WO2018165868A1 - 监测方法和监测装置 - Google Patents
监测方法和监测装置 Download PDFInfo
- Publication number
- WO2018165868A1 WO2018165868A1 PCT/CN2017/076659 CN2017076659W WO2018165868A1 WO 2018165868 A1 WO2018165868 A1 WO 2018165868A1 CN 2017076659 W CN2017076659 W CN 2017076659W WO 2018165868 A1 WO2018165868 A1 WO 2018165868A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- alarm
- abnormal
- monitoring
- threshold
- Prior art date
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
- A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/01—Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/746—Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms
Definitions
- the present invention relates to the field of electronic technologies, and in particular, to a monitoring method and a monitoring device.
- a monitoring scheme is to achieve monitoring by actively monitoring the object to report the police information to the guardian.
- the application number is CN201610378951.1 Chinese patent, which requires the child to notify the guardian through the alarm in case of danger.
- this solution can only passively receive the alarm information reported by the monitoring object, and cannot actively detect the abnormal situation. There are cases where the alarm is not as good as the alarm. Especially in some dangerous situations, the party cannot or does not have the opportunity to report it.
- Another monitoring scheme is to determine whether the monitoring object is within the guardian's setting range, and when the setting range is exceeded, an alarm prompt is performed to implement the monitoring.
- the Chinese patent of the application number CN201310647844.0 according to the signal receiving device carried by the child, performs an alarm outside the signal transmission range of the signal transmitting device carried by the adult, thereby realizing the anti-lost function.
- this scheme is limited by the distance between the monitoring object and the guardian, and the scope of application is narrow and the accuracy is low.
- the main object of the embodiments of the present invention is to provide a monitoring method and a monitoring device, which are intended to solve the technical problems of the existing monitoring scheme with poor practicability and low accuracy and low accuracy.
- a monitoring device comprising:
- the information receiving module is configured to receive the physiological parameter information of the monitoring object reported by the information collecting device; [0013] the first analyzing and determining module is configured to input the physiological parameter information into a preset body health index model for analysis, Determining whether the physical state of the monitored object is abnormal;
- the abnormality alarm module is configured to perform an abnormality alarm when the physical state of the monitored object is abnormal.
- a monitoring method provided by an embodiment of the present invention by collecting physiological parameter information of a monitoring object, and inputting physiological parameter information into a preset body health index model to analyze whether the physical state of the monitoring object is abnormal, when determining If the physical state of the monitored object is abnormal, then an abnormal alarm is issued to prevent the monitoring object of the elderly, children, etc. from being lost or abducted.
- the monitoring method of the embodiment of the present invention completely realizes automatic monitoring and automatic alarming, and is not limited to the distance between the monitoring object and the guardian, greatly expands the scope of application, and improves the practicality and implementation of the monitoring. Sex and accuracy.
- the embodiment of the present invention can construct a physical health index model and a daily activity trajectory distribution model unique to the monitoring object by means of machine learning modeling, and can utilize the collected physiological parameter information and daily activity information and environmental information in the monitoring process. Continuously adjust and update the body health index model and the daily activity trajectory distribution model, so that the two models are in an uninterrupted learning state, which greatly improves the accuracy of the test results and improves the reliability of the system.
- FIG. 1 is a flow chart of a monitoring method according to a first embodiment of the present invention
- FIG. 2 is a schematic diagram of a body health index model in an embodiment of the present invention.
- FIG. 3 is a flow chart of a monitoring method according to a second embodiment of the present invention.
- FIG. 4 is a schematic diagram of a daily activity trajectory distribution model according to an embodiment of the present invention.
- FIG. 5 is a block diagram of a monitoring device according to a third embodiment of the present invention.
- 6 is a block diagram of a monitoring device according to a fourth embodiment of the present invention.
- FIG. 7 is a block diagram of a monitoring device according to a fifth embodiment of the present invention.
- FIG. 8 is a block diagram of a monitoring system in an embodiment of the present invention.
- first”, “second” and the like in the present invention are used for descriptive purposes only, and are not to be construed as indicating or implying their relative importance or implicitly indicating the number of indicated technical features. .
- features defining “first” and “second” may include at least one of the features, either explicitly or implicitly.
- the technical solutions between the various embodiments may be combined with each other, but must be based on the realization of those skilled in the art, and when the combination of the technical solutions is contradictory or impossible to implement, it should be considered that the combination of the technical solutions does not exist. It is also within the scope of protection required by the present invention.
- the monitoring object according to the embodiment of the present invention mainly refers to a child, an elderly person, a special group (such as a mentally handicapped person, a mentally retarded person, and the like who have no independent living ability), and the like, and may also be a pet. And other objects.
- FIG. 1 a monitoring method according to a first embodiment of the present invention is proposed.
- the method includes the following steps:
- Sl l receiving physiological parameter information of the monitored object reported by the information collection device.
- the information collecting device is worn on the body of the monitoring object, and is carried by the monitoring object.
- the information collection device collects or monitors the physiological parameter information of the monitored object, and passes the actual or fixed
- the line communication method is reported to the monitoring device (for example, once every 1 minute).
- the physiological parameter information of the monitored object is monitored.
- the physiological parameter information includes at least one of parameter information such as body temperature, blood pressure, and heart rate.
- the information collecting device collects three parameter information of body temperature, blood pressure and heart rate of the monitored object.
- S12 Enter physiological parameter information into a preset body health index model for analysis, and determine whether the physical state of the monitored object is abnormal.
- the body health index model is preset and stored in the monitoring device.
- the body health index model indicates the probability distribution of the monitored subject in terms of body temperature, blood pressure, heartbeat, etc., to characterize that the monitored subject's body is in a normal state, there is no coercion or loss of subjective consciousness, or there is no sudden illness. And so on.
- a body health index model was used to monitor anomalies. In the large amount of information collected, this information is regarded as a large number of discrete data points in high-dimensional space. Our purpose is to find objects in these data point objects that are different from most other objects. These abnormal data points are called outliers. Point (Outlier).
- the body health index model may be a general body health index model obtained directly by the monitoring device from the outside, or may be a specific body health index model established by the monitoring device by collecting statistics to monitor physiological parameters of the subject in a normal state. .
- the monitoring object needs to wear a period of time (model learning creation period) information collecting device to collect a certain amount of modeling data.
- the information collection device collects the physiological parameter information of the monitored object under normal conditions (such as physical health) during the model learning creation period (such as one day), and reports or collects the collected physiological parameter information. (If reported once every minute) to the monitoring device, the monitoring device trains the body health index model of the monitored object based on the acquired physiological parameter information such as body temperature, blood pressure, and heart rate.
- the body health index model includes three-dimensional information of body temperature, blood pressure and heart rate
- the monitoring device constructs a three-dimensional coordinate system as shown in FIG. 2 according to three-dimensional information
- the X-axis is body temperature information
- the y-axis is blood pressure.
- Information, z-axis is heart rate information.
- the monitoring device inputs the physiological parameter information collected during the model learning creation period into the three-dimensional coordinate system to obtain corresponding coordinate points, and calculates a central point M according to the distribution of all the collected physiological parameter information in the three-dimensional coordinate system, which is defined as three-dimensional.
- the center point of the coordinate system as a reference point for judging whether the body state is abnormal or not, it should be noted that the center point described here does not mean sitting. Original origin. If the coordinate points of the physiological parameter information collected in the subsequent monitoring and collecting on the three-dimensional coordinate system are outside the preset range of the center point M, it is discriminated as an outlier, and the three points A, B and C in Fig. 2 are outliers. This ⁇ body health index model outputs abnormal results.
- the monitoring device receives the physiological parameter information reported by the information collecting device, and inputs the body temperature, blood pressure and heart rate in the collected physiological parameter information into the three-dimensional coordinate system of the body health index model, assuming that a coordinate point is obtained.
- I. Calculate the distance d(M, I) of the coordinate point I from the center point M of the three-dimensional coordinate system, and determine whether the distance d(M, I) is greater than the first threshold. When the distance d(M, I) is greater than the first threshold ⁇ , it indicates that the coordinate point I is an outlier, and the physical state of the monitored object is determined to be abnormal.
- the monitoring device may further determine whether the distance d(M, I) is less than a second threshold (the second threshold is less than the first threshold).
- the second threshold is less than the first threshold.
- the body health index model is updated by the coordinate point I, and the center point M is re-determined, that is, the new center point is recalculated.
- the aforementioned distance calculation may use an Euclidean distance, a Mahalanobis distance, or the like.
- the body health index model may also include any two kinds of information such as body temperature, blood pressure, and heart rate, construct a two-dimensional coordinate system, or even include only one of the information, and construct One-dimensional coordinate system.
- the invention is not limited thereto.
- the body health index model is not limited to the present invention by using the coordinate system and its center point to determine whether the body state is abnormal.
- step S13 when the physical state abnormality of the current monitoring object is determined by the body health index model, an abnormal alarm is immediately performed to remind the relevant personnel.
- the monitoring device can directly dial the relevant telephone number (such as 110) to make an alarm.
- the monitoring device when the monitoring device is a terminal device (such as a mobile terminal carried by the guardian or a fixed terminal of the monitoring center), the monitoring device may directly perform an alert prompt, including a voice prompt and/or a visual information prompt. Such as sounding an alarm, broadcasting a voice message, displaying text information, displaying image information, and the like.
- an alert prompt including a voice prompt and/or a visual information prompt.
- the monitoring device may push the alarm information to the alarm prompting device (such as a mobile phone, a tablet, etc. carried by the guardian), and the alarm prompting device receives the alarm information.
- Alert prompts immediately, including voice prompts and/or visual message prompts, such as sounding an alarm, broadcasting voice messages, displaying text messages, displaying image information, and more.
- the user such as the guardian
- the alarm prompting device is also allowed to manually perform alarm confirmation on the alarm information, and receive the user's alarm confirmation information.
- the alarm confirmation information is an alarm error
- the user confirms that the monitoring device pushes.
- the abnormal alarm is incorrect.
- the alarm prompting device immediately sends an alarm error message to the monitoring device.
- the monitoring device adjusts the body health index model by using the collected physiological parameter information to update the body health index model to improve the monitoring accuracy of the monitoring device.
- the monitoring device uses the physiological parameter information collected this time to train the body health index model to continuously update the body health index model to ensure the monitoring device. The accuracy of the monitoring.
- the user may also be allowed to set the temporary state.
- the temporary state such as fitness, sickness, etc.
- the monitoring device only uses the collected physiological parameter information to detect the physical state of the monitored object, and suspends the use of the physiological parameter.
- Information updates the body health index model. For example, when the monitoring device detects that the body of the monitoring object is abnormal and performs an abnormal alarm, after receiving the alarm error message, the monitoring device does not adjust the body health index model. This avoids inappropriate modifications to the body health index model and increases flexibility.
- the monitoring method of the embodiment of the present invention collects the physiological parameter information of the monitoring object, inputs the physiological parameter information into a preset body health index model to analyze whether the physical state of the monitoring object is abnormal, and determines the body of the monitoring object. If the status is abnormal, then an abnormal alarm is issued to prevent the elderly, children and other monitoring objects from being lost or abducted.
- the monitoring method of the embodiment of the present invention completely realizes automatic monitoring and automatic alarming, and is not limited to the distance between the monitoring object and the guardian, greatly expands the scope of application, and improves the practicality and implementation of the monitoring. Sex and accuracy.
- the embodiment of the present invention can construct a body health index model unique to the monitoring object by means of machine learning modeling, and can continuously adjust and update the body health index model by using the collected physiological parameter information during the monitoring process, so that The model is in an uninterrupted learning state, which greatly improves the accuracy of the test results and improves the reliability of the system.
- FIG. 3 a monitoring method according to a second embodiment of the present invention is proposed, and the method includes the following steps:
- S21 Receive physiological parameter information of the monitoring object reported by the information collecting device, daily activity information, and environment information of the environment in which the environment is located.
- the information collecting device is worn on the body of the monitoring object, and is carried by the monitoring object.
- the information collection device collects or monitors the physiological parameter information of the monitored object, the daily activity information, and the environmental information of the environment in which it is located, and reports it to the monitoring device through wireless communication (eg, every 1 minute). Once) physiological parameter information, daily activity information and environmental information.
- the physiological parameter information includes at least one of parameter information such as body temperature, blood pressure, and heart rate.
- the information collecting device collects three kinds of parameter information of body temperature, blood pressure, and heart rate of the monitored object.
