CN107169307A - Health risk assessment method and apparatus - Google Patents

Abstract

The present invention provides a health risk assessment method and device, the method at least includes a data collection step and a risk analysis step; the data collection step collects at least one user's physiological sign data, and saves the data; the risk analysis step, according to the The data evaluates the data level, and the data level combines the impact weight of the data to calculate the risk coefficient of the user's disease. The health risk assessment method of the present invention can combine the physiological sign data collected by the user terminal to output reliable disease risk assessment results, so that the device applying the health risk assessment method can provide users with accurate, reliable and portable medical and health services.

Description

Health risk assessment method and device

technical field

The invention relates to computer application technology, in particular to a health risk assessment method and device.

Background technique

Medical research shows that changes in indices and indexes such as heart rate, blood pressure, blood oxygen level, and sleep quality have important reference value for assessing body-related diseases, fatigue, and health assessment.

However, at present, smart wearable devices can only implement some simple sports management functions. The devices cannot store data or have limited storage capacity, and cannot be used for health analysis at the medical level. , How to combine smart terminals to enjoy reliable medical and health services in a portable way has not yet proposed an effective solution.

Contents of the invention

The invention provides a health risk assessment method and device, which can provide users with portable and reliable medical health risk assessment services in combination with data collected by an intelligent terminal.

The present invention provides a health risk assessment method, comprising: at least including a data collection step and a risk analysis step; the data collection step collects at least one user's physiological sign data and saves the data; the risk analysis step evaluates the data level according to the data , the data level is combined with the impact weight of the data to calculate the risk coefficient of the user's disease.

The present invention also provides a health risk assessment device, which at least includes a data collection module, a storage module, and a risk analysis module; a data collection module that collects at least one type of user physiological sign data and stores the data in the storage module; a risk analysis module , evaluating the data level according to the data stored in the storage module, and calculating the risk coefficient of the user's disease by combining the data level with the impact weight of the data.

The health risk assessment method provided by the present invention can output reliable health risk assessment results in combination with the physiological sign data collected by the user terminal, so that the device applying the health risk assessment method can provide users with accurate, reliable and portable medical and health care Serve. Through the output of the evaluation method, the automatic early warning risk medical treatment can replace the existing physical discomfort and then seek medical treatment, which can allow patients to receive treatment at the early stage of the disease, reduce the cost of treatment for patients, and improve the experience and timeliness of existing medical services. At the same time, the smart terminal can save the data of physiological signs when the body is abnormal for the doctor to diagnose, so that the doctor can understand the user's physical state more accurately to improve the accuracy of diagnosis.

Description of drawings

Fig. 1 is a flowchart of the health risk assessment method of the present invention.

Fig. 2 is the instant risk assessment method of S102 in Fig. 1 of the present invention;

Fig. 3 is the S102 risk prediction and evaluation method in Fig. 1 of the present invention;

Fig. 4 is the forecasting model building method of forecasting index in Fig. 3;

Fig. 5 is a schematic structural diagram of the health analysis and evaluation device of the present invention.

detailed description

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

As shown in Fig. 1, the present invention discloses a health risk assessment method, which at least includes a data collection step and a risk analysis step.

Data collection step (S101), collect at least one kind of user physiological sign data, and save the data;

Risk analysis step (S102), evaluating the data level according to the data, and calculating the risk coefficient of the user's disease by combining the data level with the impact weight of the data.

According to the risk index, the user's risk of disease is judged. The higher the index, the greater the risk of disease.

The health risk assessment method of the present invention can collect the user's physiological sign data in real time and store them in the terminal. After long-term data accumulation, the risk analysis module analyzes the stored data, and can accurately predict some chronic diseases, such as cardiovascular diseases, and the complications of chronic diseases caused by related diseases, such as The risk of cardiovascular complications is given, and risk reminders are given, and users are advised to make preventive preparations for sudden diseases.

Step S102 in Fig. 1, as shown in Fig. 2, may include the following steps to assess the immediate risk coefficient of cardiovascular disease patients:

Step 201: Compare the three index data of the user's electrocardiogram signal ECG, blood pressure BP and blood oxygen SpO collected by the sensor module with the user's normal index data, evaluate the real-time level of the collected index data, record H1 _c , S1 _c , B1 _c is the real-time evaluation grade of the three indicators.

The user's normal indicator data refers to the same kind of indicator data that is judged to be the user's physical health by the risk analysis module. This application does not limit the classification of H1 _c , S1 _c , and B1 _c grades, which are set according to actual needs.

