A health monitoring method and device based on smart watch

Through smart watches, we can detect the wearing action and physiological status data of epilepsy patients and build a prediction model, which solves the problem that the epilepsy is not effectively predicted and identified in the existing technology, and realizes the accurate prediction and identification of the onset status of epilepsy patients, and improves the improvement of the onset status of epilepsy patients. The reliability of early warnings.

CN118614886BActive Publication Date: 2024-11-15GUANGZHOU SKMEI WATCH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410690270.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-30
Publication Date
2024-11-15
Estimated Expiration
2044-05-30

AI Technical Summary

Technical Problem

Existing smart wearable devices cannot effectively predict and identify the onset status of epilepsy patients, and lack the accuracy and reliability monitoring function of epilepsy patients.

Method used

The smart watch detects the wearing action and physiological status data of epilepsy patients, generates detection standard data, combines the data of gyroscope, positioning module, blood oxygen, blood sugar and heart rate detection module to build a prediction model, monitor and warning the epilepsy status in real time, and improves the status of epilepsy Prediction accuracy and identification reliability.

Benefits of technology

Accurate prediction and identification of the onset status of epilepsy patients is achieved, the reliability of early warning is improved, and the risk of epilepsy is reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118614886B_ABST
    Figure CN118614886B_ABST
Patent Text Reader

Abstract

The present invention discloses a health monitoring method and device based on a smart watch, and the method includes: generating detection standard data in the smart watch based on the disease state data; detecting the wearing action data and physiological state data of the epileptic patient according to the detection module of the smart watch; analyzing the disease state of the epileptic patient in combination with the detection standard data and the wearing action data, marking the physiological state data corresponding to the disease state as pathological physiological data; marking the physiological data in the non-disease state as normal physiological data, extracting the physiological data in the previous preset time period of the corresponding pathological physiological data from the normal physiological data, constructing a prediction model, and obtaining the disease prediction data of the epileptic patient; generating early warning information based on the disease prediction data, and giving an early warning prompt to the epileptic patient based on the early warning information. The present invention improves the accuracy and reliability of predicting and identifying the disease state of the epileptic patient by detecting the real-time health state of the epileptic patient.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention mainly relates to the field of smart wearable technology, and specifically to a health monitoring method and device based on a smart watch. Background Art

[0002] Smart wearable devices are one of the main areas of information development at present. By setting up smart detection modules on devices such as watches and bracelets, the user's health status can be monitored through smart wearable devices.

[0003] Epilepsy is a chronic brain disease caused by brain damage. Epilepsy attacks are repetitive and transient, and are sudden. Current smart wearable devices can monitor the user's physiological state in real time, but cannot predict the onset of the user's physical illness based on the user's physiological state. Therefore, the health monitoring function of smart wearable devices needs to be improved. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art. The present invention provides a health monitoring method and device based on a smart watch, which can predict the onset status of epilepsy patients by detecting the real-time health status of epilepsy patients, and improve the accuracy and reliability of the prediction and identification of the onset status of epilepsy patients.

[0005] The present invention provides a health monitoring method based on a smart watch, the monitoring method comprising:

[0006] Inputting epilepsy patient's disease state data into the smart watch through an external application, and generating detection standard data in the smart watch based on the disease state data;

[0007] Detecting the wearing action data and physiological status data of the epileptic patient according to the detection module of the smart watch;

[0008] Combine the detection standard data with the wearing action data for analysis to determine whether the epileptic patient is in a disease state, and if so, mark the physiological state data corresponding to the disease state as pathological physiological data;

[0009] Marking the physiological data in a non-morbid state as normal physiological data, extracting the physiological data in a preset time period before the corresponding pathological physiological data from the normal physiological data, and marking them as transitional physiological data;

[0010] Construct a prediction model based on normal physiological data and transitional physiological data;

[0011] Inputting physiological state data into the prediction model according to unit time, and outputting the onset prediction data of epilepsy patients;

[0012] Early warning information is generated based on the onset prediction data, and early warning prompts are given to epilepsy patients based on the early warning information.