- the daily activity information includes at least one of activity information such as location information and speed information.
- the information collection device collects two types of information: the location and the speed of the monitoring object.
- the speed includes the moving speed, and may also include the acceleration.
- the environment information includes at least one of temperature information, humidity information, and weather information, and may further include current daytime information.
- the information collecting device collects current temperature, weather, and time.
- the weather includes sunny days, rainy days, snowy days, and so on.
- S22 Enter physiological parameter information into a preset body health index model for analysis, and determine whether the physical state of the monitored object is abnormal. When the physical state of the monitored subject is abnormal, the process proceeds to step S23; when the physical state of the monitored subject is normal, the process proceeds to step S24.
- steps S22 and S23 are the same as steps S12 and S13 in the first embodiment, and are not described herein.
- S24 Input daily activity information and environmental information into a preset daily activity trajectory distribution model for analysis, and determine whether the activity state of the monitoring object is abnormal.
- the daily activity trajectory distribution model is also preset and stored in the monitoring device.
- the daily activity trajectory distribution model combines the daily activity information and environmental information of the monitored objects, and indicates the normal activities of the monitored objects under different combinations of different conditions, different temperatures and different weather conditions. There is no coercion or loss of subjectivity. The situation of consciousness. In general, it is necessary to combine the previous model, the body health index model, to make a judgment.
- the daily activity trajectory distribution model is used to monitor anomalies. In the large amount of information collected, this information is regarded as a large number of discrete data points in high-dimensional space. Our purpose is to find objects in these data point objects that are different from most other objects. These abnormal data points are called outliers. point. [0066]
- the daily activity trajectory distribution model may be an initial daily activity trajectory distribution model obtained directly by the monitoring device from the outside, or may be established by the monitoring device by collecting statistical activity information and environmental information under normal conditions of the monitoring object. Specific body health index model.
- the monitoring object wears the information collecting device, and the information collecting device actually or fixedly collects the monitoring object under normal conditions (such as when determining the monitoring object)
- the daily activity information and environmental information of the physical state are normal, and the collected daily activity information and environmental information are reported or reported (such as once every minute) to the monitoring device, and the monitoring device is based on the daily routine obtained.
- Activity information and environmental information are used to train the daily activity trajectory distribution model of the monitored object.
- the environment information includes three dimensions: daytime, temperature, and weather, and each dimension is quantized into at least two sections, and different combinations of different sections, temperature intervals, and weather conditions correspond to different ones.
- the daily activity trajectory distribution model includes two-dimensional information of position and velocity, and the monitoring device constructs a two-dimensional coordinate system as shown in FIG. 4 according to the two-dimensional information, for example, the X axis is position information, and the y axis is For speed information.
- the monitoring device first determines the corresponding environment combination according to the environment information collected during the model learning creation period, and then inputs the position and speed in the daily activity information into the two-dimensional coordinate system corresponding to the environment combination to obtain corresponding coordinate points, according to the collected
- the distribution of all the daily activity information in the environment combination in the two-dimensional coordinate system calculates a center point N, which is defined as the center point of the two-dimensional coordinate system.
- the center point described here does not refer to the coordinate origin.
- the coordinate points of the daily activity information in the two-dimensional coordinate system acquired by the subsequent monitoring and collecting in the environment combination are judged as outliers if they are outside the preset range of the center point N, as shown in Figure 4, D, E, F
- the three points are outliers, and the daily activity trajectory distribution model outputs abnormal results.
- the monitoring device receives the daily activity information and the environment information reported by the information collecting device, determines a corresponding daily activity trajectory distribution model according to the environment combination corresponding to the environment information, and performs daily activities.
- the two-dimensional coordinate system of the daily activity trajectory distribution model corresponding to the position and velocity input in the motion information
- a coordinate point J is obtained, and the distance d (N, J) of the coordinate point J from the center point N of the three-dimensional coordinate system is calculated.
- determining whether the distance d(N, J) is greater than a third threshold When the distance d(N, J) is greater than the third threshold ⁇ , it indicates that the coordinate point J is an outlier, and it is determined that the physical state of the monitored object is abnormal.
- the monitoring device may further determine whether the distance d(N, J) is less than a fourth threshold (the fourth threshold is less than the third threshold).
- the fourth threshold is less than the third threshold.
- the body health index model is updated by using the coordinate point J, and the center N is re-determined, that is, the new center point is recalculated.
- the aforementioned distance calculation may use an Euclidean distance, a Mahalanobis distance, or the like.
- the daily activity trajectory distribution model may also include any one of position and velocity to construct a one-dimensional coordinate system.
- the invention is not limited thereto.
- the daily activity trajectory distribution model may adopt other modes in the prior art in addition to the coordinate system and its center point to determine whether the active state is abnormal.
- the present invention does not limit this.
- step S25 when the current active trajectory distribution model determines that the current active state of the monitoring object is abnormal (such as a remote place, a place that has never been visited, or an abnormal moving speed, etc.), an abnormal alarm is immediately performed. To remind the relevant personnel.
- abnormal such as a remote place, a place that has never been visited, or an abnormal moving speed, etc.
- the monitoring device can directly dial the relevant telephone number (such as 110) to make an alarm.
- the monitoring device when the monitoring device is a terminal device (such as a mobile terminal carried by the guardian or a fixed terminal of the monitoring center), the monitoring device may directly perform an alert prompt, including a voice prompt and/or a visual information prompt. Such as sounding an alarm, broadcasting a voice message, displaying text information, displaying image information, and the like.
- an alert prompt including a voice prompt and/or a visual information prompt.
- the monitoring device may push the alarm information to the alarm prompting device (such as a mobile phone, a tablet, etc. carried by the guardian), and the alarm prompting device receives the alarm information.
- Alert prompts immediately, including voice prompts and/or visual message prompts, such as sounding an alarm, broadcasting voice messages, displaying text messages, displaying image information, and more.
- the user (such as the guardian) is also allowed to manually perform alarm confirmation on the alarm information, and receive the user's alarm confirmation information.
- the alarm confirmation information is an alarm error
- the user confirms that the monitoring device pushes.
- the abnormal alarm is incorrect.
- the alarm prompting device immediately sends an alarm error message to the monitoring device.
- the monitoring device when receiving the alarm error information ⁇ for the abnormal alarm of the abnormal physical state, the monitoring device adjusts the body health index model by using the collected physiological parameter information to update the body health index model.
- the monitoring device adjusts the daily activity trajectory distribution model by using the daily activity information and the environmental information collected to update the daily activity trajectory distribution model. Thereby improving the accuracy of monitoring device monitoring.
- the monitoring device can send the alarm information to the alarm prompting device through the short message channel.
- the monitoring device pushes the alarm information to the specific application on the alarm prompting device.
- the specific application of the alarm prompting device can also be used for the user to query the current physical state and activity status of the monitored object. Meanwhile, after receiving the alarm information pushed by the monitoring device, the user can manually perform the alarm confirmation, and confirm the monitoring device push. The abnormal alarm is incorrect.
- the alarm prompting device can feed the alarm error information to the monitoring device through the specific application.
- the monitoring device automatically adjusts the model according to the alarm error information fed back by the user to obtain a more accurate analysis result.
- the monitoring device when the physical state of the monitored object is normal, uses the physiological parameter information collected this time to train the body health index model to continuously update the body health index model to ensure monitoring of the monitoring device. The accuracy. Peer, further, when the activity status of the monitoring object is normal, the monitoring device uses the daily activity information and environmental information collected to train the daily activity trajectory distribution model to update the daily activity trajectory distribution model to ensure monitoring device monitoring. Accuracy
- the user may also be allowed to set the temporary state.
- the temporary state such as fitness, sickness, travel, etc.
- the monitoring device only uses the collected information (physiological parameter information or daily activity information and environment). Information) Detecting the physical state or activity state of the monitored object, the monitoring device pauses to update the body health index model or the daily activity trajectory distribution model using the collected information (physiological parameter information or daily activity information and environmental information).
- the monitoring device detects that the monitoring object has abnormal body state or abnormal state of activity and performs an abnormality alarm
- the monitoring device does not adjust the body health index model or the daily activity trajectory distribution model, thereby avoiding Improper modification of the body health index model or the daily activity distribution trajectory model.
- This embodiment monitors the physical state and activity state of the monitored object through peers, further improves the accuracy of the detection result and the reliability of the system, and effectively prevents the monitoring objects such as the elderly and children from being lost or abducted.
- the monitoring method of the embodiment of the present invention constructs a body health index model and a daily activity trajectory distribution model unique to the monitoring object by means of machine learning modeling, and both models are in an uninterrupted learning state.
- the automatic monitoring and automatic reporting are completely realized, which greatly improves the practicability, practicality and accuracy of the monitoring, and improves the accuracy of the detection result by continuously updating the model. System reliability.
- the apparatus includes an information receiving module 110.
- the first analysis and judgment module 120 and the abnormality alarm module 130 wherein:
- the information receiving module 110 is configured to receive physiological parameter information of the monitoring object reported by the information collecting device
- the physiological parameter information includes at least one of parameter information such as body temperature, blood pressure, and heart rate.
- the information collecting device collects three parameter information of body temperature, blood pressure and heart rate of the monitored object.
- the first analysis judging module 120 is configured to input the physiological parameter information into a preset body health index model for analysis, and determine whether the physical state of the monitored object is abnormal.
- the body health index model is preset and stored in the monitoring device.
- the body health index model indicates the probability distribution of the monitored subject in terms of body temperature, blood pressure, heartbeat, etc., to characterize that the monitored subject's body is in a normal state, there is no coercion or loss of subjective consciousness, or there is no sudden illness. And so on.
- a body health index model was used to monitor anomalies. In the large amount of information collected, this information is regarded as a large number of discrete data points in high-dimensional space. Our purpose is to find most of these data point objects. Objects with different objects, these abnormal data points are called Outliers.
- the body health index model may be a general body health index model obtained directly by the monitoring device from the outside, or may be a specific body health index model established by the monitoring device by collecting statistics to monitor physiological parameters of the subject in a normal state. .
- the monitoring device includes a model creation module, and during the model learning creation period (such as one day), the monitoring object wears the information collecting device, and the information collecting device implements or fixes the collected monitoring object under normal conditions ( For example, the physiological parameter information of the body health ,, and the collected physiological parameter information is reported or reported (for example, once every minute) to the monitoring device, and the model creation module is based on the acquired body temperature, blood pressure, heart rate, etc. The physiological parameter information is used to train the body health index model of the monitored object.
- the body health index model includes three-dimensional information of body temperature, blood pressure and heart rate
- the model creation module constructs a three-dimensional coordinate system as shown in FIG. 2 according to the three-dimensional information, for example, the X-axis is body temperature information, and the y-axis is Blood pressure information, z- axis is heart rate information.
- the model creation module inputs the physiological parameter information collected during the model learning creation period into the three-dimensional coordinate system to obtain corresponding coordinate points, and calculates a central point M according to the distribution of all the collected physiological parameter information in the three-dimensional coordinate system, which is defined as The center point of the three-dimensional coordinate system is used as a reference point for judging whether the body state is abnormal.
- the first analysis determining module 120 inputs the body temperature, the blood pressure, and the heart rate in the collected physiological parameter information into the three-dimensional coordinate system of the body health index model, and assumes that a coordinate point I is obtained, and the coordinate point I is calculated.
- the distance d(M, I) of the center point M of the three-dimensional coordinate system determines whether the distance d(M, I) is greater than the first threshold. When the distance d(M, I) is greater than the first threshold ⁇ , it indicates that the coordinate point I is an outlier, and it is determined that the physical state of the monitored object is abnormal.
- the first analysis determining module 120 may further determine whether the distance d(M, I) is less than a second threshold (the second threshold is less than the second threshold) a threshold).
- the second threshold is less than the second threshold
- the body health index model is updated by the coordinate point I, and the center point M is re-determined, that is, the new center point is recalculated.
- the aforementioned distance calculation may use an Euclidean distance, a Mahalanobis distance, or the like.
- the body health index model may also include any two kinds of information such as body temperature, blood pressure, and heart rate, and construct a two-dimensional coordinate system, and may even include only one of the information. , Construct a one-dimensional coordinate system.
- the invention is not limited thereto.
- the body health index model can use other methods in the prior art in addition to the coordinate system and its center point to determine whether the body state is abnormal.
- the present invention does not limit this.
- the abnormal alarm module 130 is set to perform an abnormal alarm when the physical state of the monitored object is abnormal.
