A cardiovascular health data early warning system and method based on health tags

Abstract

The present application provides a cardiovascular health data early warning system and method based on health labels, which relates to the field of cardiovascular prediction technology and is used to improve the accuracy and timeliness of cardiovascular health prediction. The method includes: determining the vehicle driving time period and user exercise time period corresponding to the user respectively according to the vehicle driving data and sensor data; determining the driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period; determining abnormal driving physiological state data and / or abnormal exercise physiological state data; converting the abnormal driving physiological state data and / or abnormal exercise physiological state data into a physiological state feature vector matrix; inputting the physiological state feature vector matrix into a cardiovascular prediction model to obtain the user's cardiovascular health prediction result, and the cardiovascular prediction model is trained according to the sample physiological state feature vector matrix and its cardiovascular health label.

Description

Technical Field

[0001] The present application relates to the field of cardiovascular prediction technology, and in particular to a cardiovascular health data early warning system and method based on health tags. Background Art

[0002] The latest data from the Global Burden of Disease Study show that cardiovascular disease (CVD) is the leading cause of death and reduced quality of life, and its prevalence is only increasing worldwide.

[0003] The early warning of cardiovascular monitoring in this stage is usually done by doctors monitoring patients on site and giving early warnings by observing real-time data. This method not only requires patients in the recovery stage to be hospitalized for observation, but also overly relies on the subjective experience of doctors, and cannot guarantee the accuracy of cardiovascular health monitoring and the timeliness of early warnings. In addition, the method in which doctors must observe patients on site is also less flexible. Summary of the invention

[0004] The embodiments of the present application provide a cardiovascular health data early warning system and method based on health tags, which are used to improve the accuracy and timeliness of cardiovascular health predictions.

[0005] The embodiment of the present invention provides a cardiovascular health data early warning system based on health tags, the system comprising: a vehicle-mounted device, a wearable device, and a cardiovascular early warning device, the vehicle-mounted device and the wearable device are in communication connection with the cardiovascular early warning device, the vehicle-mounted device is used to send the user's vehicle driving data to the cardiovascular early warning device; the wearable device is used to send the user's physiological state data and sensor data to the cardiovascular early warning device; the cardiovascular early warning device comprises a processor, the processor is used to execute:

[0006] Acquiring vehicle driving data, physiological status data, and sensor data of the user;

[0007] Determine the vehicle driving time period and the user exercise time period corresponding to the user respectively according to the vehicle driving data and the sensor data;

[0008] Determining driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period;

[0009] Analyzing the driving physiological state data and the sports physiological state data to determine abnormal driving physiological state data and / or abnormal sports physiological state data;

[0010] converting abnormal driving physiological state data and / or abnormal sports physiological state data into a physiological state feature vector matrix;

[0011] The physiological state feature vector matrix is ​​input into a cardiovascular prediction model to obtain a cardiovascular health prediction result of the user, wherein the cardiovascular prediction model is trained based on the sample physiological state feature vector matrix and its cardiovascular health label.

[0012] In an optional embodiment provided by the present invention, the determining the vehicle driving time period and the user exercise time period corresponding to the user respectively according to the vehicle driving data and the sensor data includes:

[0013] Determine the driving start time and driving end time of each driving of the user every day according to the vehicle driving data; and determine the exercise start time and exercise end time of each exercise of the user every day according to the sensor data;

[0014] The vehicle driving time period of the user each time each day is determined by the driving start time and driving end time of each driving each day; and the user exercise time period of the user each time each day is determined by the exercise start time and exercise end time of each exercise each day.

[0015] In an optional embodiment provided by the present invention, the determining the driving physiological state data and the exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period includes:

[0016] Determine a time period including each vehicle driving time period as a first time period; determine a time period including each user exercise time period as a second time period, wherein the start time of the first time period is earlier than the driving start time of the vehicle driving time period and the end time is later than the driving end time of the vehicle driving time period; the start time of the second time period is earlier than the exercise start time of the user exercise time period and the end time is later than the exercise end time of the user exercise time period;

[0017] The physiological state data corresponding to the first time period in the physiological state data is determined as driving physiological state data, and the physiological state data corresponding to the second time period in the physiological state data is determined as sports physiological state data.

[0018] In an optional embodiment provided by the present invention, the analyzing the driving physiological state data and the exercise physiological state data to determine abnormal driving physiological state data and / or abnormal exercise physiological state data includes:

[0019] Determine physiological state data before driving, physiological state data during driving, and physiological state data after driving from the driving physiological state data;

[0020] Determine physiological state data before exercise, physiological state data during exercise, and physiological state data after exercise from the exercise physiological state data;

[0021] Matching the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving with preset driving physiological state range data to determine abnormal driving physiological state data;

[0022] The physiological state data before exercise, the physiological state data during exercise and the physiological state data after exercise are matched with preset exercise physiological state range data to determine abnormal exercise physiological state data.

