior data collected during the guest mode cannot be determined, and only transmits the behavior data collected by the behavior sensors (400) during the guest mode, which can determine the triggers as users, to the server (300).
5. The system of claim 4, wherein the plurality of sensors includes at least an angular displacement sensor mounted on the access door for recognizing an open/close state of the access door, the angular displacement sensor for recognizing the open/close state of the access door is communicatively connected to the camera device (500), the angular displacement sensor for recognizing the open/close state of the access door is configured to actively send an electrical signal indicating that the access door is open to the camera device (500) immediately when the access door is rotated 3 ° to 5 ° in the opening direction, the camera device (500) takes a plurality of pictures within 1s as a first group of pictures after receiving the electrical signal indicating that the access door is open, the angular displacement sensor for recognizing the open/close state of the access door is configured to send an electrical signal indicating that the access door is closed to the camera device (500) only when the access door is rotated in the closing direction until the access door is completely closed, the camera device (500) takes a plurality of pictures continuously after the first picture group is taken until the preset time period lasts after the electric signal that the door of the entrance is closed is received, and the second picture group is taken.
6. The system according to claim 5, characterized in that the system further comprises a smart bracelet (600) and an identification module (700), the server (300) assigns a unique identification to each user and stores the identification in the smart bracelet (600), the smart bracelet (600) sends electromagnetic waves carrying the identification to a limited communication range, each behavior sensor (400) is equipped with the identification module (700) for recognizing the electromagnetic waves carrying the identification to detect the smart bracelet (600) worn by the user and thereby identify the trigger of the behavior data collected by the behavior sensor (400),
when a first client (100) determines that the accommodation mode of the people accommodated in the living space is the user-independent mode, the first client (100) receives behavior data collected by a behavior sensor (400) and defines a trigger, which is collected by an identity recognition module (700) to the behavior data under the condition that the situation that the user does not wear the smart bracelet (600), as the user by default;
when the first client (100) determines that the accommodation mode of the people accommodated in the living space is changed from the user-independent mode to the visitor mode in response to the change state of the people in the living space, the first client (100) instructs the user to wear the smart band (600), the first client (100) receives the behavior data collected by the behavior sensor (400) and determines only the behavior data collected by the identity recognition module (700) when the smart band (600) worn by the user is detected as the behavior data of which the trigger is the user.
7. A cognitive ability assessment method based on the daily behaviour of elderly people, characterized in that the method uses a system according to one of the claims 1 to 6 for assessing the cognitive ability of a user, the system comprising a server (300) and a number of behaviour sensors (400),
wherein the server (300) collects behavior data of daily behaviors associated with the user in an implicit perception manner through the number of behavior sensors (400) deployed within the user's living space;
the server (300) evaluates the cognitive ability of the user according to at least the behavior data of the daily behaviors associated with the user to obtain a cognitive ability score of the user;
the server (300) compares the cognitive ability score of the user corresponding to the current period with the cognitive ability score of the user corresponding to the previous period periodically, and the server (300) generates first early warning information when the cognitive ability score cycle ratio of the user is reduced and exceeds a preset trigger threshold value;
the server (300) compares the cognitive ability score of the user corresponding to the current period with the cognitive ability score of the user corresponding to the previous period periodically in at least two different comparison periods, wherein the different comparison periods correspond to different preset trigger thresholds, and the longer the comparison period is, the smaller the preset trigger threshold corresponding to the comparison period with the longer period length is;
the server (300) is pre-stored with a preset behavior sequence and a preset time length of specific behaviors according with user behavior habits, after the server (300) generates first early warning information, the server (300) analyzes the accuracy of the comparison between the behavior sequence of the user in the current period and the preset behavior sequence according to received behavior data of daily behaviors associated with the user to obtain a first confidence coefficient, the server (300) analyzes a second confidence coefficient of the comparison between the time length of corresponding specific behaviors of the user in the current period and the preset time length of the specific behaviors according to the received behavior data of the daily behaviors associated with the user, a preset first confidence coefficient threshold and a preset second confidence coefficient threshold are stored in the server (300), the first confidence coefficient threshold is larger than the second confidence coefficient threshold, and the server (300) judges the second confidence coefficient when the first confidence coefficient is larger than the first confidence coefficient threshold and the second confidence coefficient is larger than the second confidence coefficient threshold The warning information is invalid when the warning information is invalid,
the server (300) directly confirms that the first early warning information is valid when the first confidence degree is less than or equal to a first confidence degree threshold value and the second confidence degree is less than or equal to a second confidence degree threshold value,
the server (300) marks the first early warning information as to-be-verified first early warning information when one condition of the first confidence coefficient less than or equal to a first confidence coefficient threshold value and the second confidence coefficient less than or equal to a second confidence coefficient threshold value is met, the server (300) waits for the behavior data of the daily behavior associated with the user in the next period sent by the first client (100) to verify the to-be-verified first early warning information, and only when the cognitive performance score of the user corresponding to the next period is lower than the cognitive performance score of the corresponding user evaluated in two previous periods of the next period by more than a preset trigger threshold value according to the server (300),
when the server (300) confirms that the first early warning information is valid, the first early warning information is sent to a second client (200) held by other people except the user;
the server (300) evaluates the cognitive ability of the user according to the following scoring formula to obtain a cognitive ability score of the user:
Figure FDA0003310665070000061
before evaluating the cognitive ability of the user to obtain the cognitive ability score of the user, the server (300) selects behavior data corresponding to at most n behaviors triggered daily as a standard task X for evaluating the cognitive ability of the user, wherein the standard task X is expressed as X ═ { X ═ X1,X2,…,XnT is a set of standard times for the average person to complete each of the n tasks, and is expressed as T ═ T1,T2,…,Tn}，TiIs the standard time for ordinary people to complete the ith task in the n tasks, and t is the set of the average time for the user to complete each task in the n tasks, and is expressed as t ═ t1,t2,…,tn}，tiIs the average time of the user to complete the ith task of the n tasks, and alpha is the time of the user to complete each task of the n tasksSet of task completions, αiRepresenting the task completion degree of the ith task in the n tasks, w is the set of weights of all the tasks in the n tasks, w isiAnd representing the weight of the ith task in the n tasks.
CN201910421469.5A 2019-05-20 2019-05-20 Cognitive ability assessment system and method based on daily behaviors of old people Active CN110136832B (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN202210356940.9A CN114724714A (en) 2019-05-20 2019-05-20 Patient cognitive ability monitoring system and method based on user space
CN202210362582.2A CN114974577A (en) 2019-05-20 2019-05-20 Patient cognitive ability analysis and early warning system and method based on behavior prediction
CN201910421469.5A CN110136832B (en) 2019-05-20 2019-05-20 Cognitive ability assessment system and method based on daily behaviors of old people