- the abnormality alarm module 130 immediately performs an abnormal alarm to remind the relevant personnel.
- the abnormality alarm module 130 can directly dial the relevant telephone number (such as 110) to perform an alarm.
- the abnormal alarm module 130 may directly perform an alarm prompt, including voice prompts and/or visual information. Tips such as sounding an alarm, broadcasting voice messages, displaying text messages, displaying image information, and so on.
- the abnormal alarm module 130 may send an alarm message to the alarm prompting device (such as a mobile phone, a tablet, etc. carried by the guardian), and the alarm prompting device receives the alarm information.
- the alarm prompting device such as a mobile phone, a tablet, etc. carried by the guardian
- the alarm prompting device receives the alarm information.
- an alert prompt is displayed, including voice prompts and/or visual information prompts, such as sounding an alarm, broadcasting a voice message, displaying text information, displaying image information, and the like.
- the monitoring device of the embodiment of the present invention analyzes the physiological parameter information of the monitoring object, inputs the physiological parameter information into a preset body health index model, and analyzes whether the physical state of the monitoring object is abnormal, and determines the body of the monitoring object. If the status is abnormal, then an abnormal alarm is issued to prevent the elderly, children and other monitoring objects from being lost or abducted.
- the monitoring method of the embodiment of the present invention completely realizes automatic monitoring and automatic alarming, and is not limited to the distance between the monitoring object and the guardian, greatly expands the scope of application, and improves the practicality and implementation of the monitoring. Sex and accuracy.
- the embodiment of the present invention can construct a body health index model unique to the monitoring object by means of machine learning modeling, improve the accuracy of the detection result, and improve the reliability of the system.
- a monitoring apparatus according to a fourth embodiment of the present invention is proposed.
- This embodiment adds a second analysis judging module 140 to the first embodiment.
- the information receiving module 110 not only receives the physiological parameter information of the monitoring object, but also receives the daily activity information of the monitoring object reported by the information collecting device and the environment information of the environment.
- the physiological parameter information includes at least one of parameter information such as body temperature, blood pressure, and heart rate.
- the information collecting device collects three parameter information of body temperature, blood pressure, and heart rate of the monitored object.
- the daily activity information includes at least one of activity information such as location information and speed information.
- the information collection device collects two types of information: the location and the speed of the monitoring object.
- the speed includes the moving speed, and may also include the acceleration.
- the environment information includes at least one of temperature information and weather information, and may further include current daytime information.
- the information collecting device collects three types of current temperature, weather, and daytime information.
- the weather includes sunny days, rainy days, snowy days, and so on.
- the second analysis and determination module 140 is configured to: input daily activity information and environment information into a preset daily activity trajectory distribution model for analysis, and determine whether the activity state of the monitoring object is abnormal.
- the daily activity trajectory distribution model is also preset and stored in the monitoring device.
- the daily activity trajectory distribution model combines the daily activity information and environmental information of the monitored objects, and indicates the normal activities of the monitored objects under different combinations of different conditions, different temperatures and different weather conditions. There is no coercion or loss of subjectivity. The situation of consciousness. In general, it is necessary to combine the previous model, the body health index model, to make judgments.
- the daily activity trajectory distribution model is used to monitor anomalies. In the large amount of information collected, this information is regarded as a large number of discrete data points in high-dimensional space. Our purpose is to find objects in these data point objects that are different from most other objects. These abnormal data points are called outliers. point.
- the daily activity trajectory distribution model may be an initial daily activity trajectory distribution model obtained by the monitoring device directly from the outside, or may be established by the monitoring device by collecting statistical activity information and environmental information under normal conditions of the monitoring object. Specific body health index model.
- the monitoring object wears the information collecting device, and the information collecting device implements or fixes the collected monitoring object under normal conditions (such as when determining the monitoring object)
- the daily activity information and environmental information of the physical state are normal, and the collected daily activity information and environmental information are reported or reported (for example, once every minute) to the monitoring device, and the model creation module of the monitoring device is based on Obtain daily activity information and environmental information to train the daily activity trajectory distribution model of the monitored object.
- the environment information includes three dimensions: daytime, temperature, and weather
- the daily activity trajectory distribution model includes two-dimensional information of position and velocity
- the model creation module constructs a two-dimensional coordinate system as shown in FIG. 4 according to the two-dimensional information, for example, the X axis is position information, y The axis is speed information.
- the model creation module first determines the corresponding environment combination according to the environment information collected during the model learning creation period, and then inputs the position and speed in the daily activity information into the two-dimensional coordinate system corresponding to the environment combination to obtain corresponding coordinate points, according to the collection.
- the distribution of all the daily activity information in the two-dimensional coordinate system of the environment combination calculates a center point N, which is defined as the center point of the two-dimensional coordinate system, as a reference point for judging whether the activity state is abnormal.
- the coordinate points of the common activity information in the two-dimensional coordinate system acquired by the subsequent monitoring ⁇ are determined to be outliers if they are outside the preset range of the center point N, as shown in Figure 4, D, E, The three points of F are outliers, and the daily activity trajectory distribution model outputs abnormal results.
- the second analysis and determination module 140 determines a corresponding daily activity trajectory distribution model according to the environment combination corresponding to the collected environmental information, and inputs the position and velocity in the daily activity information into a daily activity trajectory distribution model.
- a coordinate point J is obtained, and the distance d(N, J) of the coordinate point J from the center point N of the three-dimensional coordinate system is calculated, and whether the distance d(N, J) is greater than the third threshold is determined. .
- the distance d(N, J) is greater than the third threshold ⁇ , it indicates that the coordinate point J is an outlier point, and the body state of the monitored object is determined.
- the second analysis determining module 140 may further determine whether the distance d(N, J) is less than a fourth threshold (the fourth threshold is less than the fourth threshold) Three thresholds).
- the fourth threshold is less than the fourth threshold
- the body health index model is updated by the coordinate point J, and the center N is re-determined, that is, the new center point is newly calculated.
- the aforementioned distance calculation may use an Euclidean distance, a Mahalanobis distance, or the like.
- the daily activity trajectory distribution model may also include any one of position and velocity to construct a one-dimensional coordinate system.
- the invention is not limited thereto.
- the daily activity trajectory distribution model may adopt other modes in the prior art, and the present invention does not limit the use of the coordinate system and its center point to determine whether the activity state is abnormal.
- the first analysis determining module 120 determines whether the physical state of the monitored object is abnormal.
- the second analysis determining module 140 determines the active state of the monitored object. Is it abnormal?
- the second analysis determining module 140 may first detect whether the active state of the monitored object is abnormal, and then the first analysis determining module 120 detects whether the physical state of the monitored object is abnormal.
- the invention is not limited thereto.
- the abnormality alarm module 130 is configured to: when the physical state or the active state of the monitored object is abnormal (such as in a remote place), an abnormal alarm is performed.
- This embodiment monitors the physical state and activity state of the monitored object by peers, further improves the accuracy of the detection result and the reliability of the system, and effectively prevents the monitoring objects such as the elderly and children from being lost or abducted.
- a monitoring apparatus according to a fifth embodiment of the present invention is proposed.
- This embodiment adds a model updating module 150 based on the second embodiment, and the model updating module 150 is configured to continuously update during the monitoring process.
- the body health index model and the daily activity trajectory distribution model are configured to continuously update during the monitoring process.
- the model updating module 150 uses the physiological parameter information collected this time to train the body health index model to continuously update the body health index model to ensure accurate monitoring of the monitoring device. Sex.
- the model updating module 150 uses the daily activity information and environmental information collected to train the daily activity trajectory distribution model to update the daily activity trajectory distribution model to ensure monitoring. The accuracy of device monitoring.
- the model updating module 150 adjusts the body health index model by using the collected physiological parameter information to update the body health index model.
- the model update module 150 adjusts the daily active trajectory distribution model by using the collected daily activity information and the environmental information to update the daily active trajectory distribution model. Thereby improving the monitoring of the monitoring device Accuracy.
- the user may also be allowed to set the temporary state.
- the model update module 150 suspends the use of the collected information (physiological parameter information or daily activity information). And environmental information) Update the body health index model or the daily activity trajectory distribution model to maintain the accuracy of the monitoring device.
- the model update module 150 does not adjust the body health index model or the daily activity trajectory distribution model. This avoids inappropriate modifications to the body health index model or the daily activity distribution trajectory model, increasing flexibility.
- the first analysis determination module 120 or the second analysis determination module 140 in the embodiment may be omitted to form a new embodiment.
- the monitoring device of the embodiment of the present invention constructs a body health index model and a daily activity trajectory distribution model unique to the monitoring object by means of machine learning modeling, and both models are in an uninterrupted learning state. Use the continuously learning and updated model to predict whether the physical state and active state of the monitored object are abnormal. It has realized the functions of intelligent anti-lost, anti-turning and anti-fraud for the elderly, children, mental retardation and people suffering from depression. Compared with the prior art, the automatic monitoring and automatic reporting are completely realized, which greatly improves the practicability, practicality and accuracy of the monitoring, and improves the accuracy of the detection result by continuously updating the model. System reliability.
- a monitoring system as shown in FIG. 8 may be constructed, where the monitoring system includes an information collecting device 20 and a monitoring device 10, where:
- the information collecting device 20 is configured to collect physiological parameter information of the monitoring object, and report the physiological parameter information to the monitoring device 10.
- the monitoring device 10 configured to input physiological parameter information into a preset body health index model for analysis, to determine whether the physical state of the monitored object is abnormal; and when the physical state of the monitored object is abnormal, perform an abnormality
- the monitoring device 10 is further configured to: when the physical state of the monitored subject is normal, update the body health index model using the physiological parameter information collected this time.
- the monitoring system further includes an alerting device 30, The monitoring device 10 pushes the alarm information to the alarm prompting device 30, and the alarm prompting device 30 performs an alarm prompt according to the alarm information.
- the alarm prompting device 30 is further configured to: receive the alarm confirmation information, and feed back the alarm error information to the monitoring device 10 when the alarm confirmation information is an alarm error.
- the monitoring device 10 is further configured to: receive an alert error message, and update the body health index model using the physiological parameter information.
- the information collecting device 20 is further configured to: collect the daily activity information of the monitoring object and the environmental information of the environment, and report the daily activity information and the environmental information to the monitoring device 10; the monitoring device 10 is further configured to: The daily activity information and the environmental information are input into a preset daily activity trajectory distribution model for analysis to determine whether the activity state of the monitoring object is abnormal. When the activity state of the monitoring object is abnormal, an abnormal alarm is performed.
- the monitoring device 10 is configured to: determine whether the activity state of the monitoring object is abnormal when the physical state of the monitoring object is normal.
- the monitoring device 10 is further configured to: when the activity state of the monitoring object is normal, update the daily activity trajectory distribution model by using the collected daily activity information and environmental information.
- the monitoring device 10 is further configured to: when receiving the alarm error information ⁇ for the abnormal alarm of the abnormal physical state, update the body health index model by using the collected physiological parameter information; when receiving the abnormality for the active state The alarm error information of the abnormal alarm is used to update the daily activity trajectory distribution model by using the daily activity information and environmental information collected this time.
- the monitoring device 10 is further configured to: when the current state is in a temporary state, suspend the use of the physiological parameter information or the daily activity information and the environmental information of the current collection to update the body health index model or the daily activity trajectory distribution model.
- the monitoring method and apparatus proposed by the embodiments of the present invention adopt a mode of automatic acquisition, modeling and prediction, and solve defects and deficiencies in the practicality, practicality and accuracy of the monitoring scheme in the prior art.
- the technical solution of the present invention which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk,
- a storage medium such as ROM/RAM, disk,
- the optical disc includes a number of instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the methods described in various embodiments of the present invention.
- a monitoring method provided by an embodiment of the present invention by collecting physiological parameter information of a monitoring object, and inputting physiological parameter information into a preset body health index model to analyze whether the physical state of the monitoring object is abnormal, when determining If the physical state of the monitored object is abnormal, then an abnormal alarm is issued to prevent the monitoring object of the elderly, children, etc. from being lost or abducted.
- the monitoring method of the embodiment of the present invention completely realizes automatic monitoring and automatic alarming, and is not limited to the distance between the monitoring object and the guardian, greatly expands the scope of application, and improves the practicality and implementation of the monitoring. Sex and accuracy.