[0013] Furthermore, the step of inputting epilepsy patient's disease state data into the smart watch through an external application, and generating detection standard data in the smart watch based on the disease state data includes:

[0014] The communication module of the smart watch is connected with the external smart device signal, and the external smart device sends the historical onset status data of the epilepsy patient to the smart watch;

[0015] The smart watch extracts the motion parameter data of the epileptic patient based on the onset status data, and generates detection standard data based on the motion parameter data of the epileptic patient.

[0016] Furthermore, the detecting of the wearing action data and physiological state data of the epileptic patient by the detection module of the smart watch includes:

[0017] The gyroscope based on the smart watch detects the amplitude and frequency of hand movements of epileptic patients;

[0018] The location change data of epilepsy patients is detected based on the positioning module of the smart watch;

[0019] The blood oxygen detection module, blood sugar detection module and heart rate detection module based on the smart watch detect the physiological state data of epilepsy patients.

[0020] Further, the combined detection standard data and the wearing action data are analyzed to determine whether the epileptic patient is in a disease state, and if so, marking the physiological state data corresponding to the disease state as pathological physiological data includes:

[0021] Record the wearing action data of the epileptic patient at preset time intervals, and organize the wearing action data within the corresponding time interval into a measurement data group;

[0022] Arrange the detection standard data into a control data group, and insert one of the control data groups between any two adjacent measurement data groups;

[0023] The onset probability of the epilepsy patient is calculated in real time based on the measurement data set and the control data set. If the onset probability is greater than 85%, it is determined that the epilepsy patient is in an onset state.

[0024] Furthermore, the calculation formula for the probability of epilepsy is:

[0025]

[0026] Where W is the covariance matrix of action parameters within the same group, B is the covariance matrix of action parameters between groups of data, n is the number of action parameter data, k is the number of data groups, i is a constant, and X i is the action parameter matrix of the i-th data group, is the overall mean of the K data groups, is the mean data of the i-th data group, The probability of epilepsy.

[0027] Furthermore, the construction of a prediction model based on normal physiological data, transitional physiological data and pathological physiological data includes:

[0028] A first physiological data set of an epileptic patient is extracted per unit time from the transitional physiological data, and a second physiological data set of an epileptic patient is extracted per unit time from the normal physiological data. An onset time function within a time interval is calculated based on the first physiological data set and the second physiological data set.

[0029] Furthermore, the calculation formula of the onset time function is:

[0030]

[0031] Among them, T y is the onset time function, is the incidence coefficient of the preset time interval, is the incidence coefficient of the predicted time, T is the preset time interval, i is a constant used as a counting unit, C i is the physiological data corresponding to the i-th unit time in the first physiological data set, C′ i is the physiological data corresponding to the i-th unit time in the second physiological data set.

[0032] Furthermore, the step of inputting the physiological state data into the prediction model according to the unit time and outputting the onset prediction data of the epilepsy patient comprises:

[0033] Extracting real-time physiological data of an epileptic patient from the physiological state data per unit time, inputting the real-time physiological data of the epileptic patient into the prediction model, and replacing the physiological data in the first physiological data set with the real-time physiological data in sequence;

[0034] When the output incidence coefficient is equal to the incidence coefficient of the preset time interval When the error is less than 0.5, the predicted data of the replacement position corresponding to the real-time physiological data is recorded to obtain the disease incidence coefficient of the predicted time.

[0035] Furthermore, generating warning information based on the onset prediction data, and providing a warning prompt to the epilepsy patient based on the warning information includes:

[0036] Within the corresponding predicted onset time, the wearing action data of the epilepsy patient is acquired based on the detection of the smart watch, and the physiological state of the epilepsy patient is judged based on the wearing action data combined with the detection standard data.

[0037] The present invention also provides a health monitoring device based on a smart watch, the health monitoring device comprising:

[0038] Data acquisition module: used to input epilepsy patient's disease state data into the smart watch through an external application, and generate detection standard data in the smart watch based on the disease state data;

[0039] Data detection module: used to detect the wearing action data and physiological state data of the epileptic patient according to the detection module of the smart watch;

[0040] Disease onset judgment module: used to combine the detection standard data with the wearing action data analysis to judge whether the epileptic patient is in a disease state, and if so, mark the physiological state data corresponding to the disease state as pathological physiological data;

[0041] Data marking module: used to mark the physiological data in the non-pathological state as normal physiological data, extract the physiological data in the preset time period before the corresponding pathological physiological data from the normal physiological data, and mark it as transitional physiological data;

[0042] Model building module: used to build a prediction model based on normal physiological data and transitional physiological data;

[0043] Onset prediction module: inputs physiological state data into the prediction model according to unit time, and outputs onset prediction data of epileptic patients;

[0044] Early warning module: used to generate early warning information based on the onset prediction data, and to provide early warning prompts to epilepsy patients based on the early warning information.

[0045] The present invention provides a health monitoring method and device based on a smart watch, which monitors the wearing movements and physiological states of epilepsy patients in real time through the smart watch, predicts the onset state of epilepsy patients in combination with the morbid physiological data of the epilepsy patients in the onset state, improves the accuracy of the prediction of the onset state of epilepsy patients, and improves the accuracy and reliability of the recognition of the real-time wearing movements of epilepsy patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 is a flow chart of a health monitoring method based on a smart watch in an embodiment of the present invention;

[0048] Figure 2 is a flow chart of a method for analyzing the onset status of an epileptic patient according to an embodiment of the present invention;

[0049] Figure 3 Schematic diagram of a health monitoring device based on a smart watch in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] Embodiment 1:

[0052] Figure 1 A flow chart of a health monitoring method based on a smart watch in an embodiment of the present invention is shown. The monitoring method includes:

[0053] S11: Input the epilepsy patient's disease status data into the smart watch through an external application, and generate detection standard data in the smart watch based on the disease status data.

[0054] Specifically, the communication module based on the smart watch is connected with the external smart device signal, and the external smart device sends the historical onset status data of the epilepsy patient to the smart watch. The communication module based on the smart watch can realize the signal connection between the smart watch and the external smart device, so that an information transmission channel is established between the smart watch and the external smart device. The epilepsy patient can send his or her own historical diagnosis data to the smart watch through the external smart device, and the smart watch can confirm the epilepsy patient's onset status data based on the epilepsy patient's historical diagnosis data.

[0055] The smart watch extracts the motion parameter data of the epileptic patient's onset based on the onset status data, generates detection standard data based on the motion parameter data of the epileptic patient's onset, extracts the motion parameter data of the epileptic patient in the onset state based on the onset status data, such as rigidity, convulsion, shaking and other actions, and records the corresponding motion parameter data according to the onset status data of the epileptic patient, so as to determine whether the epileptic patient is in the onset state according to the motion parameter data.

[0056] Furthermore, based on the fact that the onset state of the same epilepsy patient at each attack is highly similar, that is, the onset state of the epilepsy patient at each attack is relatively stable, the smart watch can detect the movement state and physiological state of the epilepsy patient in real time to confirm whether the epilepsy patient is in an onset state.

[0057] Specifically, when it is confirmed that an epileptic patient is in a disease state, the wearing action data of the epileptic patient in the disease state is obtained through the smart watch, and the detection standard data is corrected and updated according to the wearing action data in the disease state, so as to improve the accuracy of the smart watch in identifying and judging the disease state of the epileptic patient.

[0058] S12: Detecting the wearing action data and physiological state data of the epileptic patient according to the detection module of the smart watch.

[0059] Specifically, the gyroscope of the smart watch is used to detect the amplitude and frequency of the hand movements of the epileptic patient. When the epileptic patient wears the smart watch, the movement trajectory and direction of the watch is detected based on the built-in gyroscope of the smart watch. Based on the movement trajectory and direction of the watch, the amplitude and frequency of the hand movements of the epileptic patient can be confirmed.

[0060] The location change data of the epilepsy patient is detected based on the positioning module of the smart watch. The location information of the epilepsy patient is detected in real time based on the positioning module inside the smart watch. Based on the location information, rescue services are provided to the epilepsy patient when the disease occurs.

[0061] The blood oxygen detection module, blood sugar detection module and heart rate detection module based on the smart watch detect the physiological state data of epilepsy patients.

[0062] The smart watch is provided with a blood oxygen detection module, and the blood oxygen detection module is provided at the position where the dial meets the hand of the epileptic patient. The blood oxygen detection module emits infrared light and red light to the hand of the epileptic patient, and the blood oxygen saturation of the epileptic patient is confirmed based on the infrared light and red light reflection of the hand of the epileptic patient.

[0063] The smart watch is provided with a blood glucose detection module, which is a non-invasive blood glucose detection module, that is, a spectral test is performed on the skin of an epileptic patient through near-infrared and mid-infrared light. The near-infrared and mid-infrared light have good penetration performance on biological fluids and soft tissues. It can analyze the reflection of the soft tissue of the hand of the epileptic patient, extract the in vivo glucose information of the epileptic patient, and thus convert the blood glucose concentration of the epileptic patient.

[0064] The smart watch is provided with a heart rate detection module, which can emit green light at a high frequency to the hands of epileptic patients through LED lamp beads. Based on the contraction and expansion of blood vessels in the hands of epileptic patients, combined with the reflection strength of green light on the hands, and based on the good absorption of green light by human blood, the contraction and expansion of blood vessels can be detected in real time based on the strength of green light reflection on the hands, thereby converting it into a heart rate detection result.

[0065] Specifically, a photoelectric detection module is set on the bottom of the dial of the smart watch, and near-infrared light, mid-infrared light, infrared light, red light and green light are emitted to the hands of epileptic patients. According to the spectral reflection of different light by epileptic patients, the physiological data such as blood oxygen concentration, blood sugar concentration and heart rate of epileptic patients can be measured and calculated.

[0066] Furthermore, the smart watch can be used to monitor and analyze the physiological data of epilepsy patients in real time to improve the reliability of the smart watch in predicting and controlling the onset of epilepsy patients.

[0067] S13: combining the detection standard data with the wearing action data for analysis to determine whether the epileptic patient is in a disease state, and if so, marking the physiological state data corresponding to the disease state as pathological physiological data.

[0068] Specifically, Figure 2 A flow chart of a method for analyzing the onset status of an epileptic patient in an embodiment of the present invention is shown, and the analysis method comprises:

[0069] S131: Record the wearing action data of the epileptic patient at preset time intervals, and organize the wearing action data within the corresponding time interval into a measurement data group.

[0070] The preset time interval is set according to the historical diagnostic data of the epilepsy patient. In the present embodiment, the preset time interval is set to twice the duration of the epilepsy patient's attack. According to the historical diagnostic data of the epilepsy patient, the duration of each epilepsy attack is obtained, and the average duration of the epilepsy attack is calculated. According to the average duration, the preset time interval corresponding to the health monitoring of the epilepsy patient is determined.

[0071] The wearing action data of epilepsy patients are detected in real time based on smart watches, and the wearing action data in each preset time interval are organized into measurement data groups.

[0072] S132: Arrange the detection standard data into reference data groups, and insert one reference data group between any two adjacent measurement data groups.

[0073] Specifically, the detection standard data is organized into a control data group, and the onset time corresponding to the control data group is set to be consistent with the preset time interval. The smart watch performs real-time detection of the wearing movements of the epileptic patient based on the detection module, and records them as a measurement data group according to the preset time interval, and copies the control data group and stores it in the next storage position of the measurement data group, so that the measurement data group and the control data group can be stored alternately in the data storage space inside the smart watch.

[0074] S133: Calculate the epilepsy patient's probability of onset in real time based on the measurement data group and the control data group. If the probability of onset is greater than 85%, determine that the epilepsy patient is in an onset state.

[0075] By analyzing the measurement data group and the control data group, the probability of onset of epilepsy patients is calculated, and the calculation formula of the probability of onset of epilepsy patients is:

[0076]

[0077] Where W is the covariance matrix of action parameters within the same group, B is the covariance matrix of action parameters between groups of data, n is the number of action parameter data, k is the number of data groups, i is a constant, and X i is the action parameter matrix of the i-th data group, is the overall mean of the K data groups, is the mean data of the i-th data group, The probability of epilepsy.

[0078] Specifically, the plurality of data groups include a plurality of measurement data groups and a plurality of control data groups. By performing covariance matrix analysis on the plurality of measurement data groups and the plurality of control data groups, a plurality of action parameters of each data group are set in a matrix form, and by calculating the differences between the data groups, wherein, If it is close to 0, the differences between the data groups are large, that is, the difference between the measured data group and the control data group is large, and the probability that the epileptic patient is in a state of illness is low; If it is close to 1, the differences between the data groups are small, that is, the difference between the measured data group and the control data group is small, and the probability that the epileptic patient is in a state of illness is high.

[0079] Specifically, the internal storage unit of the smart watch is provided with a preset storage space, and the storage space includes a fixed storage cell set at the first position and several dynamic storage cells, the fixed storage cell is used to store a control data group, and the single dynamic storage cell is used to store a measurement data group and a control data group.

[0080] Furthermore, in this embodiment, the plurality of dynamic storage cells are set to 3, that is, the storage upper limit of the storage space is 3 measurement data groups and 4 control data groups, and the measurement data groups are obtained in real time and stored left-shifted in the plurality of dynamic storage cells of the storage space to ensure that the storage space can update the measurement data groups obtained by the smart watch in real time.

[0081] Evaluation and analysis are performed based on the measurement data groups acquired in real time combined with the measurement data groups at adjacent preset time intervals to improve the reliability of smart watches in judging the onset status of epileptic patients.

[0082] S14: Marking the physiological data in the non-morbid state as normal physiological data, extracting the physiological data in a preset time period before the corresponding pathological physiological data from the normal physiological data, and marking them as transitional physiological data.

[0083] The physiological data of epilepsy patients in non-disease state are recorded, and the physiological data in non-disease state are marked as normal physiological data. The physiological data within a preset time period before the corresponding pathological physiological data are extracted from the normal physiological data, and the physiological data are marked as transitional physiological data. According to the changes in the physiological data of epilepsy patients before the onset of the disease, the accuracy of epilepsy onset prediction and monitoring can be improved.

[0084] S15: Construct a prediction model based on normal physiological data and transitional physiological data.

[0085] The onset time function is constructed based on the normal physiological data and the transitional physiological data, and the onset duration function is constructed based on the transitional physiological data and the pathological physiological data.

[0086] Specifically, a first physiological data set of an epileptic patient is extracted per unit time from the transitional physiological data, and a second physiological data set of an epileptic patient is extracted per unit time from the normal physiological data. The onset time function within the time interval is calculated based on the first physiological data set and the second physiological data set. The calculation formula of the onset time function is:

[0087]

[0088] Among them, T y is the onset time function, is the incidence coefficient of the preset time interval, is the incidence coefficient of the predicted time, T is the preset time interval, i is a constant used as a counting unit, C i is the physiological data corresponding to the i-th unit time in the first physiological data set, C′ i is the physiological data corresponding to the i-th unit time in the second physiological data set.

[0089] S16: Inputting the physiological state data into the prediction model according to unit time, and outputting the onset prediction data of the epilepsy patient.

[0090] Extract the real-time physiological data of the epilepsy patient according to the unit time in the physiological state data, input the real-time physiological data of the epilepsy patient into the prediction model, replace the physiological data in the first physiological data set with the real-time physiological data in turn, and when the output disease incidence coefficient is equal to the disease incidence coefficient of the preset time interval When the error is less than 0.5, the predicted data of the replacement position corresponding to the real-time physiological data is recorded to obtain the disease incidence coefficient of the predicted time

[0091] Furthermore, when the output disease incidence coefficient is equal to the disease incidence coefficient of the preset time interval When the error is less than 0.5, the prediction of the onset time of the epileptic patient is completed, and the real-time physiological data replaces the i-th physiological data corresponding to the first physiological data set, thereby updating the first physiological data set of the prediction model.