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910421469.5A CN110136832B (en) 2019-05-20 2019-05-20 Cognitive ability assessment system and method based on daily behaviors of old people

Related Child Applications (2)

Application Number Title Priority Date Filing Date
CN202210356940.9A Division CN114724714A (en) 2019-05-20 2019-05-20 Patient cognitive ability monitoring system and method based on user space
CN202210362582.2A Division CN114974577A (en) 2019-05-20 2019-05-20 Patient cognitive ability analysis and early warning system and method based on behavior prediction

Publications (2)

Publication Number Publication Date
CN110136832A CN110136832A (en) 2019-08-16
CN110136832B true CN110136832B (en) 2022-04-15

Family

ID=67571819

Family Applications (3)

Application Number Title Priority Date Filing Date
CN201910421469.5A Active CN110136832B (en) 2019-05-20 2019-05-20 Cognitive ability assessment system and method based on daily behaviors of old people
CN202210356940.9A Pending CN114724714A (en) 2019-05-20 2019-05-20 Patient cognitive ability monitoring system and method based on user space
CN202210362582.2A Pending CN114974577A (en) 2019-05-20 2019-05-20 Patient cognitive ability analysis and early warning system and method based on behavior prediction

Family Applications After (2)

Application Number Title Priority Date Filing Date
CN202210356940.9A Pending CN114724714A (en) 2019-05-20 2019-05-20 Patient cognitive ability monitoring system and method based on user space
CN202210362582.2A Pending CN114974577A (en) 2019-05-20 2019-05-20 Patient cognitive ability analysis and early warning system and method based on behavior prediction

Country Status (1)