- the embodiment of the present invention can construct a physical health index model and a daily activity trajectory distribution model unique to the monitoring object by means of machine learning modeling, and can utilize the collected physiological parameter information and daily activity information and environmental information in the monitoring process. Continuously adjust and update the body health index model and the daily activity trajectory distribution model, so that the two models are in an uninterrupted learning state, which greatly improves the accuracy of the test results and improves the reliability of the system.
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Heart & Thoracic Surgery (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Physiology (AREA)
- Cardiology (AREA)
- Artificial Intelligence (AREA)
- Pulmonology (AREA)
- Evolutionary Computation (AREA)
- Signal Processing (AREA)
- Psychiatry (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Fuzzy Systems (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Alarm Systems (AREA)
Abstract
一种监测方法,包括以下步骤:接收信息采集设备上报的监测对象的生理参数信息(S11);将所述生理参数信息输入预设的身体健康指数模型进行分析,判断监测对象的身体状态是否异常(S12);当监测对象的身体状态异常时,进行异常告警(S13)。还提供了一种监测装置。所述监测方法和监测装置实现了自动监测、自动告警,并且不受限于监测对象与监护人的距离,可有效防止老人、儿童等监测对象走丢或被拐骗。
Description
技术领域
[0001] 本发明涉及电子技术领域, 尤其是涉及一种监测方法和监测装置。
背景技术
[0002] 随着生活节奏的加快, 人们的工作和生活压力越来越大, 工作忙碌, 难以全天 候的细心照顾老人或儿童, 吋有老人、 小孩走丢的情况, 甚至出现儿童被拐骗 的悲剧。 针对前述问题, 现有技术中提出了针对老人和儿童的监测方案, 主要 包括以下两种:
[0003] 一种监测方案, 是通过监测对象主动向监护人上报告警信息来实现监测。 例如 , 申请号为 CN201610378951.1中国专利, 需要儿童在遇到危险吋及吋通过报警 器告知监护人。 但这种方案只能被动接收监测对象上报的告警信息, 不能主动 发现异常状况, 存在告警不及吋的情况, 特别是一些危险情况下, 当事人根本 不能或者没有机会主动上报。
[0004] 另一种监测方案, 是通过判断监测对象是否在监护人的设定范围内, 当超出设 定范围吋则进行告警提示来实现监测。 例如, 申请号为 CN201310647844.0的中 国专利, 依据儿童携带的信号接收装置在大人携带的信号发射装置的信号发射 范围外吋进行告警, 从而实现防走丢功能。 但这种方案受限于监测对象和监护 人的距离, 适用范围较窄, 准确性低。
[0005] 综上所述可知, 现有的监测方案, 实用性和实吋性较差, 准确性较低。
技术问题
[0006] 本发明实施例的主要目的在于提供一种监测方法和监测装置, 旨在解决现有的 监测方案实用性和实吋性较差, 准确性较低的技术问题。
问题的解决方案
技术解决方案
[0007] 为达以上目的, 一方面提出一种监测方法, 所述方法包括以下步骤:
[0008] 接收信息采集设备上报的监测对象的生理参数信息;
[0009] 将所述生理参数信息输入预设的身体健康指数模型进行分析, 判断所述监测对 象的身体状态是否异常;
[0010] 当所述监测对象的身体状态异常吋, 进行异常告警。
[0011] 另一方面, 提出一种监测装置, 所述装置包括:
[0012] 信息接收模块, 设置为接收信息采集设备上报的监测对象的生理参数信息; [0013] 第一分析判断模块, 设置为将所述生理参数信息输入预设的身体健康指数模型 进行分析, 判断所述监测对象的身体状态是否异常;
[0014] 异常告警模块, 设置为当所述监测对象的身体状态异常吋, 进行异常告警。
发明的有益效果
有益效果
[0015] 本发明实施例所提供的一种监测方法, 通过采集监测对象的生理参数信息, 将 生理参数信息输入到预设的身体健康指数模型中来分析监测对象的身体状态是 否异常, 当判定监测对象的身体状态异常吋, 则及吋进行异常告警, 从而有效 防止老人、 儿童等监测对象走丢或被拐骗。 与现有技术相比, 本发明实施例的 监测方法完全实现了自动监测、 自动告警, 并且不受限于监测对象与监护人的 距离, 大大扩大了适用范围, 提升了监测的实用性、 实吋性和准确性。 并且, 本发明实施例可以通过机器学习建模的方式, 构建监测对象特有的身体健康指 数模型和日常活动轨迹分布模型, 并且可以在监测过程中利用采集的生理参数 信息和日常活动信息及环境信息持续调整和更新身体健康指数模型和日常活动 轨迹分布模型, 使得两种模型处于不间断的学习状态, 大大提高了检测结果的 准确性, 提升了系统的可靠性。
对附图的简要说明
附图说明
[0016] 图 1是本发明第一实施例的监测方法的流程图;
[0017] 图 2是本发明实施例中身体健康指数模型的示意图;
[0018] 图 3是本发明第二实施例的监测方法的流程图;
[0019] 图 4是本发明实施例日常活动轨迹分布模型的示意图;
[0020] 图 5是本发明第三实施例的监测装置的模块示意图;
[0021] 图 6是本发明第四实施例的监测装置的模块示意图;
[0022] 图 7是本发明第五实施例的监测装置的模块示意图;
[0023] 图 8是本发明实施例中的监测系统的模块示意图。
[0024] 本发明目的的实现、 功能特点及优点将结合实施例, 参照附图做进一步说明。
本发明的实施方式
[0025] 应当理解, 此处所描述的具体实施例仅仅用以解释本发明, 并不用于限定本发 明。 下面将结合本发明实施例中的附图, 对本发明实施例中的技术方案进行清 楚、 完整地描述, 显然, 所描述的实施例仅仅是本发明的一部分实施例, 而不 是全部的实施例。 基于本发明中的实施例, 本领域普通技术人员在没有作出创 造性劳动前提下所获得的所有其他实施例, 都属于本发明保护的范围。
[0026] 需要说明, 本发明实施例中所有方向性指示 (诸如上、 下、 左、 右、 前、 后… …)仅用于解释在某一特定姿态 (如附图所示)下各部件之间的相对位置关系、 运 动情况等, 如果该特定姿态发生改变吋, 则该方向性指示也相应地随之改变。
[0027] 另外, 在本发明中涉及"第一"、 "第二"等的描述仅用于描述目的, 而不能理解 为指示或暗示其相对重要性或者隐含指明所指示的技术特征的数量。 由此, 限 定有"第一"、 "第二 "的特征可以明示或者隐含地包括至少一个该特征。 另外, 各 个实施例之间的技术方案可以相互结合, 但是必须是以本领域普通技术人员能 够实现为基础, 当技术方案的结合出现相互矛盾或无法实现吋应当认为这种技 术方案的结合不存在, 也不在本发明要求的保护范围之内。
[0028] 本发明实施例所述的监测对象, 主要指儿童、 老人、 特殊人群 (如心理障碍者 、 智力低下者等无独立生活能力的人群) 等需要监护的对象, 当然, 也可以是 宠物等其他对象。
[0029] 实施例一
[0030] 参见图 1, 提出本发明第一实施例的监测方法, 所述方法包括以下步骤:
[0031] Sl l、 接收信息采集设备上报的监测对象的生理参数信息。
[0032] 具体的, 信息采集设备佩戴在监测对象的身体上, 由监测对象随身携带。 信息 采集设备实吋或定吋的采集监测对象的生理参数信息, 并实吋或定吋的通过无
线通信方式向监测装置上报 (如每隔 1分钟上报一次) 监测对象的生理参数信息
[0033] 所述生理参数信息包括体温、 血压、 心率等参数信息中的至少一种。 本发明实 施例中, 信息采集设备采集监测对象的体温、 血压和心率三种参数信息。
[0034] S12、 将生理参数信息输入预设的身体健康指数模型进行分析, 判断监测对象 的身体状态是否异常。
[0035] 本发明实施例中, 预先设定了身体健康指数模型, 并存储于监测装置中。 所述 身体健康指数模型表示监测对象在体温、 血压、 心跳等维度的概率分布情况, 用以表征监测对象的身体处于正常状态, 不存在被胁迫或失去主观意识的情况 , 或者不存在突发疾病等状况。
[0036] 身体健康指数模型被用来监测异常情况。 在采集的大量信息中, 将这些信息看 作高维空间的大量离散数据点, 我们的目的是发现这些数据点对象中与大部分 其他对象不同的对象, 这些异常的数据点被称作离群点 (Outlier) 。
[0037] 所述身体健康指数模型可以是监测装置直接从外部获取的通用的身体健康指数 模型, 也可以是监测装置通过搜集统计监测对象正常状态下的生理参数而建立 的特定的身体健康指数模型。
[0038] 举例而言, 初次使用吋, 监测对象需要佩戴一段吋间 (模型学习创建周期) 的 信息采集设备来采集一定量的建模数据。 信息采集设备在模型学习创建周期 ( 如一天) 内实吋或定吋的采集监测对象正常状态下 (如身体健康吋) 的生理参 数信息, 并将采集的生理参数信息实吋或定吋的上报 (如每隔一分钟上报一次 ) 给监测装置, 监测装置根据获取到的体温、 血压、 心率等生理参数信息来训 练出监测对象的身体健康指数模型。
[0039] 本发明实施例中, 身体健康指数模型包括体温、 血压和心率三维信息, 监测装 置根据三维信息构建如图 2所示的三维坐标系, 例如, X轴为体温信息, y轴为血 压信息, z轴为心率信息。 监测装置将模型学习创建周期内采集的生理参数信息 输入到三维坐标系中获得对应的坐标点, 根据采集的所有生理参数信息在三维 坐标系中的分布, 计算出一个中心点 M, 定义为三维坐标系的中心点, 作为后续 判断身体状态是否异常的参考点, 需要注意的是, 这里所述的中心点并非指坐
标原点。 后续监测吋采集的生理参数信息在三维坐标系上的坐标点如果在中心 点 M的预设范围外, 则判别为离群点, 如图 2中的 A、 B、 C三点为离群点, 此吋 身体健康指数模型则输出异常结果。
[0040] 本步骤 S12中, 监测装置接收信息采集设备上报的生理参数信息, 并将采集的 生理参数信息中的体温、 血压和心率输入身体健康指数模型的三维坐标系中, 假设获得一个坐标点 I, 计算该坐标点 I与三维坐标系的中心点 M的距离 d(M,I), 判断该距离 d(M,I)是否大于第一阈值。 当距离 d(M,I)大于第一阈值吋, 说明该坐 标点 I是离群点, 判定监测对象的身体状态异常。
[0041] 进一步地, 当距离 d(M,I)不大于第一阈值吋, 监测装置还可以进一步判断该距 离 d(M,I)是否小于第二阈值 (第二阈值小于第一阈值)。 当距离 d(M,I)小于第二阈值 吋, 利用该坐标点 I更新身体健康指数模型, 重新确定中心点 M, 即重新计算出 新的中心点。 前述距离计算可以采用欧氏距离、 马氏距离等。
[0042] 在其它实施例中, 身体健康指数模型也可以只包括体温、 血压、 心率等参数信 息中的任意两种信息, 构建二维坐标系, 甚至可以只包括其中的任意一种信息 , 构建一维坐标系。 本发明对此不作限定。
[0043] 此外, 身体健康指数模型除了采用坐标系及其中心点来判断身体状态是否异常 夕卜, 还可以采用现有技术中的其他方式, 本发明对此不作限定。
[0044] S13、 当监测对象的身体状态异常吋, 进行异常告警。
[0045] 本步骤 S13中, 当通过身体健康指数模型判别出当前监测对象的身体状态异常 吋, 则立即进行异常告警, 以提醒相关人员。
[0046] 可选地, 监测装置可以直接拨打相关电话号码 (如 110) 进行报警。