[0023] In an optional embodiment provided by the present invention, matching the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving with preset driving physiological state range data to determine abnormal driving physiological state data includes:

[0024] Acquire preset driving physiological state range data corresponding to the driving duration of the first time period; the preset driving physiological state range data includes physiological state range data before driving, physiological state range data during driving, and physiological state range data after driving;

[0025] The data of the physiological state data before driving, the physiological state data during driving and the physiological state data after driving that are not in the corresponding physiological state range data before driving, the physiological state range data during driving and the physiological state range data after driving are determined as abnormal driving physiological state data.

[0026] In an optional embodiment provided by the present invention, matching the physiological state data before exercise, the physiological state data during exercise, and the physiological state data after exercise with preset exercise physiological state range data to determine abnormal exercise physiological state data includes:

[0027] Acquire preset exercise physiological state range data corresponding to the exercise duration of the second time period; the preset exercise physiological state range data includes physiological state range data before exercise, physiological state range data during exercise, and physiological state range data after exercise;

[0028] The data of the physiological state data before exercise, the physiological state data during exercise and the physiological state data after exercise that are not in the corresponding physiological state range data before exercise, the physiological state range data during exercise and the physiological state range data after exercise are determined as abnormal exercise physiological state data.

[0029] In an optional embodiment provided by the present invention, the processor is further configured to:

[0030] Determine a driving weight that affects the physiological data based on the data change range of the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving;

[0031] The data change range of the physiological state data before exercise, the physiological state data during exercise and the physiological state data after exercise is used to determine the exercise weight that affects the physiological data.

[0032] In an optional embodiment provided by the present invention, the converting of abnormal driving physiological state data and / or abnormal sports physiological state data into a physiological state feature vector matrix includes:

[0033] Obtaining a first state feature vector matrix according to the product of the abnormal driving physiological state data and the driving weight affecting the physiological data;

[0034] Obtaining a second state feature vector matrix according to the product of the abnormal sports physiological state data and the sports weight affecting the physiological data;

[0035] The first state eigenvector matrix and the second state eigenvector matrix are determined as physiological state eigenvector matrices.

[0036] In an optional embodiment provided by the present invention, the step of inputting the physiological state feature vector matrix into a cardiovascular prediction model to obtain a cardiovascular health prediction result of the user includes:

[0037] Inputting the first state eigenvector matrix and the second state eigenvector matrix into a cardiovascular prediction model to obtain a first prediction value and a second prediction value;

[0038] The cardiovascular health prediction result of the user is obtained by the following calculation formula:

[0039]

[0040] Among them, y is the user's cardiovascular health prediction result, y1 is the first prediction value, y2 is the second prediction value, x1 is the probability value corresponding to the first prediction value, x2 is the probability value corresponding to the second prediction value, a, b are preset probability values, a>b.

[0041] An embodiment of the present invention provides a cardiovascular health data early warning method based on a health tag, the method is applied to a cardiovascular early warning device, and the method includes:

[0042] Acquiring vehicle driving data, physiological status data, and sensor data of the user;

[0043] Determine the vehicle driving time period and the user exercise time period corresponding to the user respectively according to the vehicle driving data and the sensor data;

[0044] Determining driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period;

[0045] Analyzing the driving physiological state data and the sports physiological state data to determine abnormal driving physiological state data and / or abnormal sports physiological state data;

[0046] converting abnormal driving physiological state data and / or abnormal sports physiological state data into a physiological state feature vector matrix;

[0047] The physiological state feature vector matrix is ​​input into a cardiovascular prediction model to obtain a cardiovascular health prediction result of the user, wherein the cardiovascular prediction model is trained based on the sample physiological state feature vector matrix and its cardiovascular health label.

[0048] The present invention provides a cardiovascular health data early warning system and method based on health labels. The cardiovascular early warning device first obtains the user's vehicle driving data, physiological state data, and sensor data; determines the user's corresponding vehicle driving time period and user exercise time period respectively according to the vehicle driving data and the sensor data; determines the driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period; analyzes the driving physiological state data and the exercise physiological state data to determine abnormal driving physiological state data and / or abnormal exercise physiological state data; converts the abnormal driving physiological state data and / or abnormal exercise physiological state data into a physiological state feature vector matrix; inputs the physiological state feature vector matrix into a cardiovascular prediction model to obtain the user's cardiovascular health prediction result, and the cardiovascular prediction model is trained according to the sample physiological state feature vector matrix and its cardiovascular health label. Compared with cardiovascular monitoring and early warning, which usually requires doctors to monitor patients on site, the present invention processes the vehicle driving data and physiological state data obtained by the vehicle-mounted device and the wearable device respectively to obtain abnormal driving physiological state data and / or abnormal exercise physiological state data, and then determines the cardiovascular health prediction results corresponding to the abnormal driving physiological state data and / or abnormal exercise physiological state data based on the cardiovascular prediction model. Therefore, the present invention can automatically calculate the user's cardiovascular prediction results without manual monitoring intervention, thereby improving the accuracy and timeliness of cardiovascular health prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1A structural diagram of a cardiovascular health data early warning system based on health tags provided for this application;

[0050] Figure 2 A flowchart of a cardiovascular health data early warning method based on health tags provided in this application;

[0051] Figure 3 A flowchart of a cardiovascular prediction model training method provided in this application;

[0052] Figure 4 This is a schematic diagram of the structure of a cardiovascular early warning device provided in this application. DETAILED DESCRIPTION

[0053] In order to better understand the above-mentioned technical scheme, the technical scheme of the embodiments of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the embodiments of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments may be combined with each other.

[0054] See also Figure 1 , a cardiovascular health data early warning system based on health tags is provided in an embodiment of the present invention, the system comprises: a vehicle-mounted device, a wearable device, and a cardiovascular early warning device, wherein the vehicle-mounted device and the wearable device are communicatively connected with the cardiovascular early warning device. The vehicle-mounted device is used to send the user's vehicle driving data to the cardiovascular early warning device; the wearable device is used to send the user's physiological state data and sensor data to the cardiovascular early warning device.

[0055] Specific as Figure 2 As shown, the cardiovascular early warning device in this embodiment includes a processor, and the processor is used to execute the following cardiovascular health data early warning method based on health tags, and the execution steps of the method are:

[0056] S201, obtaining the user's vehicle driving data, physiological status data, and sensor data.

[0057] The vehicle driving data includes the user's driving time, driving habits, driving speed, mileage, etc.; the physiological state data includes the user's heart rate, blood pressure, heartbeat, etc., which are not specifically limited in this embodiment. This embodiment can count the user's exercise time, steps, running steps, cycling distance, etc. through sensor data.

[0058] S202: Determine a vehicle driving time period and a user exercise time period corresponding to the user according to the vehicle driving data and the sensor data.

[0059] The vehicle driving time period is the time period in which the user drives the vehicle within a unit time. If the unit time is 1 day, the time period in which the user drives the vehicle within a day can be determined based on the vehicle driving data. The time period can be one or more; the user exercise time period is the time period in which the user exercises within a unit time. If the unit time is 1 day, the time period in which the user exercises within a day can be determined based on the user exercise time period. For example, two time periods are counted, one for exercise from 7:00 to 8:00 in the morning and the other for exercise from 18:00 to 18:30 in the evening.

[0060] In an optional embodiment provided by the present application, the vehicle driving time period and user exercise time period corresponding to the user are determined respectively according to the vehicle driving data and the sensor data, including: determining the driving start time and the driving end time of each driving of the user every day according to the vehicle driving data; and determining the exercise start time and the exercise end time of each exercise of the user every day according to the sensor data; determining the vehicle driving time period of the user each time every day by the driving start time and the driving end time of each driving every day; and determining the user exercise time period of the user each time every day by the exercise start time and the exercise end time of each exercise every day.

[0061] S203: Determine driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period.

[0062] In an optional embodiment provided in the present application, the driving physiological state data and the exercise physiological state data are determined from the physiological state data based on the vehicle driving time period and the user exercise time period, including: determining the time period including each vehicle driving time period as the first time period; determining the time period including each user exercise time period as the second time period, determining the physiological state data corresponding to the first time period in the physiological state data as the driving physiological state data, and determining the physiological state data corresponding to the second time period in the physiological state data as the exercise physiological state data.

[0063] The start time of the first time period is earlier than the driving start time of the vehicle driving time period and the end time is later than the driving end time of the vehicle driving time period; the start time of the second time period is earlier than the exercise start time of the user exercise time period and the end time is later than the exercise end time of the user exercise time period. For example, if the vehicle driving time period is 18:03-19:05, the first time period can be 18:00-19:10; if the user exercise time period is 8:10-9:10, the second time period is 18:05-19:15.

[0064] S204: Analyze the driving physiological state data and the sports physiological state data to determine abnormal driving physiological state data and / or abnormal sports physiological state data.

[0065] In this embodiment, the driving physiological state data is analyzed to determine abnormal driving physiological state data, and the exercise physiological state data is analyzed to determine abnormal exercise physiological state data. It should be noted that, since the user has good driving habits and exercise habits, there may not be abnormal driving physiological state data and / or abnormal exercise physiological state data during the analysis of the driving physiological state data and the exercise physiological state data. In this case, this embodiment can directly convert the driving physiological state data and the exercise physiological state data into a physiological state feature vector matrix, and then input the physiological state feature vector matrix into the cardiovascular prediction model to obtain the user's cardiovascular health prediction result.

[0066] In an optional embodiment provided by the present application, the analyzing the driving physiological state data and the exercise physiological state data to determine abnormal driving physiological state data and / or abnormal exercise physiological state data includes:

[0067] S2041, determining physiological state data before driving, physiological state data during driving, and physiological state data after driving from the driving physiological state data.

[0068] Among them, the physiological state data before driving and the physiological state data after driving are statistically averaged, and the physiological state data during driving selects the physiological state data of the time period in which the data before and after change exceeds a certain range. For example, if the vehicle driving time period is 18:03-19:05, the first time period can be 18:00-19:10, and the average value of the physiological state data between 18:00-18:03 is determined as the physiological state data before driving, and the average value of the physiological state data between 19:05-19:10 is determined as the physiological state data after driving.

[0069] S2042, determining the physiological state data before exercise, the physiological state data during exercise, and the physiological state data after exercise from the exercise physiological state data.

[0070] Among them, the physiological state data before exercise and the physiological state data after exercise are statistically averaged, and the physiological state data during exercise selects the physiological state data of the time period in which the data before and after change exceeds a certain amplitude. For example, if the user's exercise time period is 18:03-19:05, the second time period can be 18:00-19:10, and the average value of the physiological state data between 18:00-18:03 is determined as the physiological state data before exercise, and the average value of the physiological state data between 19:05-19:10 is determined as the physiological state data after exercise.

[0071] S2043, matching the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving with preset driving physiological state range data to determine abnormal driving physiological state data.

[0072] In an optional embodiment provided in the present application, the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving are matched with preset driving physiological state range data to determine abnormal driving physiological state data, including: obtaining preset driving physiological state range data corresponding to the driving time of the first time period; the preset driving physiological state range data includes the physiological state range data before driving, the physiological state range data during driving, and the physiological state range data after driving; determining the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving that are not in the corresponding physiological state range data before driving, the physiological state range data during driving, and the physiological state range data after driving as abnormal driving physiological state data.

[0073] S2044, matching the physiological state data before exercise, the physiological state data during exercise, and the physiological state data after exercise with preset exercise physiological state range data to determine abnormal exercise physiological state data.

[0074] In an optional embodiment provided by the present application, the physiological state data before exercise, the physiological state data during exercise, and the physiological state data after exercise are matched with preset exercise physiological state range data to determine abnormal exercise physiological state data, including: obtaining preset exercise physiological state range data corresponding to the exercise duration of the second time period; the preset exercise physiological state range data includes the physiological state range data before exercise, the physiological state range data during exercise, and the physiological state range data after exercise; the physiological state data before exercise, the physiological state data during exercise, and the physiological state data after exercise are not in the corresponding physiological state range data before exercise, the physiological state range data during exercise, and the physiological state range data after exercise as abnormal exercise physiological state data.

[0075] S205: Convert the abnormal driving physiological state data and / or the abnormal sports physiological state data into a physiological state feature vector matrix.

[0076] In an optional embodiment provided in the present application, the driving weight that affects the physiological data can be determined by the data change amplitude of the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving; the exercise weight that affects the physiological data can be determined by the data change amplitude of the physiological state data before exercise, the physiological state data during exercise, and the physiological state data after exercise.

[0077] In an optional embodiment provided in the present application, the converting of abnormal driving physiological state data and / or abnormal motion physiological state data into a physiological state feature vector matrix includes: obtaining a first state feature vector matrix according to the product of the abnormal driving physiological state data and the driving weight affecting the physiological data; obtaining a second state feature vector matrix according to the product of the abnormal motion physiological state data and the motion weight affecting the physiological data; and determining the first state feature vector matrix and the second state feature vector matrix as physiological state feature vector matrices.

[0078] S206, inputting the physiological state feature vector matrix into the cardiovascular prediction model to obtain the cardiovascular health prediction result of the user, the cardiovascular prediction model is trained based on the sample physiological state feature vector matrix and its cardiovascular health label.

[0079] Among them, the cardiovascular health label can be set according to the severity of vascular disease in the sample data center. For example, the cardiovascular health label can be represented by numbers 1-5. The larger the number, the higher the severity of the cardiovascular disease it represents.

[0080] In an optional embodiment provided in the present application, inputting the physiological state feature vector matrix into a cardiovascular prediction model to obtain a cardiovascular health prediction result of the user includes:

[0081] Inputting the first state eigenvector matrix and the second state eigenvector matrix into a cardiovascular prediction model to obtain a first prediction value and a second prediction value;

[0082] The cardiovascular health prediction result of the user is obtained by the following calculation formula:

[0083]

[0084] Among them, y is the user's cardiovascular health prediction result, y1 is the first prediction value, y2 is the second prediction value, x1 is the probability value corresponding to the first prediction value, x2 is the probability value corresponding to the second prediction value, a, b are preset probability values, a>b.

[0085] Furthermore, after obtaining the cardiovascular health prediction result of the user through the cardiovascular prediction model, the present embodiment can also obtain the user's dietary habit data, drug treatment data, and work and rest habit data, and then determine the user's cardiovascular health prediction result based on the dietary habit data, drug treatment data, and work and rest habit data. Specifically, the present embodiment can determine the user's cardiovascular health prediction result by matching the data in a preset mapping table, which stores scores corresponding to various dietary habit data, drug treatment data, and work and rest habit data, and then determine the cardiovascular health prediction result by statistically calculating the scores.

[0086] In this embodiment, after obtaining the cardiovascular health prediction result of the user through the preset mapping table, the driving weight affecting the physiological data and the exercise weight affecting the physiological data can be adjusted based on the cardiovascular health prediction result. For example, the driving weight determined by step S205 is 0.45, the exercise weight is 0.55, the first prediction value obtained by the cardiovascular prediction model is 4, the second prediction value is 5, and the cardiovascular health prediction result of the user obtained through the preset mapping table is 4, that is, the first prediction result is the same as the cardiovascular health prediction result of the user obtained through the preset mapping table, then the driving weight in the first state eigenvector matrix corresponding to the first prediction value is increased, and the exercise weight is decreased. Specifically, the driving weight is adjusted from 0.45 to 0.5, and the exercise weight is adjusted from 0.55 to 0.5.

[0087] The embodiment of the present application provides a cardiovascular health data warning method based on health labels. The cardiovascular warning device first obtains the user's vehicle driving data, physiological state data, and sensor data; determines the user's corresponding vehicle driving time period and user exercise time period according to the vehicle driving data and the sensor data; determines the driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period; analyzes the driving physiological state data and the exercise physiological state data to determine abnormal driving physiological state data and / or abnormal exercise physiological state data; converts the abnormal driving physiological state data and / or abnormal exercise physiological state data into a physiological state feature vector matrix; and inputs the physiological state feature vector matrix into a cardiovascular prediction model to obtain the user's cardiovascular health prediction result, wherein the cardiovascular prediction model is trained according to the sample physiological state feature vector matrix and its cardiovascular health label. Compared with cardiovascular monitoring and early warning, which usually requires doctors to monitor patients on site, the present invention processes the vehicle driving data and physiological state data obtained by the vehicle-mounted device and the wearable device respectively to obtain abnormal driving physiological state data and / or abnormal exercise physiological state data, and then determines the cardiovascular health prediction results corresponding to the abnormal driving physiological state data and / or abnormal exercise physiological state data based on the cardiovascular prediction model. Therefore, the present invention can automatically calculate the user's cardiovascular prediction results without manual monitoring intervention, thereby improving the accuracy and timeliness of cardiovascular health prediction.

[0088] Specific as Figure 3 As shown, the present application provides a flow chart of a cardiovascular prediction model training method, and the training process of the cardiovascular prediction model is:

[0089] S301, obtaining sample data for training a cardiovascular prediction model and its corresponding cardiovascular health labels.

[0090] Among them, the sample data includes vehicle driving data, physiological status data, and sensor data.

[0091] S302: Determine the vehicle driving time period and the user exercise time period corresponding to the user according to the vehicle driving data and the sensor data.

[0092] S303: Determine driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period.

[0093] S304: Analyze the driving physiological state data and the sports physiological state data to determine abnormal driving physiological state data and / or abnormal sports physiological state data.

[0094] S305: Convert the abnormal driving physiological state data and / or the abnormal sports physiological state data into a physiological state feature vector matrix.

[0095] It should be noted that the description contents of steps S302 to S305 in this embodiment are the same as those in Figure 2 The description of the corresponding steps in is the same, and this embodiment will not be repeated here.

[0096] S306, training the model according to the physiological state feature vector matrix and its cardiovascular health label to obtain a cardiovascular prediction model.

[0097] Among them, the physiological state characteristic vector matrix includes a first state characteristic vector matrix determined according to the product of abnormal driving physiological state data and the driving weight affecting the physiological data; and a second state characteristic vector matrix determined according to the product of abnormal motion physiological state data and the motion weight affecting the physiological data.

[0098] In this embodiment, the model is trained according to the physiological state eigenvector matrix and its cardiovascular health label to obtain the cardiovascular prediction model, including: inputting the first state eigenvector matrix and the second state eigenvector matrix into the cardiovascular prediction model to obtain the first prediction value and the second prediction value, and calculating the final prediction value according to the first prediction value and the second prediction value, calculating the loss value through the final prediction value and the cardiovascular health label, and when the loss value is less than the preset value, stopping the training of the cardiovascular prediction model.

[0099] Among them, the final prediction value can be obtained by calculating the following formula:

[0100]

[0101] Among them, y is the final prediction value, y1 is the first prediction value, y2 is the second prediction value, x1 is the probability value corresponding to the first prediction value, x2 is the probability value corresponding to the second prediction value, a and b are preset probability values, a>b.

[0102] Specifically, in this embodiment, the loss value can be calculated by the following function loss formula:

[0103]

[0104] Among them, n is the number of sample data, y 1i is the first predicted value corresponding to the i-th sample data, y 2i is the second predicted value corresponding to the i-th sample data, z i is the cardiovascular health label corresponding to the i-th sample data, y iis the final prediction value corresponding to the i-th sample data, α is the driving weight, β is the motion weight, and γ is the preset weight coefficient. Among them, γ>α+β.

[0105] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0106] In one embodiment, a cardiovascular early warning device is provided, which corresponds one-to-one to the cardiovascular health data early warning method based on health tags in the above embodiment. Figure 4 As shown, the functional modules of the cardiovascular early warning device are described in detail as follows:

[0107] An acquisition module 41 is used to acquire the vehicle driving data, physiological state data, and sensor data of the user;

[0108] A determination module 42, configured to determine a vehicle driving time period and a user exercise time period corresponding to the user according to the vehicle driving data and the sensor data;

[0109] The determination module 42 is further used to determine driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period;

[0110] The determination module 42 is further used to analyze the driving physiological state data and the sports physiological state data to determine abnormal driving physiological state data and / or abnormal sports physiological state data;

[0111] A conversion module 43, used for converting abnormal driving physiological state data and / or abnormal sports physiological state data into a physiological state feature vector matrix;

[0112] The prediction module 44 is used to input the physiological state feature vector matrix into a cardiovascular prediction model to obtain a cardiovascular health prediction result of the user. The cardiovascular prediction model is trained based on the sample physiological state feature vector matrix and its cardiovascular health label.

[0113] In an optional embodiment provided by the present invention, the determination module 42 is specifically configured to:

[0114] Determine the driving start time and driving end time of each driving of the user every day according to the vehicle driving data; and determine the exercise start time and exercise end time of each exercise of the user every day according to the sensor data;

[0115] The vehicle driving time period of the user each time each day is determined by the driving start time and driving end time of each driving each day; and the user exercise time period of the user each time each day is determined by the exercise start time and exercise end time of each exercise each day.

[0116] In an optional embodiment provided by the present invention, the determination module 42 is specifically configured to:

[0117] Determine a time period including each vehicle driving time period as a first time period; determine a time period including each user exercise time period as a second time period, wherein the start time of the first time period is earlier than the driving start time of the vehicle driving time period and the end time is later than the driving end time of the vehicle driving time period; the start time of the second time period is earlier than the exercise start time of the user exercise time period and the end time is later than the exercise end time of the user exercise time period;

[0118] The physiological state data corresponding to the first time period in the physiological state data is determined as driving physiological state data, and the physiological state data corresponding to the second time period in the physiological state data is determined as sports physiological state data.

[0119] In an optional embodiment provided by the present invention, the determination module 42 is specifically configured to:

[0120] Determine physiological state data before driving, physiological state data during driving, and physiological state data after driving from the driving physiological state data;

[0121] Determine physiological state data before exercise, physiological state data during exercise, and physiological state data after exercise from the exercise physiological state data;

[0122] Matching the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving with preset driving physiological state range data to determine abnormal driving physiological state data;

[0123] The physiological state data before exercise, the physiological state data during exercise and the physiological state data after exercise are matched with preset exercise physiological state range data to determine abnormal exercise physiological state data.

[0124] In an optional embodiment provided by the present invention, the determination module 42 is specifically configured to:

[0125] Acquire preset driving physiological state range data corresponding to the driving duration of the first time period; the preset driving physiological state range data includes physiological state range data before driving, physiological state range data during driving, and physiological state range data after driving;

[0126] The data of the physiological state data before driving, the physiological state data during driving and the physiological state data after driving that are not in the corresponding physiological state range data before driving, the physiological state range data during driving and the physiological state range data after driving are determined as abnormal driving physiological state data.

[0127] In an optional embodiment provided by the present invention, the determination module 42 is specifically configured to:

[0128] Acquire preset exercise physiological state range data corresponding to the exercise duration of the second time period; the preset exercise physiological state range data includes physiological state range data before exercise, physiological state range data during exercise, and physiological state range data after exercise;

[0129] The data of the physiological state data before exercise, the physiological state data during exercise and the physiological state data after exercise that are not in the corresponding physiological state range data before exercise, the physiological state range data during exercise and the physiological state range data after exercise are determined as abnormal exercise physiological state data.

[0130] In an optional embodiment provided by the present invention, the determination module 42 is further configured to:

[0131] Determine a driving weight that affects the physiological data based on the data change range of the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving;

[0132] The data change range of the physiological state data before exercise, the physiological state data during exercise and the physiological state data after exercise is used to determine the exercise weight that affects the physiological data.

[0133] In an optional embodiment provided by the present invention, the conversion module 43 is specifically configured to:

[0134] Obtaining a first state feature vector matrix according to the product of the abnormal driving physiological state data and the driving weight affecting the physiological data;

[0135] Obtaining a second state feature vector matrix according to the product of the abnormal sports physiological state data and the sports weight affecting the physiological data;

[0136] The first state eigenvector matrix and the second state eigenvector matrix are determined as physiological state eigenvector matrices.

[0137] In an optional embodiment provided by the present invention, the prediction module 44 is specifically used for:

[0138] Inputting the first state eigenvector matrix and the second state eigenvector matrix into a cardiovascular prediction model to obtain a first prediction value and a second prediction value;

[0139] The cardiovascular health prediction result of the user is obtained by the following calculation formula:

[0140]

[0141] Among them, y is the user's cardiovascular health prediction result, y1 is the first prediction value, y2 is the second prediction value, x1 is the probability value corresponding to the first prediction value, x2 is the probability value corresponding to the second prediction value, a, b are preset probability values, a>b.

[0142] For the specific definition of the cardiovascular early warning device, please refer to the definition of the cardiovascular health data early warning method based on health tags above, which will not be repeated here. Each module in the above-mentioned device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0143] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0144] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A cardiovascular health data early warning system based on health tags, characterized in that: The system comprises: a vehicle-mounted device, a wearable device, and a cardiovascular early warning device. The vehicle-mounted device and the wearable device are in communication with the cardiovascular early warning device. The vehicle-mounted device is used to send the user's vehicle driving data to the cardiovascular early warning device; the wearable device is used to send the user's physiological state data and sensor data to the cardiovascular early warning device; the cardiovascular early warning device comprises a processor, and the processor is used to execute: Acquiring vehicle driving data, physiological status data, and sensor data of the user; Determine the vehicle driving time period and the user exercise time period corresponding to the user respectively according to the vehicle driving data and the sensor data; Determining driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period; Analyzing the driving physiological state data and the sports physiological state data to determine abnormal driving physiological state data and / or abnormal sports physiological state data; converting abnormal driving physiological state data and / or abnormal sports physiological state data into a physiological state feature vector matrix; Inputting the physiological state feature vector matrix into a cardiovascular prediction model to obtain a cardiovascular health prediction result of the user, wherein the cardiovascular prediction model is trained based on the sample physiological state feature vector matrix and its cardiovascular health label; The determining the vehicle driving time period and the user exercise time period corresponding to the user respectively according to the vehicle driving data and the sensor data includes: Determine the driving start time and driving end time of each driving of the user every day according to the vehicle driving data; and determine the exercise start time and exercise end time of each exercise of the user every day according to the sensor data; Determine the vehicle driving time period of the user each time each day by the driving start time and driving end time of each driving each day; and determine the user exercise time period of the user each time each day by the exercise start time and exercise end time of each exercise each day; The determining of driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period includes: Determine a time period including each vehicle driving time period as a first time period; determine a time period including each user exercise time period as a second time period, wherein the start time of the first time period is earlier than the driving start time of the vehicle driving time period and the end time is later than the driving end time of the vehicle driving time period; the start time of the second time period is earlier than the exercise start time of the user exercise time period and the end time is later than the exercise end time of the user exercise time period; Determine the physiological state data corresponding to the first time period in the physiological state data as driving physiological state data, and determine the physiological state data corresponding to the second time period in the physiological state data as sports physiological state data; The analyzing the driving physiological state data and the sports physiological state data to determine abnormal driving physiological state data and / or abnormal sports physiological state data includes: Determine physiological state data before driving, physiological state data during driving, and physiological state data after driving from the driving physiological state data; Determine physiological state data before exercise, physiological state data during exercise, and physiological state data after exercise from the exercise physiological state data; Matching the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving with preset driving physiological state range data to determine abnormal driving physiological state data; Matching the physiological state data before exercise, the physiological state data during exercise, and the physiological state data after exercise with the preset exercise physiological state range data to determine abnormal exercise physiological state data; The processor is further configured to: Determine a driving weight that affects the physiological data based on the data change range of the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving; Determine the exercise weight that affects the physiological data by using the data change amplitude of the physiological state data before exercise, the physiological state data during exercise, and the physiological state data after exercise; The step of converting abnormal driving physiological state data and / or abnormal sports physiological state data into a physiological state feature vector matrix includes: Obtaining a first state feature vector matrix according to the product of the abnormal driving physiological state data and the driving weight affecting the physiological data; Obtaining a second state feature vector matrix according to the product of the abnormal sports physiological state data and the sports weight affecting the physiological data; determining the first state eigenvector matrix and the second state eigenvector matrix as physiological state eigenvector matrices; The step of inputting the physiological state feature vector matrix into a cardiovascular prediction model to obtain a cardiovascular health prediction result of the user includes: Inputting the first state eigenvector matrix and the second state eigenvector matrix into a cardiovascular prediction model to obtain a first prediction value and a second prediction value; The cardiovascular health prediction result of the user is obtained by the following calculation formula: in, Predict cardiovascular health results for users, is the first predicted value, is the second predicted value, is the probability value corresponding to the first predicted value, is the probability value corresponding to the second predicted value, is the preset probability value, ; Specifically, the loss value is calculated by the following function loss formula; in, is the number of sample data, is the first predicted value corresponding to the i-th sample data, is the second predicted value corresponding to the i-th sample data, is the cardiovascular health label corresponding to the i-th sample data, is the final predicted value corresponding to the i-th sample data, for , is the preset weight coefficient; .

2. The system according to claim 1, characterized in that The step of matching the physiological state data before driving, the physiological state data during driving, and the physiological state data after driving with preset driving physiological state range data to determine abnormal driving physiological state data includes: Acquire preset driving physiological state range data corresponding to the driving duration of the first time period; the preset driving physiological state range data includes physiological state range data before driving, physiological state range data during driving, and physiological state range data after driving; The data of the physiological state data before driving, the physiological state data during driving and the physiological state data after driving that are not in the corresponding physiological state range data before driving, the physiological state range data during driving and the physiological state range data after driving are determined as abnormal driving physiological state data.

3. The system according to claim 1, characterized in that The step of matching the physiological state data before exercise, the physiological state data during exercise, and the physiological state data after exercise with preset exercise physiological state range data to determine abnormal exercise physiological state data includes: Acquire preset exercise physiological state range data corresponding to the exercise duration of the second time period; the preset exercise physiological state range data includes physiological state range data before exercise, physiological state range data during exercise, and physiological state range data after exercise; The data of the physiological state data before exercise, the physiological state data during exercise and the physiological state data after exercise that are not in the corresponding physiological state range data before exercise, the physiological state range data during exercise and the physiological state range data after exercise are determined as abnormal exercise physiological state data.

4. A cardiovascular health data early warning method based on health labels, characterized in that: The method is applied to the system as claimed in claim 1, and the method comprises: Acquiring vehicle driving data, physiological status data, and sensor data of the user; Determine the vehicle driving time period and the user exercise time period corresponding to the user respectively according to the vehicle driving data and the sensor data; Determining driving physiological state data and exercise physiological state data from the physiological state data based on the vehicle driving time period and the user exercise time period; Analyzing the driving physiological state data and the sports physiological state data to determine abnormal driving physiological state data and / or abnormal sports physiological state data; converting abnormal driving physiological state data and / or abnormal sports physiological state data into a physiological state feature vector matrix; The physiological state feature vector matrix is ​​input into a cardiovascular prediction model to obtain a cardiovascular health prediction result of the user, wherein the cardiovascular prediction model is trained based on the sample physiological state feature vector matrix and its cardiovascular health label.

Images