Step 202: Calculate the immediate risk assessment coefficient Risk1=A·{H1 _c , S1 _c , B1 _c }, where the influence weight vector A={a ₁ , a ₂ , a ₃ }. The influence weight vector can be set with reference to statistical results, or set by a professional doctor according to the user's physical condition.

Real-time risk assessment can realize real-time monitoring of the user's physiological signs data, real-time monitoring of the user's physical abnormalities, and remind the user to seek medical treatment in time. At the same time, the smart terminal can save the data of physiological signs when the body is abnormal for the doctor to diagnose, so that the doctor can understand the user's physical state more accurately to improve the accuracy of diagnosis.

Step S102 in Fig. 1, as shown in Fig. 3, may also include the following steps, to evaluate the incidence prediction risk coefficient of cardiovascular disease patients:

Step 301: Based on the three index data of the user's electrocardiogram signal ECG, blood pressure BP and blood oxygen SpO stored in the storage module, predict the values of the user's three indexes within a predetermined period.

The predetermined period includes short-term and long-term, which can be set by users according to their own needs.

Step 302: By comparing the predicted values of the three indicators with the data of the normal indicators, evaluate the grades of the predicted indicators, and record H2 _c , S2 _c , and B2 _c as the predicted evaluation grades of the three indicators.

Step 303: Calculate the risk assessment prediction coefficient Risk2=A·{H2 _c , S2 _c , B2 _c }, where the influence weight vector A={a ₁ , a ₂ , a ₃ }.

The smart terminal of this application can not only realize real-time risk assessment, but also predict the risk of some potential chronic diseases through the long-term trend collected by the sensor module, warn the user in advance, take corresponding preventive measures, or seek medical treatment in advance.

Further, the value of the predictive index in step S301 in Figure 3 may use an existing predictive model, or as shown in Figure 4, including:

Step 401: Construct a model, assuming that the value L _i of the sample index collected for the i-th time of the sample L is expressed as a linear combination of values collected for the past p (p≤i) times, plus white noise for the i-th time, namely:

L _i ＝θ ₀ +φ ₁ L _i-1 +…+φ _p L _ip +a _i , where {a _i ,i=0,±1,±2,…} is white noise, φ ₁ , φ ₂ , ..., φ _p is a constant coefficient, and θ ₀ is a constant.

Step 402: Calculate constant coefficients φ ₁ , φ ₂ , ..., φ _p and θ ₀ based on the sample data;

Step 403: Predicting the index value, knowing the k-th user sample index data, predicting the k+l-th index value, setting i=k+l, and obtaining the k+l-th index value is:

L _k+l ＝θ ₀ +φ ₁ L _k+l-1 +φ ₂ L _k+l-2 + . . . +φ _p L _k+lp +a _k+l .

In the method of the present invention, when data is stored, it is stored in a preset data packet format;

The preset data packet format is encoded according to the format of packet header plus data. The packet header includes start bit, device number, time stamp, physiological signs and status data name and check digit; the start bit is used to distinguish each data packet; the device number is used to The time stamp is used to record the collection time; the physiological sign data name is used to distinguish the physiological sign data collected by different sensors and the different status information of the user, and the check digit is used to verify whether there is an error in data transmission.

In the method of the present invention, the risk analysis step also includes sleep state detection, and the sleep state detection at least includes the following steps:

Step 501: analyzing the user's posture data, extracting the user's sleep time and deep sleep time;

Step 502: According to the user's sleep time and deep sleep time, as well as the user's heart rate HR and blood oxygen SpO data within two periods, determine the user's sleep state.

Wherein, step 501 may further include the following steps:

Step 501-1, automatically enable sleep state detection through posture detection;

Step 501-2, start timing the sleep of the user, and collect acceleration sensor data and heart rate, heart rate and blood oxygen concentration data collected by the blood oxygen sensor;

Step 501-3: Analyze the data collected by the acceleration sensor in real time to determine the user's posture. When the acceleration collection data is stable for a long time, start timing the deep sleep of the user, and stop the deep sleep timing once the user has a posture change;

Step 501-4, during the sleep process and the deep sleep process, collect the user's heart rate and blood oxygen data through the heart rate and blood oxygen sensors at regular intervals;

Step 501-5. After the user wakes up, the sleep state monitoring function is automatically terminated through posture detection. Through the analysis of posture and heart rate data, the amount of sleep time and deep sleep time of the night, as well as the blood oxygen concentration and heart rate during sleep are extracted.

In the method of the present invention, the risk analysis step also includes fatigue state detection, and the fatigue state detection at least includes the following steps:

Step 601: Detect the fluctuation characteristics of the electrocardiographic signal ECG and the heart rate variation signal HRV stored in the storage module, and convert the fluctuation characteristics into the initial value of the fatigue coefficient;

Step 602: Calculate the final fatigue coefficient by weighting the initial value of the fatigue coefficient with the user's heart rate HR, blood oxygen SpO data, and user's sleep state.

The health risk assessment method provided by the present invention can output reliable health risk assessment results in combination with the physiological sign data collected by the user terminal, so that the intelligent terminal applying the health risk assessment method can provide users with accurate, reliable and portable medical treatment. health service. Through the output of the evaluation method, the automatic early warning risk medical treatment can replace the existing physical discomfort and then seek medical treatment, which can allow patients to receive treatment at the early stage of the disease, reduce the cost of treatment for patients, and improve the experience and timeliness of existing medical services. At the same time, the smart terminal can save the data of physiological signs when the body is abnormal for the doctor to diagnose, so that the doctor can understand the user's physical state more accurately to improve the accuracy of diagnosis.

As shown in Fig. 5, the present invention also discloses a health risk assessment device, which at least includes a data collection module, a storage module and a risk analysis module.

The data collection module collects at least one kind of user's physiological sign data, and stores the data in the storage module. The risk analysis module evaluates the data level according to the data stored in the storage module, and the data level combines the impact weight of the data to calculate the risk coefficient of the user's disease.

The health risk assessment device of the present invention may be a smart watch, or a wristband, or other wearable devices, or part of the devices in the user terminal.

For example, the device of the present invention monitors the application description of the user suffering from primary heart disease in real time, including:

Step A-1, the sensor module of the device collects the ECG signal ECG data of the ECG sensor in real time, the heart rate HR, blood pressure BP and blood oxygen SpO concentration data collected by the blood oxygen sensor;

Step A-2, the risk analysis module analyzes the heart rate, blood pressure and blood oxygen data collected in real time; once the heart rate HR data is abnormal, the blood oxygen SpO concentration drops sharply, and the change of blood pressure HR conforms to the characteristics of primary heart disease, evaluate the user The risk factor of the disease, if it is determined that the user has a primary heart attack and may have an emergency such as cardiac arrest, the evaluation result will be displayed on the display screen of the device;

Step A-3, or, further, the device requests medical assistance from relevant institutions and related persons through the communication module, and sends the collected user's ECG data to relevant institutions to report the user's real-time status.

Also, as an example, the application description of the real-time monitoring of serious accidental falls of elderly users by the device of the present invention:

Step B-1, the sensor module of the device collects the user attitude data of the acceleration sensor in real time;

Step B-2. When the risk analysis module detects a drastic change in the user's posture, and after the drastic posture change, the user's posture changes very little or does not change, indicating that the user may fall;

Step B-3. The risk analysis module analyzes the changes in the user’s heart rate HR, blood pressure, and BP sign data collected by the sensor module in real time at the same time. If the data changes are abnormal, it means that the user has accidentally fallen, and the evaluation result is displayed on the device display screen;

Step B-5, or, further, the device requests medical assistance from relevant institutions and related persons through the communication module, and sends the collected physiological sign data to relevant institutions to report the user's real-time status.

The above shows that the device of the present invention can monitor users suffering from chronic sudden diseases in real time, such as high blood pressure, heart disease, etc.; or elderly users, sudden accidents such as falling down. The device monitors the user's posture and physiological sign data in real time. Once the user has a sudden illness or accident, the device will send the user's situation to the relevant organization or related person through the emergency communication function to request corresponding medical measures and services. .

The risk analysis module uses a DSP chip as a microcontroller to analyze and collect data; the storage module uses SPI NandFlash as a storage device. This type of chip has the characteristics of small size, few pins and large capacity, and can store collected data for a long time. The large-capacity storage design of the storage module can collect and store user data for a long time, making the smart terminal more accurate in predicting chronic diseases such as cardiovascular diseases.

In this application, the user's physiological sign data includes posture, heart rate HR, blood oxygen SpO, blood pressure BP, electrocardiographic signal ECG, and heart rate variability signal HRV, etc.

There are 5 types of sensors in the sensor module to detect the user's posture and collect physiological sign data, among which the acceleration sensor is used to detect the user's posture and related motion information; the blood oxygen and heart rate sensors are used to collect the user's blood oxygen SpO and heart rate HR data, and simultaneously solve Calculate the blood pressure BP data; the infrared body temperature sensor is used to collect the user's body temperature data; the electrocardiogram sensor is used to collect the user's ECG signal ECG data, and analyze each acquisition by extracting the QRS band in the ECG signal ECG collected by the electrocardiogram sensor The interval standard deviation between the R wave peaks of each cycle of the signal of the QRS complex wave band is used to reflect the heart rate variability signal HRV.

The smart terminal collects the user's physiological sign data in real time, packs them into data packets in a certain format, and stores them in the storage module. The data packet format is encoded according to the format of packet header plus data, where the packet header includes start bit, device number, time stamp, physiological signs and status data name and check digit. The start bit is used to distinguish each data packet; the device number is used to distinguish different devices that display the user; the timestamp is used to record the collection time; the physiological sign data name is mainly used to distinguish the physiological sign data collected by different sensors and the different states of the user Information, such as distinguishing whether the collected data is ECG data or heart rate data; the check digit is used to verify whether the data transmission is wrong.

Furthermore, before the data is packaged or stored, it is preprocessed to eliminate abnormal data.

In the device of the present invention, the risk analysis module is also an instant risk analysis module, which includes:

Instant grade evaluation module: compare the three index data of the user's electrocardiogram signal ECG, blood pressure BP and blood oxygen SpO collected by the sensor module with the user's normal index data, evaluate the instant grade of the index value, record H1 _c , S1 _c , B1 _c is the immediate evaluation grade of the three indicators;

Immediate risk calculation module: Calculate the immediate risk assessment coefficient Risk1=A·{H1 _c , S1 _c , B1 _c }, where the influence weight vector A={a ₁ , a ₂ , a ₃ }.

In the device of the present invention, the risk analysis module also includes a risk prediction module, and the risk prediction module at least includes the following modules:

Prediction module: Based on the three index data of the user's electrocardiogram signal ECG, blood pressure BP, and blood oxygen SpO stored in the storage module, predict the values of the user's three indexes within a predetermined period;

Forecast grade evaluation module: By comparing the predicted value of the three indicators with the normal index data, evaluate the grade of the predicted index value, record H2 _c , S2 _c , B2 _c as the forecast evaluation grade of the three indicators;

Risk prediction calculation module: calculate the risk assessment prediction coefficient Risk2=A·{H2 _c , S2 _c , B2 _c }, where the influence weight vector A={a ₁ , a ₂ , a ₃ }.

Further, the predictive index values include:

Model building module: to build a model, the value L _i of the sample index collected for the i-th time of the sample L is expressed as a linear combination of the values collected for the past p (p≤i) times, plus white noise for the i-th time, namely:

L _i ＝θ ₀ +φ ₁ L _i-1 +…+φ _p L _ip +a _i , where {a _i ,i=0,±1,±2,…} is white noise, φ ₁ , φ ₂ , ..., φ _p is a constant coefficient, and θ ₀ is a constant;

Model coefficient calculation module: based on sample data, calculate constant coefficients φ ₁ , φ ₂ , ..., φ _p and θ ₀ ;

Model prediction module: predicting index value, knowing the k-th user sample index data, predicting the index value of the k+lth time, setting i=k+l, and obtaining the k+l-th index value as:

L _k+l ＝θ ₀ +φ ₁ L _k+l-1 +φ ₂ L _k+l-2 + . . . +φ _p L _k+lp +a _k+l .

In another embodiment of the present invention, the risk analysis module also includes a sleep state detection module, specifically including:

Sleep time detection module: analyze the user's posture data, extract the user's sleep time and deep sleep time;

Sleep state analysis module: judge the user's sleep state according to the user's sleep time and deep sleep time, as well as the user's heart rate HR and blood oxygen SpO data within two periods of time.

In another embodiment of the present invention, the risk analysis module also includes a fatigue state detection module, specifically including:

Fatigue coefficient initial value calculation module: the device collects the user's electrocardiogram signal (ECG), heart rate variability signal (HRV), heart rate and blood oxygen data through sensors. The risk analysis module detects the fluctuation characteristics of the electrocardiographic signal ECG and the heart rate variation signal HRV stored in the storage module, and converts the fluctuation characteristics into the initial value of the fatigue coefficient.

Fatigue coefficient comprehensive calculation module: The risk analysis module weights the initial value of the fatigue coefficient with heart rate HR, blood oxygen SpO data, and user sleep status to obtain the final fatigue coefficient.

In addition to Figure 5, the device of the present invention can also be combined with a platform or server to obtain more functions or services. At this time, the terminal is used as a collection terminal of a medical and health service platform to collect physiological sign data for the platform to analyze and evaluate the user's health status; And used for medical diagnosis services to users. The terminal can be connected to the smart phone through Bluetooth, and the collected data can be sent to the smart phone and other devices, and then the data can be sent to the platform server through the Internet, or the mobile phone can directly send the collected data to the platform server.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the technical solutions of the present invention are Should be included within the protection scope of the present invention.

Claims (16)

1. A health risk assessment method, characterized in that at least comprising a data collection step and a risk analysis step; A data collection step, collecting at least one type of user physiological sign data, and saving the data; In the risk analysis step, the data grade is evaluated according to the data, and the data grade is combined with the impact weight of the data to calculate the risk coefficient of the user's disease. 2. The method according to claim 1, wherein the user's physiological sign data includes posture, heart rate HR, blood oxygen SpO, blood pressure BP, electrocardiographic signal ECG and heart rate variability signal HRV. 3. The method according to claim 2, wherein the risk analysis step comprises: Step 201: Compare the three index data of the user's electrocardiogram signal ECG, blood pressure BP, and blood oxygen SpO collected by the sensor module with the user's normal index data, and evaluate the instant grades H1 _c , S1 _c , B1 _c of the index values ; Step 202: Calculate the immediate risk assessment coefficient Risk1=A·{H1 _c , S1 _c , B1 _c }, where the influence weight vector A={a ₁ , a ₂ , a ₃ }. 4. method according to claim 2, is characterized in that, described risk analysis step also comprises: Step 301: Based on the three index data of the user's electrocardiogram signal ECG, blood pressure BP and blood oxygen SpO stored in the storage module, predict the values of the three indexes of the user within a predetermined period; Step 302: By comparing the predicted value of the three indicators with the normal indicator data, evaluate the level of the predicted indicator value, record _H2c , _S2c , _B2c as the predicted evaluation level of the three indicators; Step 303: Calculate the risk assessment prediction coefficient Risk2=A·{H2 _c , S2 _c , B2 _c }, where the influence weight vector A={a ₁ , a ₂ , a ₃ }. 5. The method according to claim 4, wherein the predicted values of the three indicators of the user within a predetermined period include: Step 401: Build a model. The value L _i of the sample index collected for the i-th time of the sample L is expressed as a linear combination of the values collected for the past p (p≤i) times, plus white noise for the i-th time, namely: L _i ＝θ ₀ +φ ₁ L _i-1 +…+φ _p L _ip +a _i , where {a _i ,i=0,±1,±2,…} is white noise, φ ₁ , φ ₂ , ..., φ _p is a constant coefficient, and θ ₀ is a constant; Step 402: Calculate constant coefficients φ ₁ , φ ₂ , ..., φ _p and θ ₀ based on the sample data; Step 403: Predicting the index value, knowing the k-th user sample index data, predicting the k+l-th index value, setting i=k+l, and obtaining the k+l-th index value is: L _k+l ＝θ ₀ +φ ₁ L _k+l-1 +φ ₂ L _k+l-2 + . . . +φ _p L _k+lp +a _k+l . 6. The method according to claim 1, wherein the data is stored in a preset data packet format; The preset data packet format is encoded according to the format of header plus data, and the header includes start bit, device number, time stamp, physiological signs and status data name and check digit; the start bit is used to distinguish each segment data packet; the device number is used to distinguish and display different devices of the user; the timestamp is used to record the collection time; the physiological sign data name is used to distinguish the physiological sign data collected by different sensors and the different status information of the user, so The above parity bit is used to check whether there is an error in data transmission. 7. The method according to claim 2, wherein the risk analysis step also includes sleep state detection, and the sleep state detection includes at least the following steps: Step 501: analyzing the user's posture data, extracting the user's sleep time and deep sleep time; Step 502: According to the user's sleep time and deep sleep time, as well as the user's heart rate HR and blood oxygen SpO data within the two time periods, determine the user's sleep state. 8. The method according to claim 7, wherein the risk analysis step also includes fatigue state detection, and the fatigue state detection includes at least the following steps: Step 601: Detect the fluctuation characteristics of the electrocardiogram signal ECG and the heart rate variation signal HRV stored in the storage module, and convert the fluctuation characteristics into the initial value of the fatigue coefficient; Step 602: Calculate the final fatigue coefficient by weighting the initial value of the fatigue coefficient with the user's heart rate HR, blood oxygen SpO data and the user's sleep state. 9. A health risk assessment device, characterized in that it at least includes a data collection module, a storage module and a risk analysis module; A data collection module, collecting at least one type of user physiological sign data, and storing the data in the storage module; The risk analysis module evaluates the data level according to the data stored in the storage module, and the data level combines the impact weight of the data to calculate the risk coefficient of the user's disease. 10. The device according to claim 9, wherein the user's physiological sign data includes posture, heart rate HR, blood oxygen SpO, blood pressure BP, electrocardiogram signal ECG and heart rate variability signal HRV. 11. The device according to claim 10, wherein the risk analysis module comprises a real-time risk analysis module, and the real-time risk analysis module comprises at least: Immediate grade evaluation module: compare the three index data of the user's electrocardiogram signal ECG, blood pressure BP, and blood oxygen SpO collected by the sensor module with the user's normal index data, and evaluate the instant grades of the index values H1 _c , S1 _c , _B1c ; Immediate risk calculation module: Calculate the immediate risk assessment coefficient Risk1=A·{H1 _c , S1 _c , B1 _c }, where the influence weight vector A={a ₁ , a ₂ , a ₃ }. 12. The device according to claim 10, wherein the risk analysis module further comprises a risk prediction module, and the risk prediction module comprises at least: Prediction module: based on the three index data of the user's electrocardiogram signal ECG, blood pressure BP and blood oxygen SpO stored in the storage module, predict the values of the three indexes of the user within a predetermined period; Prediction level evaluation module: by comparing the predicted value of the three indicators with the normal index data, evaluate the level of the predicted indicator value, and record _H2c , _S2c , and _B2c as the forecasted evaluation level of the three indicators; Risk prediction calculation module: calculate the risk assessment prediction coefficient Risk2=A·{H2 _c , S2 _c , B2 _c }, where the influence weight vector A={a ₁ , a ₂ , a ₃ }. 13. The device according to claim 12, wherein the predicted values of the three indicators of the user within a predetermined period include: Model building module: to build a model, the value L _i of the sample index collected for the i-th time of the sample L is expressed as a linear combination of the values collected for the past p (p≤i) times, plus white noise for the i-th time, namely: L _i ＝θ ₀ +φ ₁ L _i-1 +…+φ _p L _ip +a _i , where {a _i ,i=0,±1,±2,…} is white noise, φ ₁ , φ ₂ , ..., φ _p is a constant coefficient, and θ ₀ is a constant; Model coefficient calculation module: based on sample data, calculate constant coefficients φ ₁ , φ ₂ , ..., φ _p and θ ₀ ; Model prediction module: predicting index value, knowing the k-th user sample index data, predicting the index value of the k+lth time, setting i=k+l, and obtaining the k+l-th index value as: L _k+l ＝θ ₀ +φ ₁ L _k+l-1 +φ ₂ L _k+l-2 + . . . +φ _p L _k+lp +a _k+l . 14. The device according to claim 9, wherein the data is stored in a preset data packet format; The preset data packet format is encoded according to the format of header plus data, and the header includes start bit, device number, time stamp, physiological signs and status data name and check digit; the start bit is used to distinguish each segment data packet; the device number is used to distinguish and display different devices of the user; the timestamp is used to record the collection time; the physiological sign data name is used to distinguish the physiological sign data collected by different sensors and the different status information of the user, so The above parity bit is used to check whether there is an error in data transmission. 15. The device according to claim 10, wherein the risk analysis module further comprises a sleep state detection module, and the sleep state detection module comprises at least: Sleep time detection module: analyze the user's posture data, extract the user's sleep time and deep sleep time; Sleep state analysis module: judge the user's sleep state according to the user's sleep time and deep sleep time, as well as the user's heart rate HR and blood oxygen SpO data within the two periods. 16. The device according to claim 15, wherein the risk analysis module further comprises a fatigue state detection module, and the fatigue state detection module comprises at least: The fatigue coefficient initial value calculation module: detect the fluctuation characteristics of the electrocardiogram signal ECG and the heart rate variation signal HRV stored in the storage module, and convert the fluctuation characteristics into the initial value of the fatigue coefficient; Fatigue coefficient comprehensive calculation module: calculate the final fatigue coefficient by weighting the initial value of the fatigue coefficient with the user's heart rate HR, blood oxygen SpO data and the user's sleep state.