[0047] 可选地, 当监测装置为终端设备 (如监护人员随身携带的移动终端或监控中心 的固定终端) 吋, 监测装置可以直接进行告警提示, 包括语音提示和 /或可视信 息提示, 如发出警报声、 播报语音信息、 显示文字信息、 显示图像信息等。
[0048] 可选地, 当监测装置为云端服务器吋, 监测装置则可以向告警提示设备 (如监 护人员随身携带的手机、 平板等移动终端) 推送告警信息, 告警提示设备接收 到告警信息后, 立即进行告警提示, 包括语音提示和 /或可视信息提示, 如发出 警报声、 播报语音信息、 显示文字信息、 显示图像信息等。
[0049] 进一步地, 告警提示设备进行告警提示后, 还允许用户 (如监护人) 对告警信 息手动进行告警确认, 接收用户的告警确认信息, 当告警确认信息为告警错误 , 即用户确认监测装置推送的异常告警不正确吋, 告警提示设备则立即向监测 装置反馈告警错误信息。 监测装置接收到告警错误信息吋, 则利用本次采集的 生理参数信息对身体健康指数模型进行调整, 以更新身体健康指数模型, 以提 高监测装置监测的准确性。
[0050] 进一步地, 本实施例中, 当监测对象的身体状态正常吋, 监测装置利用本次采 集的生理参数信息训练身体健康指数模型, 以对身体健康指数模型进行持续更 新, 以保证监测装置监测的准确性。
[0051] 进一步地, 还可以允许用户设置临吋状态, 在临吋状态下 (如健身吋、 生病吋 等) , 监测装置只利用采集的生理参数信息检测监测对象的身体状态, 暂停利 用生理参数信息更新身体健康指数模型。 例如, 当监测装置检测到监测对象身 体状态异常而进行异常告警后, 接收到告警错误信息吋, 监测装置不对身体健 康指数模型进行调整。 从而避免对身体健康指数模型的不恰当修改, 提高了灵 活性。
[0052] 本发明实施例的监测方法, 通过采集监测对象的生理参数信息, 将生理参数信 息输入到预设的身体健康指数模型中来分析监测对象的身体状态是否异常, 当 判定监测对象的身体状态异常吋, 则及吋进行异常告警, 从而有效防止老人、 儿童等监测对象走丢或被拐骗。 与现有技术相比, 本发明实施例的监测方法完 全实现了自动监测、 自动告警, 并且不受限于监测对象与监护人的距离, 大大 扩大了适用范围, 提升了监测的实用性、 实吋性和准确性。
[0053] 并且, 本发明实施例可以通过机器学习建模的方式, 构建监测对象特有的身体 健康指数模型, 并且可以在监测过程中利用采集的生理参数信息持续调整和更 新身体健康指数模型, 使得模型处于不间断的学习状态, 大大提高了检测结果 的准确性, 提升了系统的可靠性。
[0054] 参见图 3, 提出本发明第二实施例的监测方法, 所述方法包括以下步骤:
[0055] S21、 接收信息采集设备上报的监测对象的生理参数信息、 日常活动信息和所 处环境的环境信息。
具体的, 信息采集设备佩戴在监测对象的身体上, 由监测对象随身携带。 信息 采集设备实吋或定吋的采集监测对象的生理参数信息、 日常活动信息和所处环 境的环境信息, 并实吋或定吋的通过无线通信方式向监测装置上报 (如每隔 1分 钟上报一次) 生理参数信息、 日常活动信息和环境信息。
[0057] 所述生理参数信息包括体温、 血压、 心率等参数信息中的至少一种, 本发明实 施例中, 信息采集设备采集监测对象的体温、 血压和心率三种参数信息。
[0058] 所述日常活动信息包括位置信息、 速度信息等活动信息中的至少一种, 本发明 实施例中, 信息采集设备采集监测对象的位置和速度两种信息。 其中, 速度包 括移动速度, 还可以包括加速度。
[0059] 所述环境信息包括温度信息、 湿度信息和天气信息中的至少一种, 还可以包括 当前的吋间信息, 本发明实施例中, 信息采集设备采集当前的温度、 天气和吋 间三种信息。 其中, 天气包括晴天、 雨天、 雪天等。
[0060] S22、 将生理参数信息输入预设的身体健康指数模型进行分析, 判断监测对象 的身体状态是否异常。 当监测对象的身体状态异常吋, 进入步骤 S23 ; 当监测对 象的身体状态正常吋, 进入步骤 S24。
[0061] S23、 进行异常告警。
[0062] 本实施例中, 步骤 S22和 S23分别与第一实施例中的步骤 S12和 S13相同, 在此不 赘述。
[0063] S24、 将日常活动信息和环境信息输入预设的日常活动轨迹分布模型进行分析 , 判断监测对象的活动状态是否异常。
[0064] 本实施例中, 还预先设定了日常活动轨迹分布模型, 并存储于监测装置中。 日 常活动轨迹分布模型结合了监测对象的日常活动信息和环境信息等影响因素, 表示监测对象在不同吋间、 不同温度、 不同天气状况等组合条件下的正常活动 情况, 不存在被胁迫或失去主观意识的情况。 一般的, 此处需要结合上一个模 型即身体健康指数模型来做出判断。
[0065] 日常活动轨迹分布模型被用来监测异常情况。 在采集的大量信息中, 将这些信 息看作高维空间的大量离散数据点, 我们的目的是发现这些数据点对象中与大 部分其他对象不同的对象, 这些异常的数据点被称作离群点。
[0066] 所述日常活动轨迹分布模型可以是监测装置直接从外部获取的初始的日常活动 轨迹分布模型, 也可以是监测装置通过搜集统计监测对象正常状态下的日常活 动信息和环境信息而建立的特定的身体健康指数模型。
[0067] 举例而言, 在模型学习创建周期内 (如一天吋间) , 监测对象佩戴信息采集设 备, 信息采集设备实吋或定吋的采集监测对象在正常状态下 (如当判定监测对 象的身体状态正常吋) 的日常活动信息和环境信息, 并将采集的日常活动信息 和环境信息实吋或定吋的上报 (如每隔一分钟上报一次) 给监测装置, 监测装 置根据获取到的日常活动信息和环境信息来训练出监测对象的日常活动轨迹分 布模型。
[0068] 本发明实施例中, 环境信息包括吋间、 温度和天气三个维度, 将各个维度进行 量化分为至少两个区间, 不同的吋间段、 温度区间和天气状况的组合对应不同 的日常活动轨迹分布模型。 例如, 将吋间分为四个区间, 分别为以下吋间段: 0 0:00 05:00, 05:01-09:00, 09:01-18:00, 18:01-23:59; 温度分为三个区间, 分别 为: -40°~ -5°, -4°~ 15° , 16°~ 40°; 天气状况分为晴天、 雨天和雪天三个区间。 根据环境信息的三个维度的不同区间进行组合, 一共可以得到 4*3*3=12种组合 , 对于每一种组合都单独建立一个日常活动轨迹分布模型。
[0069] 本发明实施例中, 日常活动轨迹分布模型包括位置和速度二维信息, 监测装置 根据二维信息构建如图 4所示的二维坐标系, 例如, X轴为位置信息, y轴为速度 信息。 监测装置首先根据模型学习创建周期内采集的环境信息确定对应的环境 组合, 然后将日常活动信息中的位置和速度输入到该环境组合对应的二维坐标 系中获得对应的坐标点, 根据采集的该环境组合下所有的日常活动信息在二维 坐标系中的分布, 计算出一个中心点 N, 定义为二维坐标系的中心点, 作为后续 判断活动状态是否异常的参考点, 需要注意的是, 这里所述的中心点并非指坐 标原点。 后续监测吋采集的该环境组合下的日常活动信息在二维坐标系上的坐 标点如果在中心点 N的预设范围外, 则判别为离群点, 如图 4中的 D、 E、 F三点 为离群点, 此吋日常活动轨迹分布模型则输出异常结果。
[0070] 本步骤 S24中, 监测装置接收信息采集设备上报的日常活动信息和环境信息, 根据环境信息对应的环境组合确定对应的日常活动轨迹分布模型, 并将日常活
动信息中的位置和速度输入对应的日常活动轨迹分布模型的二维坐标系中, 假 设获得一个坐标点 J, 计算该坐标点 J与三维坐标系的中心点 N的距离 d(N,J), 判 断该距离 d(N,J)是否大于第三阈值。 当距离 d(N,J)大于第三阈值吋, 说明该坐标 点 J是离群点, 判定监测对象的身体状态异常。
[0071] 进一步地, 当距离 d(N,J)不大于第三阈值吋, 监测装置还可以进一步判断该距 离 d(N,J)是否小于第四阈值 (第四阈值小于第三阈值)。 当距离 d(N,J)小于第四阈值 吋, 利用该坐标点 J更新身体健康指数模型, 重新确定中心 N, 即重新计算出新 的中心点。 前述距离计算可以采用欧氏距离、 马氏距离等。
[0072] 在其它实施例中, 日常活动轨迹分布模型也可以只包括位置和速度中的任意一 种信息, 构建一维坐标系。 本发明对此不作限定。
[0073] 此外, 日常活动轨迹分布模型除了采用坐标系及其中心点来判断活动状态是否 异常外, 还可以采用现有技术中的其他方式, 本发明对此不作限定。
[0074] S25、 当监测对象的活动状态异常吋, 进行异常告警。
[0075] 本步骤 S25中, 当通过日常活动轨迹分布模型判别出监测对象当前的活动状态 异常 (如到了偏远的地方、 从未去过的地方或者移动速度异常等) 吋, 则立即 进行异常告警, 以提醒相关人员。
[0076] 可选地, 监测装置可以直接拨打相关电话号码 (如 110) 进行报警。
[0077] 可选地, 当监测装置为终端设备 (如监护人员随身携带的移动终端或监控中心 的固定终端) 吋, 监测装置可以直接进行告警提示, 包括语音提示和 /或可视信 息提示, 如发出警报声、 播报语音信息、 显示文字信息、 显示图像信息等。
[0078] 可选地, 当监测装置为云端服务器吋, 监测装置则可以向告警提示设备 (如监 护人员随身携带的手机、 平板等移动终端) 推送告警信息, 告警提示设备接收 到告警信息后, 立即进行告警提示, 包括语音提示和 /或可视信息提示, 如发出 警报声、 播报语音信息、 显示文字信息、 显示图像信息等。
[0079] 进一步地, 告警提示设备进行告警提示后, 还允许用户 (如监护人) 对告警信 息手动进行告警确认, 接收用户的告警确认信息, 当告警确认信息为告警错误 , 即用户确认监测装置推送的异常告警不正确吋, 告警提示设备则立即向监测 装置反馈告警错误信息。
[0080] 本实施例中, 当接收到针对身体状态异常的异常告警的告警错误信息吋, 监测 装置则利用本次采集的生理参数信息对身体健康指数模型进行调整, 以更新身 体健康指数模型。 当接收到针对活动状态异常的异常告警的告警错误信息, 监 测装置则利用本次采集的日常活动信息和环境信息对日常活动轨迹分布模型进 行调整, 以更新日常活动轨迹分布模型。 从而提高监测装置监测的准确性。
[0081] 具体实施吋, 监测装置可以通过短信通道向告警提示设备发送告警信息, 本实 施例中, 监测装置向告警提示设备上的特定应用推送告警信息。 告警提示设备 的特定应用还可以供用户査询监测对象当前的身体状态、 活动状态等信息, 同 吋, 在接收到监测装置推送的告警信息后, 用户可手动进行告警确认, 在确认 监测装置推送的异常告警不正确吋, 告警提示设备则可以通过该特定应用将告 警错误信息反馈给监测装置, 监测装置将根据用户反馈的告警错误信息自动进 行模型调整, 以得到更准确的分析结果。
[0082] 进一步地, 本实施例中, 当监测对象的身体状态正常吋, 监测装置利用本次采 集的生理参数信息训练身体健康指数模型, 以对身体健康指数模型进行持续更 新, 保证监测装置监测的准确性。 同吋, 进一步地, 当监测对象的活动状态正 常吋, 监测装置则利用本次采集的日常活动信息和环境信息训练日常活动轨迹 分布模型, 以对日常活动轨迹分布模型进行更新, 保证监测装置监测的准确性
[0083] 进一步地, 还可以允许用户设置临吋状态, 在临吋状态下 (如健身吋、 生病吋 、 旅游吋等) , 监测装置只利用采集的信息 (生理参数信息或者日常活动信息 及环境信息) 检测监测对象的身体状态或活动状态, 监测装置暂停利用采集的 信息 (生理参数信息或者日常活动信息及环境信息) 更新身体健康指数模型或 日常活动轨迹分布模型。
[0084] 例如, 当监测装置检测到监测对象身体状态异常或活动状态异常而进行异常告 警后, 接收到告警错误信息吋, 监测装置不对身体健康指数模型或日常活动轨 迹分布模型进行调整, 从而避免对身体健康指数模型或日常活动分布轨迹模型 的不恰当修改。
[0085] 本实施例中, 先检测监测对象的身体状态是否异常, 再检测监测对象的活动状
态是否异常。 实际上, 在其它实施例中, 也可以反过来, 先检测监测对象的活 动状态是否异常, 再检测监测对象的身体状态是否异常。 本发明对此不作限定
[0086] 本实施例通过同吋对监测对象的身体状态和活动状态进行监测, 进一步提高了 检测结果的准确性以及系统的可靠性, 有效防止老人、 儿童等监测对象走丢或 被拐骗。
[0087] 本发明实施例的监测方法, 通过机器学习建模的方式, 构建监测对象特有的身 体健康指数模型和日常活动轨迹分布模型, 且两个模型都处于不间断的学习状 态。 使用不断学习、 更新的模型实吋预测监测对象的身体状态和活动状态是否 发生异常。 实现了对老人、 小孩、 智力低下以及患有抑郁症的人群智能防走丢 、 防拐防骗的功能。 与现有技术相比, 完全实现了自动监测、 自动上报, 大大 提升了监测的实用性、 实吋性和准确性, 同吋通过对模型的不断更新, 提高了 检测结果的准确性, 提升了系统的可靠性。
[0088] 实施例三
[0089] 参见图 5, 提出本发明第三实施例的监测装置, 所述装置包括信息接收模块 110
、 第一分析判断模块 120和异常告警模块 130, 其中:
[0090] 信息接收模块 110: 设置为接收信息采集设备上报的监测对象的生理参数信息
[0091] 所述生理参数信息包括体温、 血压、 心率等参数信息中的至少一种。 本发明实 施例中, 信息采集设备采集监测对象的体温、 血压和心率三种参数信息。
[0092] 第一分析判断模块 120: 设置为将生理参数信息输入预设的身体健康指数模型 进行分析, 判断监测对象的身体状态是否异常。
[0093] 本发明实施例中, 预先设定了身体健康指数模型, 并存储于监测装置中。 所述 身体健康指数模型表示监测对象在体温、 血压、 心跳等维度的概率分布情况, 用以表征监测对象的身体处于正常状态, 不存在被胁迫或失去主观意识的情况 , 或者不存在突发疾病等状况。
[0094] 身体健康指数模型被用来监测异常情况。 在采集的大量信息中, 将这些信息看 作高维空间的大量离散数据点, 我们的目的是发现这些数据点对象中与大部分
其他对象不同的对象, 这些异常的数据点被称作离群点 (Outlier) 。
[0095] 所述身体健康指数模型可以是监测装置直接从外部获取的通用的身体健康指数 模型, 也可以是监测装置通过搜集统计监测对象正常状态下的生理参数而建立 的特定的身体健康指数模型。
[0096] 举例而言, 监测装置包括模型创建模块, 在模型学习创建周期内 (如一天吋间 ) , 监测对象佩戴信息采集设备, 信息采集设备实吋或定吋的采集监测对象正 常状态下 (如身体健康吋) 的生理参数信息, 并将采集的生理参数信息实吋或 定吋的上报 (如每隔一分钟上报一次) 给监测装置, 模型创建模块根据获取到 的体温、 血压、 心率等生理参数信息来训练出监测对象的身体健康指数模型。
[0097] 本发明实施例中, 身体健康指数模型包括体温、 血压和心率三维信息, 模型创 建模块根据三维信息构建如图 2所示的三维坐标系, 例如, X轴为体温信息, y轴 为血压信息, z轴为心率信息。 模型创建模块将模型学习创建周期内采集的生理 参数信息输入到三维坐标系中获得对应的坐标点, 根据采集的所有生理参数信 息在三维坐标系中的分布, 计算出一个中心点 M, 定义为三维坐标系的中心点, 作为后续判断身体状态是否异常的参考点。 后续监测吋采集的生理参数信息在 三维坐标系上的坐标点如果在中心点 M的预设范围外, 则判别为离群点, 如图 2 中的 A、 B、 C三点为离群点, 此吋身体健康指数模型则输出异常结果。
[0098] 监测过程中, 第一分析判断模块 120将采集的生理参数信息中的体温、 血压和 心率输入身体健康指数模型的三维坐标系中, 假设获得一个坐标点 I, 计算该坐 标点 I与三维坐标系的中心点 M的距离 d(M,I), 判断该距离 d(M,I)是否大于第一阈 值。 当距离 d(M,I)大于第一阈值吋, 说明该坐标点 I是离群点, 判定监测对象的 身体状态异常。
[0099] 进一步地, 当距离 d(M,I)不大于第一阈值吋, 第一分析判断模块 120还可以进一 步判断该距离 d(M,I)是否小于第二阈值 (第二阈值小于第一阈值)。 当距离 d(M,I)小 于第二阈值吋, 利用该坐标点 I更新身体健康指数模型, 重新确定中心点 M, 即 重新计算出新的中心点。 前述距离计算可以采用欧氏距离、 马氏距离等。
[0100] 在其它实施例中, 身体健康指数模型也可以只包括体温、 血压、 心率等参数信 息中的任意两种信息, 构建二维坐标系, 甚至可以只包括其中的任意一种信息
, 构建一维坐标系。 本发明对此不作限定。
[0101] 此外, 身体健康指数模型除了采用坐标系及其中心点来判断身体状态是否异常 夕卜, 还可以采用现有技术中的其他方式, 本发明对此不作限定。
[0102] 异常告警模块 130: 设置为当监测对象的身体状态异常吋, 进行异常告警。
[0103] 具体的, 当第一分析判断模块 120通过身体健康指数模型判别出当前监测对象 的身体状态异常吋, 异常告警模块 130则立即进行异常告警, 以提醒相关人员。
[0104] 可选地, 异常告警模块 130可以直接拨打相关电话号码 (如 110) 进行报警。
[0105] 可选地, 当监测装置为终端设备 (如监护人员随身携带的移动终端或监控中心 的固定终端) 吋, 异常告警模块 130可以直接进行告警提示, 包括语音提示和 /或 可视信息提示, 如发出警报声、 播报语音信息、 显示文字信息、 显示图像信息 等。
[0106] 可选地, 当监测装置为云端服务器吋, 异常告警模块 130则可以向告警提示设 备 (如监护人员随身携带的手机、 平板等移动终端) 推送告警信息, 告警提示 设备接收到告警信息后, 立即进行告警提示, 包括语音提示和 /或可视信息提示 , 如发出警报声、 播报语音信息、 显示文字信息、 显示图像信息等。
[0107] 本发明实施例的监测装置, 通过采集监测对象的生理参数信息, 将生理参数信 息输入到预设的身体健康指数模型中来分析监测对象的身体状态是否异常, 当 判定监测对象的身体状态异常吋, 则及吋进行异常告警, 从而有效防止老人、 儿童等监测对象走丢或被拐骗。 与现有技术相比, 本发明实施例的监测方法完 全实现了自动监测、 自动告警, 并且不受限于监测对象与监护人的距离, 大大 扩大了适用范围, 提升了监测的实用性、 实吋性和准确性。
[0108] 并且, 本发明实施例可以通过机器学习建模的方式, 构建监测对象特有的身体 健康指数模型, 提高了检测结果的准确性, 提升了系统的可靠性。
[0109] 实施例四
[0110] 参照图 6, 提出本发明第四实施例的监测装置, 本实施例在第一实施例的基础 上增加了第二分析判断模块 140。
[0111] 本实施例中, 信息接收模块 110不但接收监测对象的生理参数信息, 还接收信 息采集设备上报的监测对象的日常活动信息和所处环境的环境信息。
[0112] 所述生理参数信息包括体温、 血压、 心率等参数信息中的至少一种, 本发明实 施例中, 信息采集设备采集监测对象的体温、 血压和心率三种参数信息。
[0113] 所述日常活动信息包括位置信息、 速度信息等活动信息中的至少一种, 本发明 实施例中, 信息采集设备采集监测对象的位置和速度两种信息。 其中, 速度包 括移动速度, 还可以包括加速度。
[0114] 所述环境信息包括温度信息和天气信息中至少一种, 还可以包括当前的吋间信 息, 本发明实施例中, 信息采集设备采集当前的温度、 天气和吋间三种信息。 其中, 天气包括晴天、 雨天、 雪天等。
[0115] 本实施例中, 第二分析判断模块 140设置为: 将日常活动信息和环境信息输入 预设的日常活动轨迹分布模型进行分析, 判断监测对象的活动状态是否异常。
[0116] 本实施例还预先设定了日常活动轨迹分布模型, 并存储于监测装置中。 日常活 动轨迹分布模型结合了监测对象的日常活动信息和环境信息等影响因素, 表示 监测对象在不同吋间、 不同温度、 不同天气状况等组合条件下的正常活动情况 , 不存在被胁迫或失去主观意识的情况。 一般的, 此处需要结合上一个模型即 身体健康指数模型来做出判断。
[0117] 日常活动轨迹分布模型被用来监测异常情况。 在采集的大量信息中, 将这些信 息看作高维空间的大量离散数据点, 我们的目的是发现这些数据点对象中与大 部分其他对象不同的对象, 这些异常的数据点被称作离群点。
[0118] 所述日常活动轨迹分布模型可以是监测装置直接从外部获取的初始的日常活动 轨迹分布模型, 也可以是监测装置通过搜集统计监测对象正常状态下的日常活 动信息和环境信息而建立的特定的身体健康指数模型。
[0119] 举例而言, 在模型学习创建周期内 (如一天吋间) , 监测对象佩戴信息采集设 备, 信息采集设备实吋或定吋的采集监测对象在正常状态下 (如当判定监测对 象的身体状态正常吋) 的日常活动信息和环境信息, 并将采集的日常活动信息 和环境信息实吋或定吋的上报 (如每隔一分钟上报一次) 给监测装置, 监测装 置的模型创建模块根据获取到的日常活动信息和环境信息来训练出监测对象的 日常活动轨迹分布模型。
[0120] 本发明实施例中, 环境信息包括吋间、 温度和天气三个维度, 模型创建模块将
各个维度进行量化分为至少两个区间, 不同的吋间段、 温度区间和天气状况的 组合对应不同的日常活动轨迹分布模型。 例如, 将吋间分为四个区间, 分别为 以下吋间段: 00:00 05:00, 05:01-09:00, 09:01-18:00, 18:01-23:59; 温度分为 三个区间, 分别为: -40°~ -5°, -4°~ 15° , 16°~ 40°; 天气状况分为晴天、 雨天和 雪天三个区间。 根据环境信息的三个维度的不同区间进行组合, 一共可以得到 4 *3*3=12种组合, 对于每一种组合都单独建立一个日常活动轨迹分布模型。
[0121] 本发明实施例中, 日常活动轨迹分布模型包括位置和速度二维信息, 模型创建 模块根据二维信息构建如图 4所示的二维坐标系, 例如, X轴为位置信息, y轴为 速度信息。 模型创建模块首先根据模型学习创建周期内采集的环境信息确定对 应的环境组合, 然后将日常活动信息中的位置和速度输入到该环境组合对应的 二维坐标系中获得对应的坐标点, 根据采集的该环境组合下所有的日常活动信 息在二维坐标系中的分布, 计算出一个中心点 N, 定义为二维坐标系的中心点, 作为后续判断活动状态是否异常的参考点。 后续监测吋采集的该环境组合下的 曰常活动信息在二维坐标系上的坐标点如果在中心点 N的预设范围外, 则判别为 离群点, 如图 4中的 D、 E、 F三点为离群点, 此吋日常活动轨迹分布模型则输出 异常结果。
[0122] 本实施例中, 第二分析判断模块 140根据采集的环境信息对应的环境组合确定 对应的日常活动轨迹分布模型, 并将日常活动信息中的位置和速度输入对应的 日常活动轨迹分布模型的二维坐标系中, 假设获得一个坐标点 J, 计算该坐标点 J 与三维坐标系的中心点 N的距离 d(N,J), 判断该距离 d(N, J)是否大于第三阈值。 当 距离 d(N,J)大于第三阈值吋, 说明该坐标点 J是离群点, 判定监测对象的身体状态 计吊。
[0123] 进一步地, 当距离 d(N,J)不大于第三阈值吋, 第二分析判断模块 140还可以进一 步判断该距离 d(N,J)是否小于第四阈值 (第四阈值小于第三阈值)。 当距离 d(N,J)小 于第四阈值吋, 利用该坐标点 J更新身体健康指数模型, 重新确定中心 N, 即重 新计算出新的中心点。 前述距离计算可以采用欧氏距离、 马氏距离等。
[0124] 在其它实施例中, 日常活动轨迹分布模型也可以只包括位置和速度中的任意一 种信息, 构建一维坐标系。 本发明对此不作限定。
[0125] 此外, 日常活动轨迹分布模型除了采用坐标系及其中心点来判断活动状态是否 异常外, 还可以采用现有技术中的其他方式, 本发明对此不作限定。
[0126] 作为优选, 本实施例中, 首先由第一分析判断模块 120判断监测对象的身体状 态是否异常, 当监测对象的身体状态正常吋, 第二分析判断模块 140才判断监测 对象的活动状态是否异常。
[0127] 当然, 在其它实施例中, 也可以反过来, 先由第二分析判断模块 140检测监测 对象的活动状态是否异常, 再由第一分析判断模块 120检测监测对象的身体状态 是否异常。 本发明对此不作限定。
[0128] 本实施例中, 异常告警模块 130设置为: 当监测对象的身体状态或活动状态异 常 (如到了偏远的地方) 吋, 进行异常告警。
[0129] 本实施例通过同吋对监测对象的身体状态和活动状态进行监测, 进一步提高了 检测结果的准确性以及系统的可靠性, 有效防止老人、 儿童等监测对象走丢或 被拐骗。
[0130] 实施例五
[0131] 参见图 7, 提出本发明第五实施例的监测装置, 本实施例在第二实施例的基础 上增加了模型更新模块 150, 所述模型更新模块 150设置为在监测过程中持续更 新身体健康指数模型和日常活动轨迹分布模型。
[0132] 具体的, 当监测对象的身体状态正常吋, 模型更新模块 150利用本次采集的生 理参数信息训练身体健康指数模型, 以对身体健康指数模型进行持续更新, 以 保证监测装置监测的准确性。 同吋, 进一步地, 当监测对象的活动状态正常吋 , 模型更新模块 150则利用本次采集的日常活动信息和环境信息训练日常活动轨 迹分布模型, 以对日常活动轨迹分布模型进行更新, 保证监测装置监测的准确 性。
[0133] 进一步地, 当接收到针对身体状态异常的异常告警的告警错误信息吋, 模型更 新模块 150则利用本次采集的生理参数信息对身体健康指数模型进行调整, 以更 新身体健康指数模型。 当接收到针对活动状态异常的异常告警的告警错误信息 , 模型更新模块 150则利用本次采集的日常活动信息和环境信息对日常活动轨迹 分布模型进行调整, 以更新日常活动轨迹分布模型。 从而提高监测装置监测的
准确性。
[0134] 进一步地, 还可以允许用户设置临吋状态, 在临吋状态下 (如健身吋、 生病吋 、 旅游吋等) , 模型更新模块 150暂停利用采集的信息 (生理参数信息或者日常 活动信息及环境信息) 更新身体健康指数模型或日常活动轨迹分布模型, 以保 持监测装置检测的准确性。
[0135] 例如, 当监测装置检测到监测对象身体状态异常或活动状态异常而进行异常告 警后, 接收到告警错误信息吋, 模型更新模块 150不对身体健康指数模型或日常 活动轨迹分布模型进行调整, 从而避免对身体健康指数模型或日常活动分布轨 迹模型的不恰当修改, 提高了灵活性。
[0136] 可选地, 也可以省略本实施例中的第一分析判断模块 120或第二分析判断模块 1 40而形成新的实施例。
[0137] 本发明实施例的监测装置, 通过机器学习建模的方式, 构建监测对象特有的身 体健康指数模型和日常活动轨迹分布模型, 且两个模型都处于不间断的学习状 态。 使用不断学习、 更新的模型实吋预测监测对象的身体状态和活动状态是否 发生异常。 实现了对老人、 小孩、 智力低下以及患有抑郁症的人群智能防走丢 、 防拐防骗的功能。 与现有技术相比, 完全实现了自动监测、 自动上报, 大大 提升了监测的实用性、 实吋性和准确性, 同吋通过对模型的不断更新, 提高了 检测结果的准确性, 提升了系统的可靠性。
[0138] 为实现本发明实施例的监测方案, 在具体实施吋, 可以搭建如图 8所示的监测 系统, 所述监测系统包括信息采集设备 20、 监测装置 10, 其中:
[0139] 信息采集设备 20: 设置为采集监测对象的生理参数信息, 并将生理参数信息上 报给监测装置 10。
[0140] 监测装置 10: 设置为将生理参数信息输入预设的身体健康指数模型进行分析, 判断监测对象的身体状态是否异常; 当监测对象的身体状态异常吋, 进行异常
[0141] 进一步地, 监测装置 10还设置为: 当监测对象的身体状态正常吋, 利用本次采 集的生理参数信息更新身体健康指数模型。
[0142] 进一步地, 当监测装置 10为云端服务器吋, 监测系统还包括告警提示设备 30,
监测装置 10向告警提示设备 30推送告警信息, 告警提示设备 30根据告警信息进 行告警提示。
[0143] 进一步地, 告警提示设备 30还设置为: 接收告警确认信息, 当告警确认信息为 告警错误吋, 向监测装置 10反馈告警错误信息。 监测装置 10还设置为: 接收告 警错误信息, 利用生理参数信息更新身体健康指数模型。
[0144] 进一步地, 信息采集设备 20还设置为: 采集监测对象的日常活动信息和所处环 境的环境信息, 并将日常活动信息和环境信息上报给监测装置 10; 监测装置 10 还设置为: 将日常活动信息和环境信息输入预设的日常活动轨迹分布模型进行 分析, 判断监测对象的活动状态是否异常; 当监测对象的活动状态异常吋, 进 行异常告警。
[0145] 更进一步地, 监测装置 10设置为: 当监测对象的身体状态正常吋, 才判断监测 对象的活动状态是否异常。
[0146] 进一步地, 监测装置 10还设置为: 当监测对象的活动状态正常吋, 利用本次采 集的日常活动信息和环境信息更新日常活动轨迹分布模型。
[0147] 进一步地, 监测装置 10还设置为: 当接收到针对身体状态异常的异常告警的告 警错误信息吋, 利用本次采集的生理参数信息更新身体健康指数模型; 当接收 到针对活动状态异常的异常告警的告警错误信息, 利用本次采集的日常活动信 息和环境信息更新日常活动轨迹分布模型。
[0148] 进一步地, 监测装置 10还设置为: 当当前处于临吋状态吋, 暂停利用本次采集 的生理参数信息或日常活动信息及环境信息更新身体健康指数模型或日常活动 轨迹分布模型。
[0149] 需要说明的是: 上述实施例提供的监测系统与监测方法实施例属于同一构思, 其具体实现过程详见方法实施例, 且方法实施例中的技术特征在系统实施例中 均对应适用, 这里不再赘述。
[0150] 本发明实施例提出的监测方法和装置, 采用自动采集、 建模和预测的模式, 解 决了现有技术中的监测方案在实用性、 实吋性和准确性上的缺陷和不足。
[0151] 通过以上的实施方式的描述, 本领域的技术人员可以清楚地了解到上述实施例 方法可借助软件加必需的通用硬件平台的方式来实现, 当然也可以通过硬件,
但很多情况下前者是更佳的实施方式。 基于这样的理解, 本发明的技术方案本 质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来, 该计 算机软件产品存储在一个存储介质 (如 ROM/RAM、 磁碟、 光盘) 中, 包括若干 指令用以使得一台终端设备 (可以是手机, 计算机, 服务器, 或者网络设备等 ) 执行本发明各个实施例所述的方法。
[0152] 以上参照附图说明了本发明的优选实施例, 并非因此局限本发明的权利范围。
本领域技术人员不脱离本发明的范围和实质, 可以有多种变型方案实现本发明 , 比如作为一个实施例的特征可用于另一实施例而得到又一实施例。 凡在运用 本发明的技术构思之内所作的任何修改、 等同替换和改进, 均应在本发明的权 利范围之内。
工业实用性
[0153] 本发明实施例所提供的一种监测方法, 通过采集监测对象的生理参数信息, 将 生理参数信息输入到预设的身体健康指数模型中来分析监测对象的身体状态是 否异常, 当判定监测对象的身体状态异常吋, 则及吋进行异常告警, 从而有效 防止老人、 儿童等监测对象走丢或被拐骗。 与现有技术相比, 本发明实施例的 监测方法完全实现了自动监测、 自动告警, 并且不受限于监测对象与监护人的 距离, 大大扩大了适用范围, 提升了监测的实用性、 实吋性和准确性。 并且, 本发明实施例可以通过机器学习建模的方式, 构建监测对象特有的身体健康指 数模型和日常活动轨迹分布模型, 并且可以在监测过程中利用采集的生理参数 信息和日常活动信息及环境信息持续调整和更新身体健康指数模型和日常活动 轨迹分布模型, 使得两种模型处于不间断的学习状态, 大大提高了检测结果的 准确性, 提升了系统的可靠性。
Claims
权利要求书
[权利要求 1] 一种监测方法, 包括以下步骤:
接收信息采集设备上报的监测对象的生理参数信息;
将所述生理参数信息输入预设的身体健康指数模型进行分析, 判断所 述监测对象的身体状态是否异常;
当所述监测对象的身体状态异常吋, 进行异常告警。
[权利要求 2] 根据权利要求 1所述的监测方法, 其中, 所述判断所述监测对象的身 体状态是否异常的步骤之后还包括:
当所述监测对象的身体状态正常吋, 利用所述生理参数信息更新所述 身体健康指数模型。
[权利要求 3] 根据权利要求 1所述的监测方法, 其中, 所述进行异常告警包括: 向告警提示设备推送告警信息, 以使所述告警提示设备根据所述告警 信息进行告警提示。
[权利要求 4] 根据权利要求 3所述的监测方法, 其中, 所述向告警提示设备推送告 警信息的步骤之后还包括:
当接收到所述告警提示设备反馈的告警错误信息吋, 利用所述生理参 数信息更新所述身体健康指数模型。
[权利要求 5] 根据权利要求 1所述的监测方法, 其中, 所述方法还包括:
接收所述信息采集设备上报的监测对象的日常活动信息和所处环境的 环境信息;
将所述日常活动信息和所述环境信息输入预设的日常活动轨迹分布模 型进行分析, 判断所述监测对象的活动状态是否异常;
当所述监测对象的活动状态异常吋, 进行异常告警。
[权利要求 6] 根据权利要求 5所述的监测方法, 其中, 当所述监测对象的身体状态 正常吋, 才判断所述监测对象的活动状态是否异常。
[权利要求 7] 根据权利要求 6所述的监测方法, 其中, 所述判断所述监测对象的活 动状态是否异常的步骤之后还包括:
当所述监测对象的活动状态正常吋, 利用所述日常活动信息和所述环
境信息更新所述日常活动轨迹分布模型。
根据权利要求 5所述的监测方法, 其中, 所述监测装置进行异常告警 包括:
向告警提示设备推送告警信息, 以使所述告警提示设备根据所述告警 信息进行告警提示。
根据权利要求 8所述的监测方法, 其中, 所述向告警提示设备推送告 警信息的步骤之后还包括:
当接收到针对所述身体状态异常的异常告警的告警错误信息吋, 利用 所述生理参数信息更新所述身体健康指数模型;
当接收到针对所述活动状态异常的异常告警的告警错误信息, 利用所 述曰常活动信息和所述环境信息更新所述日常活动轨迹分布模型。 根据权利要求 1所述的监测方法, 其中, 所述生理参数信息包括体温 、 血压和心率, 所述判断所述监测对象的身体状态是否异常包括: 将所述体温、 血压和心率输入所述身体健康指数模型的三维坐标系中 , 在所述三维坐标系中获得一个坐标点;
计算所述坐标点与所述三维坐标系的中心点的距离, 判断所述距离是 否大于第一阈值; 所述三维坐标系的中心点为: 根据采集的所有生理 参数信息在三维坐标系中的分布而计算出的中心点;
当所述距离大于所述第一阈值吋, 判定所述监测对象的身体状态异常 根据权利要求 10所述的监测方法, 其中, 所述判断所述距离是否大于 第一阈值的步骤之后还包括:
当所述距离不大于所述第一阈值吋, 判断所述距离是否小于第二阈值 当所述距离小于所述第二阈值吋, 利用所述坐标点重新确定所述中心 点; 其中, 所述第一阈值大于所述第二阈值。
根据权利要求 5所述的监测方法, 其中, 所述环境信息包括吋间、 温 度和天气, 不同的吋间段、 温度区间和天气状况的组合对应不同的曰
常活动轨迹分布模型。
根据权利要求 12所述的监测方法, 其中, 所述日常活动信息包括位置 和速度, 所述判断所述监测对象的活动状态是否异常包括: 将所述位置和速度输入对应的日常活动轨迹分布模型的二维坐标系中 , 在所述二维坐标系中获得一个坐标点;
计算所述坐标点与所述二维坐标系的中心点的距离, 判断所述距离是 否大于第三阈值; 所述二维坐标系的中心点为: 根据采集的环境组合 下所有的日常活动信息在二维坐标系中的分布而计算出的中心点; 当所述距离大于所述第三阈值吋, 判定所述监测对象的活动状态异常 根据权利要求 13所述的监测方法, 其中, 所述判断所述距离是否大于 第三阈值的步骤之后还包括:
当所述距离不大于所述第三阈值吋, 判断所述距离是否小于第四阈值 当所述距离小于所述第四阈值吋, 利用所述坐标点重新确定所述中心 点; 其中, 所述第三阈值大于所述第四阈值。
根据权利要求 1所述的监测方法, 其中, 所述方法还包括: 当当前处于临吋状态吋, 暂停利用所述生理参数信息更新所述身体健 康指数模型。
一种监测装置, 包括:
信息接收模块, 设置为接收信息采集设备上报的监测对象的生理参数 f π息;
第一分析判断模块, 设置为将所述生理参数信息输入预设的身体健康 指数模型进行分析, 判断所述监测对象的身体状态是否异常; 异常告警模块, 设置为当所述监测对象的身体状态异常吋, 进行异常 根据权利要求 16所述的监测装置, 其中, 还包括模型更新模块, 所述 模型更新模块设置为: 当所述监测对象的身体状态正常吋, 利用所述
生理参数信息更新所述身体健康指数模型。
[权利要求 18] 根据权利要求 16所述的监测装置, 其中, 所述异常告警模块设置为: 向所述告警提示设备推送告警信息, 以使所述告警提示设备根据所述 告警信息进行告警提示。
[权利要求 19] 根据权利要求 18所述的监测装置, 其中, 还包括模型更新模块, 所述 模型更新模块设置为: 当接收到所述告警提示设备反馈的告警错误信 息吋, 利用所述生理参数信息更新所述身体健康指数模型。
[权利要求 20] 根据权利要求 16所述的监测装置, 其中,
所述信息接收模块还设置为: 接收所述信息采集设备上报的监测对象 的曰常活动信息和所处环境的环境信息;
所述装置还包括第二分析判断模块, 所述第二分析判断模块设置为: 将所述日常活动信息和所述环境信息输入预设的日常活动轨迹分布模 型进行分析, 判断所述监测对象的活动状态是否异常;
所述异常告警模块还设置为: 当所述监测对象的活动状态异常吋, 进 行异常告警。
[权利要求 21] 根据权利要求 20所述的监测装置, 其中, 所述第二分析判断模块设置 为: 当所述监测对象的身体状态正常吋, 才判断所述监测对象的活动 状态是否异常。
[权利要求 22] 根据权利要求 21所述的监测装置, 其中, 还包括模型更新模块, 所述 模型更新模块设置为:
当所述监测对象的活动状态正常吋, 利用所述日常活动信息和所述环 境信息更新所述日常活动轨迹分布模型。
[权利要求 23] 根据权利要求 20所述的监测装置, 其中, 所述异常告警模块设置为: 向所述告警提示设备推送告警信息, 以使所述告警提示设备根据所述 告警信息进行告警提示。
[权利要求 24] 根据权利要求 23所述的监测装置, 其中, 还包括模型更新模块, 所述 模型更新模块设置为:
当接收到针对所述身体状态异常的异常告警的告警错误信息吋, 利用
所述生理参数信息更新所述身体健康指数模型;
当接收到针对所述活动状态异常的异常告警的告警错误信息, 利用所 述曰常活动信息和所述环境信息更新所述日常活动轨迹分布模型。 根据权利要求 16所述的监测装置, 其中, 所述生理参数信息包括体温 、 血压和心率, 所述第一分析判断模块设置为:
将所述体温、 血压和心率输入所述身体健康指数模型的三维坐标系中 , 在所述三维坐标系中获得一个坐标点; 计算所述坐标点与所述三维 坐标系的中心点的距离, 判断所述距离是否大于第一阈值; 当所述距 离大于所述第一阈值吋, 判定所述监测对象的身体状态异常; 其中, 所述三维坐标系的中心点为: 根据采集的所有生理参数信息在三维坐 标系中的分布而计算出的中心点。
根据权利要求 25所述的监测装置, 其中, 所述第一分析判断模块还设 置为: 当所述距离不大于所述第一阈值吋, 判断所述距离是否小于第 二阈值; 当所述距离小于所述第二阈值吋, 利用所述坐标点重新确定 所述中心点; 其中, 所述第一阈值大于所述第二阈值。
根据权利要求 20所述的监测装置, 其中, 所述环境信息包括吋间、 温 度和天气, 不同的吋间段、 温度区间和天气状况的组合对应不同的曰 常活动轨迹分布模型。
根据权利要求 27所述的监测装置, 其中, 所述日常活动信息包括位置 和速度, 所述第二分析判断模块设置为:
将所述位置和速度输入对应的日常活动轨迹分布模型的二维坐标系中 , 在所述二维坐标系中获得一个坐标点; 计算所述坐标点与所述二维 坐标系的中心点的距离, 判断所述距离是否大于第三阈值; 当所述距 离大于所述第三阈值吋, 判定所述监测对象的活动状态异常; 其中, 所述二维坐标系的中心点为: 根据采集的环境组合下所有的日常活动 信息在二维坐标系中的分布而计算出的中心点。
根据权利要求 28所述的监测装置, 其中, 所述第二分析判断模块还设 置为: 当所述距离不大于所述第三阈值吋, 判断所述距离是否小于第
四阈值; 当所述距离小于所述第四阈值吋, 利用所述坐标点重新确定 所述中心点; 其中, 所述第三阈值大于所述第四阈值。
[权利要求 30] 根据权利要求 16所述的监测装置, 其中, 还包括模型更新模块, 所述 模型更新模块设置为:
当当前处于临吋状态吋, 暂停利用所述生理参数信息更新所述身体健 康指数模型。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2017/076659 WO2018165868A1 (zh) | 2017-03-14 | 2017-03-14 | 监测方法和监测装置 |
CN201780002430.9A CN108401415B (zh) | 2017-03-14 | 2017-03-14 | 监测方法和监测装置 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2017/076659 WO2018165868A1 (zh) | 2017-03-14 | 2017-03-14 | 监测方法和监测装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2018165868A1 true WO2018165868A1 (zh) | 2018-09-20 |
Family
ID=63095107
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2017/076659 WO2018165868A1 (zh) | 2017-03-14 | 2017-03-14 | 监测方法和监测装置 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN108401415B (zh) |
WO (1) | WO2018165868A1 (zh) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3660802A1 (en) * | 2018-11-27 | 2020-06-03 | Koninklijke Philips N.V. | Predicting critical alarms |
CN110009869A (zh) * | 2019-04-01 | 2019-07-12 | 珠海格力电器股份有限公司 | 活动信息的监控方法、装置和系统 |
CN110415823B (zh) * | 2019-07-30 | 2020-08-04 | 杭州思锐信息技术股份有限公司 | 一种基于机器学习的安全状态数据处理方法及系统 |
CN110737330A (zh) * | 2019-09-06 | 2020-01-31 | 北京小米移动软件有限公司 | 提醒出行异常的方法及装置、存储介质 |
CN111143146B (zh) * | 2019-12-26 | 2023-05-23 | 深圳大普微电子科技有限公司 | 一种存储装置的健康状态预测方法及系统 |
CN112446803A (zh) * | 2020-11-30 | 2021-03-05 | 五八到家有限公司 | 监测系统及方法、设备、介质 |
CN112700874A (zh) * | 2021-01-26 | 2021-04-23 | 青岛咕噜邦科技有限公司 | 一种智慧社区模块化管理系统及方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080294019A1 (en) * | 2007-05-24 | 2008-11-27 | Bao Tran | Wireless stroke monitoring |
CN202282004U (zh) * | 2011-06-02 | 2012-06-20 | 上海巨浪信息科技有限公司 | 基于情景感知与活动分析的移动健康管理系统 |
CN104958076A (zh) * | 2015-07-02 | 2015-10-07 | 李云峰 | 基于天气环境信息的人身健康智能监测预警方法及装置 |
CN105960198A (zh) * | 2013-12-05 | 2016-09-21 | 慕曼德保健公司 | 病患的智能医疗监测 |
CN106339576A (zh) * | 2016-07-20 | 2017-01-18 | 美的集团股份有限公司 | 一种健康管理方法和系统 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN100340207C (zh) * | 2006-01-20 | 2007-10-03 | 华南理工大学 | 一种智能化监护系统报警方法 |
WO2011087927A1 (en) * | 2010-01-14 | 2011-07-21 | Venture Gain LLC | Multivariate residual-based health index for human health monitoring |
CN102406507B (zh) * | 2010-09-20 | 2013-10-23 | 天津职业技术师范大学 | 基于人体生理信号的汽车驾驶员疲劳监测方法 |
CN105447304B (zh) * | 2015-11-10 | 2018-09-07 | 重庆邮电大学 | 一种基于自学习算法的预警系统及移动终端 |
-
2017
- 2017-03-14 WO PCT/CN2017/076659 patent/WO2018165868A1/zh active Application Filing
- 2017-03-14 CN CN201780002430.9A patent/CN108401415B/zh not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080294019A1 (en) * | 2007-05-24 | 2008-11-27 | Bao Tran | Wireless stroke monitoring |
CN202282004U (zh) * | 2011-06-02 | 2012-06-20 | 上海巨浪信息科技有限公司 | 基于情景感知与活动分析的移动健康管理系统 |
CN105960198A (zh) * | 2013-12-05 | 2016-09-21 | 慕曼德保健公司 | 病患的智能医疗监测 |
CN104958076A (zh) * | 2015-07-02 | 2015-10-07 | 李云峰 | 基于天气环境信息的人身健康智能监测预警方法及装置 |
CN106339576A (zh) * | 2016-07-20 | 2017-01-18 | 美的集团股份有限公司 | 一种健康管理方法和系统 |
Also Published As
Publication number | Publication date |
---|---|
CN108401415B (zh) | 2020-11-17 |
CN108401415A (zh) | 2018-08-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2018165868A1 (zh) | 监测方法和监测装置 | |
US10485452B2 (en) | Fall detection systems and methods | |
US10368748B2 (en) | System and method of evaluating an association between a wireless sensor and a monitored patient | |
CN203122372U (zh) | 基于Android技术和物联网的实时健康监测及智能预警系统 | |
JP7550410B2 (ja) | 体調評価方法、熱中症発症リスク評価方法、および、プログラム | |
US20170231528A1 (en) | Method and system for continuous monitoring of a medical condition in patients | |
CN104077887A (zh) | 人体跌倒健康监测方法及装置 | |
CN103690285A (zh) | 一种智能护理装置 | |
WO2014182720A1 (en) | Mobile personal emergency response system | |
CN104042196A (zh) | 实时健康监测及智能预警系统与方法 | |
JPWO2003096892A1 (ja) | 脈拍異常監視装置および脈拍異常警報システム | |
CN205003738U (zh) | 自助式社区健康管理系统 | |
WO2018151004A1 (ja) | 異変通知システムおよび異変通知方法 | |
EP2541474A1 (en) | Monitoring a user activity using a mobile device | |
CN204798006U (zh) | 一种看护系统 | |
CN103646503A (zh) | 一种看护系统 | |
Srinivasan et al. | Elder care system using IoT and machine learning in AWS cloud | |
US20170249823A1 (en) | System for Tracking Wellness and Scheduling of Caregiving | |
CN109247926A (zh) | 一种特定区域内人体健康监测及跟踪系统 | |
KR20160015431A (ko) | 사용자 안전도 및 위치 기반의 응급환자 관리 방법 및 시스템 | |
CN106230991A (zh) | 一种患者监控方法及装置 | |
CN205621071U (zh) | 一种养老院养老安全保障系统 | |
CN112767202A (zh) | 一种远程监测平台及监测方法 | |
CN117095509A (zh) | 一种老年健康数据分析系统 | |
Ravi et al. | Real time driver fatigue detection and smart rescue system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17900374 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 17900374 Country of ref document: EP Kind code of ref document: A1 |