[0092] Furthermore, the disease incidence coefficient of the preset time interval The initial value is retained to avoid error accumulation in the first physiological data set during multiple data updates, thereby avoiding affecting the accuracy of the onset time prediction function.

[0093] S17: Generate warning information based on the onset prediction data, and issue a warning prompt to the epilepsy patient based on the warning information.

[0094] According to the onset prediction data, early warning information for epilepsy patients is generated, and the early warning information includes the predicted time of the onset of epilepsy patients. Based on the early warning information, the epilepsy patients are given information alarm prompts through the sound module of the smart watch, thereby ensuring that the epilepsy patients can make adjustments in advance and reduce the risk of epilepsy.

[0095] Specifically, within the corresponding predicted onset time, the wearing action data of the epilepsy patient is acquired based on the detection of the smart watch, and the physiological state of the epilepsy patient is judged based on the wearing action data combined with the detection standard data; if the epilepsy patient is in an onset state, the smart watch can notify the relevant contacts and medical staff of the epilepsy patient based on the early warning information.

[0096] Furthermore, after confirming that the epileptic patient is in a disease state, the wearing action data of the epileptic patient is obtained and recorded based on the smart watch, and the detection standard data is updated based on the wearing action data in the disease state, thereby improving the accuracy of identifying the disease state of the epileptic patient.

[0097] The embodiment of the present invention provides a health monitoring method based on a smart watch, which monitors the wearing movements and physiological states of epilepsy patients in real time through the smart watch, predicts the onset state of epilepsy patients in combination with the morbid physiological data of the epilepsy patients in the onset state, improves the accuracy of the prediction of the onset state of epilepsy patients, and improves the accuracy and reliability of the recognition of the real-time wearing movements of epilepsy patients.

[0098] Embodiment 2:

[0099] Figure 3 A schematic diagram of a health monitoring device based on a smart watch in an embodiment of the present invention is shown, and the health monitoring device includes:

[0100] Data acquisition module 10: used to input epilepsy patient's disease status data into the smart watch through an external application, and generate detection standard data in the smart watch based on the disease status data.

[0101] Specifically, the communication module based on the smart watch is connected with the external smart device signal, and the external smart device sends the historical onset status data of the epilepsy patient to the smart watch. The communication module based on the smart watch can realize the signal connection between the smart watch and the external smart device, so that an information transmission channel is established between the smart watch and the external smart device. The epilepsy patient can send his or her own historical diagnosis data to the smart watch through the external smart device, and the smart watch can confirm the epilepsy patient's onset status data based on the epilepsy patient's historical diagnosis data.

[0102] Data detection module 20: used to detect the wearing action data and physiological state data of the epilepsy patient according to the detection module of the smart watch.

[0103] Specifically, the gyroscope of the smart watch is used to detect the amplitude and frequency of the hand movements of the epileptic patient. When the epileptic patient wears the smart watch, the movement trajectory and direction of the watch is detected based on the built-in gyroscope of the smart watch. Based on the movement trajectory and direction of the watch, the amplitude and frequency of the hand movements of the epileptic patient can be confirmed.

[0104] The location change data of the epilepsy patient is detected based on the positioning module of the smart watch. The location information of the epilepsy patient is detected in real time based on the positioning module inside the smart watch. Based on the location information, rescue services are provided to the epilepsy patient when the disease occurs.

[0105] The blood oxygen detection module, blood sugar detection module and heart rate detection module based on the smart watch detect the physiological state data of epilepsy patients.

[0106] The disease onset judgment module 30 is used to combine the detection standard data with the wearing action data analysis to determine whether the epileptic patient is in a disease state, and if so, mark the physiological state data corresponding to the disease state as pathological physiological data.

[0107] The wearing action data of the epileptic patient are recorded at preset time intervals, and the wearing action data within the corresponding time intervals are organized into a measurement data group.

[0108] The detection standard data are sorted into reference data groups, and one reference data group is inserted between any two adjacent measurement data groups.

[0109] The onset probability of the epilepsy patient is calculated in real time based on the measurement data set and the control data set. If the onset probability is greater than 85%, it is determined that the epilepsy patient is in an onset state.

[0110] The data marking module 40 is used to mark the physiological data in the non-morbid state as normal physiological data, extract the physiological data in the previous preset time period corresponding to the pathological physiological data from the normal physiological data, and mark it as transitional physiological data.

[0111] The physiological data of epilepsy patients in non-disease state are recorded, and the physiological data in non-disease state are marked as normal physiological data. The physiological data within a preset time period before the corresponding pathological physiological data are extracted from the normal physiological data, and the physiological data are marked as transitional physiological data. According to the changes in the physiological data of epilepsy patients before the onset of the disease, the accuracy of epilepsy onset prediction and monitoring can be improved.

[0112] Model building module 50: used to build a prediction model based on normal physiological data and transitional physiological data.

[0113] Specifically, a first physiological data set of an epileptic patient is extracted per unit time from the transitional physiological data, and a second physiological data set of an epileptic patient is extracted per unit time from the normal physiological data. The onset time function within the time interval is calculated based on the first physiological data set and the second physiological data set. The calculation formula of the onset time function is:

[0114]

[0115] Among them, T y is the onset time function, is the incidence coefficient of the preset time interval, is the incidence coefficient of the predicted time, T is the preset time interval, i is a constant used as a counting unit, C iis the physiological data corresponding to the i-th unit time in the first physiological data set, C′ i is the physiological data corresponding to the i-th unit time in the second physiological data set.

[0116] The onset prediction module 60 is used to input the physiological state data into the prediction model according to the unit time, and output the onset prediction data of the epileptic patient.

[0117] Extract the real-time physiological data of the epilepsy patient according to the unit time in the physiological state data, input the real-time physiological data of the epilepsy patient into the prediction model, replace the physiological data in the first physiological data set with the real-time physiological data in turn, and when the output disease incidence coefficient is equal to the disease incidence coefficient of the preset time interval When the error is less than 0.5, the predicted data of the replacement position corresponding to the real-time physiological data is recorded to obtain the disease incidence coefficient of the predicted time

[0118] The early warning module 70 is used to generate early warning information based on the onset prediction data, and to issue early warning prompts to epilepsy patients based on the early warning information.

[0119] According to the onset prediction data, early warning information for epilepsy patients is generated, and the early warning information includes the predicted time of the onset of epilepsy patients. Based on the early warning information, the epilepsy patients are given information alarm prompts through the sound module of the smart watch, thereby ensuring that the epilepsy patients can make adjustments in advance and reduce the risk of epilepsy.

[0120] An embodiment of the present invention provides a health monitoring device based on a smart watch, which monitors the wearing movements and physiological states of epilepsy patients in real time through the smart watch, predicts the onset state of epilepsy patients in combination with the morbid physiological data of the epilepsy patients in the onset state, improves the accuracy of the prediction of the onset state of epilepsy patients, and improves the accuracy and reliability of the recognition of the real-time wearing movements of epilepsy patients.

[0121] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0122] In addition, the embodiments of the present invention are described in detail above. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A health monitoring device based on a smart watch, characterized in that: The health monitoring device comprises: Data acquisition module: used to input epilepsy patient's disease state data into the smart watch through an external application, and generate detection standard data in the smart watch based on the disease state data; Data detection module: used to detect the wearing action data and physiological state data of the epileptic patient according to the detection module of the smart watch; Disease onset judgment module: used to combine the detection standard data with the wearing action data analysis to judge whether the epileptic patient is in a disease state, and if so, mark the physiological state data corresponding to the disease state as pathological physiological data; Data marking module: used to mark the physiological data in the non-pathological state as normal physiological data, extract the physiological data in the preset time period before the corresponding pathological physiological data from the normal physiological data, and mark it as transitional physiological data; Model building module: used to build a prediction model based on normal physiological data and transitional physiological data; Onset prediction module: inputs physiological state data into the prediction model according to unit time, and outputs onset prediction data of epileptic patients; Early warning module: used to generate early warning information based on the onset prediction data, and to provide early warning prompts to epilepsy patients based on the early warning information; The combined detection standard data and the wearing action data are analyzed to determine whether the epileptic patient is in a disease state, and if so, marking the physiological state data corresponding to the disease state as pathological physiological data includes: Record the wearing action data of the epileptic patient at preset time intervals, and organize the wearing action data within the corresponding time interval into a measurement data group; Arrange the detection standard data into a control data group, and insert one of the control data groups between any two adjacent measurement data groups; Calculating the probability of an epileptic patient's onset in real time based on the measurement data group and the control data group, and if the probability of an epileptic patient's onset is greater than 85%, determining that the epileptic patient is in an onset state; The calculation formula for the probability of epilepsy is: ; ; ; in, is the covariance matrix of action parameters within the same group, B is the covariance matrix of action parameters between groups of data, n is the number of action parameter data, k is the number of data groups, i is a constant, is the action parameter matrix of the i-th data group, is the overall mean of the K data groups, is the mean data of the i-th data group, The probability of epilepsy in patients with epilepsy; The construction of a prediction model based on normal physiological data, transitional physiological data and pathological physiological data includes: Extracting a first physiological data set of an epileptic patient by unit time from the transitional physiological data, extracting a second physiological data set of an epileptic patient by unit time from the normal physiological data, and calculating an onset time function within a time interval based on the first physiological data set and the second physiological data set; The calculation formula of the onset time function is: ; ; in, is the onset time function, is the incidence coefficient of the preset time interval, is the incidence coefficient of the predicted time, T is the preset time interval, i is a constant used as a counting unit, is the physiological data corresponding to the i-th unit time in the first physiological data set, is the physiological data corresponding to the i-th unit time in the second physiological data set.

2. The smart watch-based health monitoring device according to claim 1, characterized in that: The step of inputting epilepsy patient's disease state data into the smart watch through an external application and generating detection standard data in the smart watch based on the disease state data includes: The communication module of the smart watch is connected with the external smart device signal, and the external smart device sends the historical onset status data of the epilepsy patient to the smart watch; The smart watch extracts the motion parameter data of the epileptic patient based on the onset status data, and generates detection standard data based on the motion parameter data of the epileptic patient.

3. The smart watch-based health monitoring device according to claim 1, characterized in that: The detecting of the wearing action data and physiological state data of the epileptic patient by the detection module of the smart watch includes: The gyroscope based on the smart watch detects the amplitude and frequency of hand movements of epileptic patients; The location change data of epilepsy patients is detected based on the positioning module of the smart watch; The blood oxygen detection module, blood sugar detection module and heart rate detection module based on the smart watch detect the physiological state data of epilepsy patients.

4. The smart watch-based health monitoring device according to claim 1, characterized in that: The step of inputting physiological state data into the prediction model according to unit time and outputting epilepsy patient's onset prediction data comprises: Extracting real-time physiological data of an epileptic patient from the physiological state data per unit time, inputting the real-time physiological data of the epileptic patient into the prediction model, and replacing the physiological data in the first physiological data set with the real-time physiological data in sequence; When the output incidence coefficient is equal to the incidence coefficient of the preset time interval When the error is less than 0.5, the predicted data of the replacement position corresponding to the real-time physiological data is recorded to obtain the disease incidence coefficient of the predicted time.

5. The smart watch-based health monitoring device according to claim 1, characterized in that: Generating warning information based on the onset prediction data, and providing a warning prompt to the epilepsy patient based on the warning information includes: Within the corresponding predicted onset time, the wearing action data of the epilepsy patient is acquired based on the detection of the smart watch, and the physiological state of the epilepsy patient is judged based on the wearing action data combined with the detection standard data.