Country Link
CN (3) CN110136832B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110502303B (en) * 2019-08-20 2023-07-11 银康(中山)科技有限公司 Method, device, electronic equipment and storage medium for evaluating ability of old people
CN110472891B (en) * 2019-08-26 2023-10-17 上海蔚来汽车有限公司 Method, system and storage medium for evaluating or managing power-up service capability or resource
CN110752033A (en) * 2019-08-30 2020-02-04 康美健康云服务有限公司 Informationization method and system for cognitive assessment and readable storage medium
CN111128383A (en) * 2019-12-30 2020-05-08 成都麦杰康科技有限公司 Information processing method, information processing device, electronic equipment and storage medium
CN113962444A (en) * 2021-09-29 2022-01-21 湖北美和易思教育科技有限公司 Student quality literacy prediction system based on reinforcement learning

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1564219A (en) * 2004-03-25 2005-01-12 浙江工业大学 Household security device for lonely living edged people
CN106092133A (en) * 2016-05-27 2016-11-09 北京灵龄科技有限责任公司 Start the decision method of guest mode, device
CN107004062A (en) * 2014-12-19 2017-08-01 皇家飞利浦有限公司 For assessing the personal equipment, system and method for performing one or more movable abilities
CN107886103A (en) * 2016-09-29 2018-04-06 日本电气株式会社 For identifying the method, apparatus and system of behavior pattern
CN108348194A (en) * 2015-09-15 2018-07-31 联邦科学和工业研究机构 Mobility monitors
CN109752963A (en) * 2019-01-02 2019-05-14 沈阳工程学院 A kind of house security system based on Internet of Things

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104881591B (en) * 2015-06-25 2018-03-30 贾建平 cognitive disorder data processing method and processing system
US11004567B2 (en) * 2017-08-15 2021-05-11 Koko Home, Inc. System and method for processing wireless backscattered signal using artificial intelligence processing for activities of daily life
CN109473173B (en) * 2018-09-30 2021-04-13 华中科技大学 Old person cognitive function defect assessment system and device based on video

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1564219A (en) * 2004-03-25 2005-01-12 浙江工业大学 Household security device for lonely living edged people
CN107004062A (en) * 2014-12-19 2017-08-01 皇家飞利浦有限公司 For assessing the personal equipment, system and method for performing one or more movable abilities
CN108348194A (en) * 2015-09-15 2018-07-31 联邦科学和工业研究机构 Mobility monitors
CN106092133A (en) * 2016-05-27 2016-11-09 北京灵龄科技有限责任公司 Start the decision method of guest mode, device
CN107886103A (en) * 2016-09-29 2018-04-06 日本电气株式会社 For identifying the method, apparatus and system of behavior pattern
CN109752963A (en) * 2019-01-02 2019-05-14 沈阳工程学院 A kind of house security system based on Internet of Things

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于隐式感知的老人日常活动识别系统;姚小慧;《中国优秀硕士学位论文全文数据库信息科技辑》;20190215(第02期);全文 *

Also Published As

Publication number Publication date
CN110136832A (en) 2019-08-16
CN114974577A (en) 2022-08-30
CN114724714A (en) 2022-07-08

Similar Documents

Publication Publication Date Title
CN110136832B (en) Cognitive ability assessment system and method based on daily behaviors of old people
US11050577B2 (en) Automatically learning and controlling connected devices
US11176793B1 (en) Predictive alarm analytics
AU2020265650B2 (en) Machine learning motion sensing with auxiliary sensors
CN100355228C (en) Anomalous detecting device and anomalous detection system
US20210123768A1 (en) Automated mapping of sensors at a location
CN109891436A (en) Security system and its control method based on deep learning neural network
US11810437B2 (en) Integrated security for multiple access control systems
CN106773763B (en) A kind of information linkage processing method and device
US10943153B2 (en) Ultrasound analytics for actionable information
US20230196106A1 (en) Training image classifiers
US11544924B1 (en) Investigation system for finding lost objects
US11734932B2 (en) State and event monitoring
Niu et al. Recognizing ADLs of one person household based on non-intrusive environmental sensing
US20220222943A1 (en) Intelligent pausing of recording by a property monitoring system
US20220101066A1 (en) Reducing false detections for night vision cameras
Choi et al. IoT-based occupants counting with smart building state variables
US11550276B1 (en) Activity classification based on multi-sensor input
US11624519B2 (en) Carbon monoxide purge system for a property
US20230081909A1 (en) Training an object classifier with a known object in images of unknown objects
US20230039967A1 (en) Airborne pathogen detection through networked biosensors
US11900706B1 (en) Object of interest detector distance-based multi-thresholding
Niu et al. Analyzing Indoor Environment Sensing Data for Recognizing ADLs of One Person Household

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant