Disease prevention system based on AI technology and intelligent wearing

Abstract

The invention discloses a disease prevention system based on an AI technology and intelligent wearing, and belongs to the field of intelligent wearing, the disease prevention system comprises an intelligent wearing device and an AI analysis platform, the intelligent wearing device comprises a data acquisition module, a vibration notification module and a display notification module, the AI analysis platform comprises a data receiving module, a data analysis module, an early warning notification module and a data import module; the data acquisition module is used for acquiring running monitoring data and health monitoring data of a user and food pictures shot by the user through a mobile phone before eating each time, and sending the acquired data to the data receiving module; and the data receiving module is used for receiving and integrating all the data and sending the integrated data to the data analysis module. According to the invention, the health data of the user can be monitored in real time to prevent diseases, and the daily exercise and diet of the user can be monitored and analyzed to prevent diseases.

Description

Technical Field

[0001] The present invention relates to the field of smart wearables, and in particular to a disease prevention system based on AI technology and smart wearables. Background Art

[0002] With the advancement and development of science and technology, smart wearable devices have gradually become popular in people's daily lives. Common smart wearable devices include smart watches, smart bracelets, smart glasses, smart headphones, smart clothing, smart insoles, etc., and smart watches are the most widely used among them. Smart watches have built-in intelligent systems, are equipped with smart phone systems and are connected to the Internet to achieve multi-functions. In addition to having all the functions of a watch, they can also monitor the user's health data, such as heart rate, blood pressure, and blood oxygen. When the user is doing outdoor sports, if the heart rate, blood pressure, and blood oxygen are abnormal, it means that the user's health is not good and he needs to pay attention to disease prevention in advance.

[0003] Although existing smart watches can monitor users' health data in real time to prevent diseases, this method of disease prevention is too single. If users do not pay attention to their daily exercise and diet, they will also be at risk of contracting diseases. Therefore, how to monitor and analyze users' daily exercise and diet to prevent diseases is a difficult problem that we urgently need to solve.

[0004] Therefore, those skilled in the art provide a disease prevention system based on AI technology and smart wearables to solve the problems raised in the above background technology. Summary of the invention

[0005] In view of the defects in the prior art, the present invention provides a disease prevention system based on AI technology and smart wear, including a smart wearable device and an AI analysis platform, wherein the smart wearable device includes a data acquisition module, a vibration notification module and a display notification module, and the AI ​​analysis platform includes a data receiving module, a data analysis module, an early warning notification module and a data import module; The data acquisition module is used to collect the user's running monitoring data, health monitoring data and food pictures taken by the user with a mobile phone before each meal, and send the collected data to the data receiving module; the data receiving module is used to receive all the data for integration, and send the integrated data to the data analysis module; the data import module is used to import preset data to the data analysis module; the data analysis module is used to receive all the data for analysis, and output the analysis results to the early warning notification module, the vibration notification module and the display notification module; the early warning notification module decides whether to send an early warning notification to the user's guardian based on the analysis results; the vibration notification module notifies the user through the vibration of the smart wearable device, and adjusts the vibration frequency of the smart wearable device based on the analysis results; the display notification module is used to display the analysis results to the user and provide a flashing reminder.

[0006] As a further solution of the present invention: the specific process of the data analysis module analyzing the collected food pictures is as follows: Perform image recognition on food pictures to obtain the type of food identified, marked as A; Search all reviews of food type A on the public food safety platform through big data, screen the searched food reviews, and then perform text recognition on the screened food reviews, and plan the positive reviews into the first sequence, and the negative reviews into the second sequence; Mark the number of reviews in the first sequence as B1, and the number of reviews in the second sequence as B2, and calculate the proportion of negative reviews of the identified food b1=B2 / (B1+B2); Search all the production processes of type A food through big data, analyze the production process that matches the identified food, and then mark the total number of steps in the analyzed production process as C; Match the identified food with the food database, calculate the calories of the identified food and mark it as D; Compare the negative review ratio b1 of the identified food with the preset value b2, the average number of processes C with the preset value c, and the calorie D with the preset value d; If any of b1>b2, C>c, D>d appears in the comparison result, it means that the food that the current user is about to eat has a first-level risk; If any two of b1>b2, C>c, and D>d appear in the comparison results, it means that the food that the current user is about to eat has a secondary risk; If all items of b1>b2, C>c, and D>d appear in the comparison results, it means that the food that the current user is about to eat has a third-level risk; If all items of b1≤b2, C≤c, and D≤d appear in the comparison results, it means that the food that the current user is about to eat does not pose a risk.

[0007] As a further solution of the present invention: the specific process of screening the searched food reviews is: All food evaluations of food type A found through big data search on the food safety public platform are marked as Pi, i=1···n, where n is a positive integer; Mark the city where the review corresponding to Pi is posted as pi, and then mark the city where the user is located as puser; If pi and p are in the same city, pi is planned to the first cluster. If pi and p are not in the same city, proceed to the next step. If pi and p are from the same city in the same province, pi will be planned to the second cluster. If pi and p are not from the same city in the same province, proceed to the next step. If pi and p are cities in adjacent provinces, pi is planned to the third cluster. If pi and p are cities in different provinces, pi is planned to the fourth cluster. All the food evaluations Pi corresponding to each pi in the first cluster are selected, half of the food evaluations Pi corresponding to each pi in the second cluster are randomly selected, one third of the food evaluations Pi corresponding to each pi in the third cluster are randomly selected, and one quarter of the food evaluations Pi corresponding to each pi in the fourth cluster are randomly selected; Output the filtered food reviews.

[0008] As a further solution of the present invention: the specific process of analyzing the production process matching the identified food is: Mark all the production processes of type A food searched through big data as Wi, i=1···n, where n is a positive integer; Get the origin of the production process Wi, marked as wi; Identify wi and select wi that are in the same area as user p in the user's city, where the same area means that the straight-line distance between wi and user p in the user's city is within 100 kilometers; The finished food images of the production process Wi corresponding to the selected wi are selected from the public platform through big data; The screened images are compared with the food images collected by the data collection module for similarity, and the production process corresponding to the image with the highest similarity is used as the production process that matches the identified food.

[0009] As a further solution of the present invention: the data acquisition module collects running monitoring data through a sensor component built into the smart wearable device, and the running monitoring data includes the user's running posture, stride, and step frequency data, wherein the data analysis module analyzes the collected running monitoring data in a specific process as follows: The collected user's running posture is marked as E, the running stride is marked as F, and the running frequency is marked as G; The posture E is compared with the preset posture e for similarity. If the similarity between E and e is less than the preset similarity, it means that the current user's running posture is abnormal; Compare the stride F with the preset stride f. If F>f, it means that the current user's running stride is abnormal. Compare the step frequency G with the preset step frequency g. If G>g, it means that the current user's running step frequency is abnormal. If any of the following items is found in the comparison result: abnormal running posture, abnormal running stride, or abnormal running frequency, it means that the current user's running exercise has a level 1 risk; If any two of the following are found in the comparison results: abnormal running posture, abnormal running stride, or abnormal running frequency, it means that the current user's running exercise has a secondary risk; If the comparison results show abnormal running posture, abnormal running stride, and abnormal running frequency, it means that the current user's running exercise has a third-level risk; If the comparison results do not show any of the abnormal running posture, abnormal running stride, or abnormal running frequency, it means that there is no risk in the current user's running exercise.

[0010] As a further solution of the present invention: the health monitoring data includes the monitored user heart rate data, blood pressure data and blood oxygen data, and the specific process of the data analysis module analyzing the collected health monitoring data is: The collected user's heart rate data is marked as X, blood pressure data is marked as Y, and blood oxygen data is marked as Z; Match X with a preset range value. If X is within the preset range value, it means that the user's heart rate is normal. If X is not within the preset range value, it means that the user's heart rate is abnormal. Match Y with the preset range value. If Y is within the preset range value, it means that the user's blood pressure is normal. If Y is not within the preset range value, it means that the user's blood pressure is abnormal. Match Z with the preset range value. If Z is within the preset range value, it means that the user's blood oxygen is normal. If Z is not within the preset range value, it means that the user's blood oxygen is abnormal. If any of abnormal heart rate, abnormal blood pressure, or abnormal blood oxygen is found in the analysis results, it means that the current user's physiological indicators are at level 1 risk; If any two of abnormal heart rate, abnormal blood pressure, and abnormal blood oxygen are found in the analysis results, it means that the current user's physiological indicators have secondary risks; If the analysis results show abnormal heart rate, abnormal blood pressure, and abnormal blood oxygen, it means that the current user's physiological indicators are at level 3 risk; If the analysis results do not show any of abnormal heart rate, abnormal blood pressure, or abnormal blood oxygen, it means that there is no risk in the current user's physiological indicators.

[0011] As a further solution of the present invention: the specific process of the vibration notification module notifying the user is as follows: If the analysis results show a level 1 risk, the vibration frequency of the smart wearable device will be adjusted to 60 vibrations per minute; If a level 2 risk appears in the analysis results, the vibration frequency of the smart wearable device will be adjusted to 100 vibrations per minute; If a third-level risk appears in the analysis results, the vibration frequency of the smart wearable device will be adjusted to two hundred vibrations per minute.

[0012] As a further solution of the present invention: the specific process of the display notification module for flashing reminder is: If a level 1 risk appears in the analysis results, the border of the display screen of the smart wearable device will flash white light; If a secondary risk appears in the analysis results, the border of the display screen of the smart wearable device will flash yellow; If a third-level risk appears in the analysis results, the border of the display screen of the smart wearable device will flash red.

[0013] As a further solution of the present invention: the specific process of the early warning notification module deciding whether to send an early warning notification to the user's guardian according to the analysis result is: Assign weight values ​​to the analysis results. If a first-level risk appears in the analysis results, assign a weight value V=1 to the analysis results; if a second-level risk appears in the analysis results, assign a weight value V=2 to the analysis results; if a third-level risk appears in the analysis results, assign a weight value V=3 to the analysis results; if multiple risks appear in the analysis results at the same time, add up the weight values ​​corresponding to each risk; Count the sum of the weight values ​​V corresponding to all analysis results in the past three days, marked as Vtotal; Vtotal is compared with the preset value v. If Vtotal>v, the early warning notification module sends an early warning notification to the user's guardian. If Vtotal≤v, the early warning notification module does not send an early warning notification to the user's guardian.

[0014] As a further solution of the present invention: the early warning notification is sent to the user's guardian in the form of a text message.

[0015] The beneficial effects of the present invention are embodied in: This application can not only monitor the user's health data in real time to prevent diseases, but also monitor and analyze the user's daily exercise and diet to prevent diseases. Among them, this application uses the camera function of the smart wearable device to take pictures of the food that the user eats, and then analyzes the calories, processing procedures, and online health reviews of the food from multiple angles to determine whether the food to be consumed is risky and remind the user in time. In addition, for the user's running exercise, the user's posture, stride, and step frequency are analyzed during running, and the user is reminded in time when the user's running is abnormal, to prevent the negative effects of running from causing damage to the user's body. In general, this application monitors the user's health in an all-round way from the user's diet, exercise, and physiological indicators, effectively preventing the occurrence of diseases and improving the safety and health of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0017] Figure 1 This is a structural block diagram of a disease prevention system based on AI technology and smart wearables. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] As mentioned in the background technology of this application, research has found that although existing smart watches can monitor users' health data in real time to prevent diseases, this method of preventing diseases is too single. If users do not pay attention to their daily exercise and diet, they will also be at risk of contracting diseases, which has certain defects.

[0020] In order to solve the above-mentioned defects, the present application discloses a disease prevention system based on AI technology and smart wearables, which can not only monitor the user's health data in real time to prevent diseases, but also monitor and analyze the user's daily exercise and diet to prevent diseases.

[0021] The following will describe in detail how the solution of the present application solves the above technical problems in conjunction with the accompanying drawings.

[0022] See also Figure 1In an embodiment of the present invention, a disease prevention system based on AI technology and smart wearables includes a smart wearable device and an AI analysis platform, wherein the smart wearable device includes a data acquisition module, a vibration notification module and a display notification module, and the AI ​​analysis platform includes a data receiving module, a data analysis module, an early warning notification module and a data import module; the data acquisition module is used to collect the user's running monitoring data, health monitoring data and food pictures taken by the user through a mobile phone before each meal, and send the collected data to the data receiving module; the data receiving module is used to receive all data for integration, and send the integrated data to the data analysis module; the data import module is used to import preset data to the data analysis module; the data analysis module is used to receive all data for analysis, and output the analysis results to the early warning notification module, the vibration notification module and the display notification module; the early warning notification module decides whether to send an early warning notification to the user's guardian according to the analysis results; the vibration notification module notifies the user through the vibration of the smart wearable device, and adjusts the vibration frequency of the smart wearable device according to the analysis results; the display notification module is used to display the analysis results to the user and flash a reminder. It should be noted that the mobile phone and the smart wearable device are connected via Bluetooth or other wireless methods. After taking the food picture, the mobile phone sends it to the smart wearable device for preprocessing, and then the data acquisition module collects the preprocessed food picture. Among them, image preprocessing is an important part of image processing. It aims to improve image quality and improve the efficiency and accuracy of subsequent processing steps. Image preprocessing usually includes the following steps: (1) Image digitization: converting analog images into digital images, which is the basis of image processing. Digital images are composed of pixels, and each pixel has a specific color value and brightness value. (2) Image denoising: removing noise from the image, such as random noise, salt and pepper noise, etc. Noise will affect the clarity and quality of the image, so it is necessary to perform denoising through filtering and other methods. Common denoising methods include mean filtering, median filtering, Gaussian filtering, etc. (3) Image enhancement: enhancing useful information in the image, while suppressing or removing useless information to improve the visual effect of the image. Image enhancement methods include contrast enhancement, sharpening, histogram equalization, etc. These methods can make the contours of objects in the image clearer and the details more obvious. (4) Image restoration: Restore information that may be lost or degraded during the acquisition, transmission or storage of an image. Image restoration usually addresses issues such as blur, distortion, and loss of an image to improve the fidelity and clarity of the image. (5) Image geometric transformation: Perform operations such as translation, rotation, scaling, and affine transformation on an image to change the geometric position or shape of the image. These transformations are often used for image correction, alignment, and standardization. (6) Image normalization: Adjust the pixel values ​​of an image to a certain range, usually normalizing the pixel values ​​to between 0 and 1.Normalization helps eliminate the brightness differences between different images, making subsequent processing more stable and reliable. (7) Image binarization: Convert grayscale images into binary images, that is, the pixels in the image have only two colors, black and white. Binarization is the basis of processing steps such as image segmentation and edge detection, which helps to simplify image information and improve processing efficiency. (8) Image filtering: Process images through filters to suppress or enhance certain frequency components in the image. Filtering operations include smoothing filtering and sharpening filtering. Smoothing filtering can remove image noise, and sharpening filtering can enhance image edge information. (9) Image segmentation: Divide the image into multiple regions with specific properties for subsequent analysis and processing. Image segmentation is an important problem in image processing and computer vision. Commonly used segmentation methods include threshold-based segmentation, edge-based segmentation, and region-based segmentation.

[0023] In this embodiment, the specific process of the data analysis module analyzing the collected food pictures is as follows: performing image recognition on the food pictures to obtain the type of identified food, which is marked as A; searching all the evaluations of food of type A on the food safety public platform through big data, screening the searched food evaluations, and then performing text recognition on the screened food evaluations, planning the positive evaluations into the first sequence, and planning the negative evaluations into the second sequence; marking the number of evaluations in the first sequence as B1, and the number of evaluations in the second sequence as B2, and calculating the proportion of negative evaluations of the identified food b1=B2 / (B1+B2); searching all the production processes of food of type A through big data, analyzing the production process that matches the identified food, and then marking the total number of processes in the analyzed production process as C. ; Match the identified food with the food database, calculate the calories of the identified food and mark it as D; compare the negative review ratio b1 of the identified food with the preset value b2, compare the average number of processes C with the preset value c, and compare the calories D with the preset value d; if any one of b1>b2, C>c, D>d appears in the comparison result, it means that the food that the current user is going to eat has a first-level risk; if any two of b1>b2, C>c, D>d appear in the comparison result, it means that the food that the current user is going to eat has a second-level risk; if all of b1>b2, C>c, D>d appear in the comparison result, it means that the food that the current user is going to eat has a third-level risk; if all of b1≤b2, C≤c, D≤d appear in the comparison result, it means that the food that the current user is going to eat does not have a risk. This application uses the camera function of the smart wearable device to take photos of the food the user is eating, and then conducts a multi-angle analysis from the calories, processing steps, and online health reviews of the food to determine whether the food to be consumed has risks and promptly remind the user.

[0024] In this embodiment, the specific process of screening the searched food evaluations is as follows: mark all food evaluations of type A food on the food safety public platform searched through big data as Pi, i=1···n, where n is a positive integer; mark the city where the evaluation corresponding to Pi is published as pi, and then mark the city where the user is located as p user; if pi and p user are in the same city, pi is planned to the first cluster, if pi and p user are not in the same city, proceed to the next step; if pi and p user are in the same province, pi is planned to the second cluster, if pi and p user are not in the same province, proceed to the next step; if pi and p user are in adjacent provinces, pi is planned to the third cluster, if pi and p user are not in adjacent provinces, pi is planned to the fourth cluster; all food evaluations Pi corresponding to each pi in the first cluster are screened out, half of the food evaluations Pi corresponding to each pi in the second cluster are randomly screened out, one third of the food evaluations Pi corresponding to each pi in the third cluster are randomly screened out; one quarter of the food evaluations Pi corresponding to each pi in the fourth cluster are randomly screened out; and the screened food evaluations are output. Food reviews are the most intuitive feedback from online users on the food they eat. When screening reviews of corresponding foods, this application follows the regional differentiation rules. The ingredients of the same type of food in the same region usually come from the same place, which means that the food reviews published closer to the user's location are more representative. For this reason, this application adopts differentiated extraction for food reviews in different regions, so that the final screened food reviews are not only more representative, but also more accurate.

[0025] In this embodiment, the specific process of analyzing the production process matching the identified food is as follows: mark all the production processes of the A type of food searched through big data as Wi, i=1···n, where n is a positive integer; obtain the origin of the production process Wi, marked as wi; identify wi, and filter out the wi that belongs to the same area as the user p in the user's city, where the same area means that the straight-line distance between wi and the user p in the user's city is within 100 kilometers; filter out the finished food pictures of the production process Wi corresponding to the filtered wi from the public platform through big data; compare the filtered pictures with the food pictures collected by the data acquisition module for similarity, and use the production process corresponding to the picture with the highest similarity as the production process matching the identified food. Different food production processes represent different health levels. Generally speaking, the simpler the production process, the healthier the food, and the more complex the production process, the more additives used or unhealthy substances produced in the production process. Among them, foods with simple production processes, such as fresh fruits, vegetables, whole grains, and unprocessed meats, usually retain the nutrients of the food itself, with no or few additional chemicals or preservatives added. These foods are often rich in fiber, vitamins, minerals, and other nutrients that are beneficial to the human body, and are therefore considered relatively healthy choices. However, some foods with complex production processes may add a large amount of unhealthy ingredients such as additives, preservatives, pigments, and sugars in order to improve taste, extend shelf life, or reduce costs. These ingredients may have a negative impact on human health, such as increasing the risk of chronic diseases such as cardiovascular disease, diabetes, and obesity. This application first filters out the food production process in the user's area, and then filters out the production process that is most similar to the food the user wants to eat, and then determines the total number of production process steps for the identified food.

[0026] In this embodiment, the data acquisition module collects running monitoring data through a sensor component built into the smart wearable device, and the running monitoring data includes the user's running posture, stride, and step frequency data, wherein the data analysis module analyzes the collected running monitoring data in a specific process as follows: marking the collected user's running posture as E, the running stride as F, and the running step frequency as G; comparing the posture E with the preset posture e for similarity, if the similarity between E and e is less than the preset similarity, it means that the current user's running posture is abnormal; comparing the stride F with the preset stride f, if F>f, it means that the current user's running stride is abnormal; comparing the step frequency G with the preset step frequency g A comparison is performed. If G>g, it means that the current user's running frequency is abnormal; if any one of the following items appears in the comparison result, it means that the current user's running exercise has a first-level risk; if any two of the following items appear in the comparison result, it means that the current user's running exercise has a second-level risk; if all of the following items appear in the comparison result, it means that the current user's running exercise has a third-level risk; if any one of the following items does not appear in the comparison result, it means that the current user's running exercise has no risk. Obtaining information such as the user's running posture, stride, frequency, and landing method through smart wearable devices mainly relies on a variety of built-in sensors and advanced algorithms in the device. The following are the specific steps and principles for implementing this process through smart wearable devices: 1. Sensor Technology (1.1) Acceleration sensor: Accelerometers are one of the most common sensors in smart wearable devices, used to measure the acceleration changes of the device in three-dimensional space. When running, by measuring the acceleration changes of the wrist or other parts of the body, the user's step frequency and stride length can be inferred.

[0027] Cadence calculation: By analyzing the periodic changes in the acceleration sensor data, the number of times the user's feet touch the ground per minute, i.e., the cadence, can be calculated.

[0028] Stride Estimation: Combining cadence and GPS location data (a well-known and common function of smart wearable devices), the user's stride length can be estimated. However, more accurate stride measurement usually requires combining other sensor data.

[0029] (1.2) Gyroscope: The gyroscope is used to monitor changes in angular velocity and can measure the rotation and tilt angle of the device in space. When running, the gyroscope can help analyze the user's body tilt angle and rotation speed to evaluate the stability and smoothness of the running posture.

[0030] (1.3) Geomagnetic sensor: The geomagnetic sensor is used to determine the orientation of the device. Although it is not directly used for running posture analysis, it can be used in conjunction with the gyroscope to provide more comprehensive motion status information.

[0031] 2. Algorithm Analysis (2.1) Posture recognition algorithm: By analyzing the data from sensors such as accelerometers and gyroscopes, combined with advanced posture recognition algorithms, the user's running posture can be evaluated in real time. These algorithms can identify key parameters such as the user's body tilt and rotation angle, and give corresponding posture evaluation results.

[0032] (2.2) Sports data analysis: The algorithms built into smart wearable devices will conduct in-depth analysis of the collected motion data, including basic information such as cadence, stride, speed, distance, as well as more complex running posture parameters.

[0033] In this embodiment, the health monitoring data includes the monitored user's heart rate data, blood pressure data and blood oxygen data. The specific process of the data analysis module analyzing the collected health monitoring data is as follows: the collected user's heart rate data is marked as X, the blood pressure data is marked as Y, and the blood oxygen data is marked as Z; X is matched with a preset range value, if X is within the preset range value, it means that the user's heart rate is normal, if X is not within the preset range value, it means that the user's heart rate is abnormal; Y is matched with a preset range value, if Y is within the preset range value, it means that the user's blood pressure is normal, if Y is not within the preset range value, it means that the user's blood pressure is abnormal; Z is matched with a preset range value, if If Z is within the preset range, it means that the user's blood oxygen is normal. If Z is not within the preset range, it means that the user's blood oxygen is abnormal. If any one of abnormal heart rate, abnormal blood pressure, and abnormal blood oxygen appears in the analysis results, it means that the current user's physiological indicators have a first-level risk. If any two of abnormal heart rate, abnormal blood pressure, and abnormal blood oxygen appear in the analysis results, it means that the current user's physiological indicators have a second-level risk. If all of abnormal heart rate, abnormal blood pressure, and abnormal blood oxygen appear in the analysis results, it means that the current user's physiological indicators have a third-level risk. If any one of abnormal heart rate, abnormal blood pressure, and abnormal blood oxygen does not appear in the analysis results, it means that the current user's physiological indicators have no risk. This setting is convenient for quickly detecting whether the user's heart rate, blood pressure, and blood oxygen are abnormal.

[0034] In this embodiment, the specific process of the vibration notification module notifying the user is as follows: if the analysis result shows a level 1 risk, the vibration frequency of the smart wearable device is adjusted to 60 vibrations per minute; if the analysis result shows a level 2 risk, the vibration frequency of the smart wearable device is adjusted to 100 vibrations per minute; if the analysis result shows a level 3 risk, the vibration frequency of the smart wearable device is adjusted to 200 vibrations per minute. This setting allows the user to clearly and intuitively understand the risk level without looking at the smart wearable device.

[0035] In this embodiment, the specific process of the display notification module flashing reminder is: if the analysis result shows a level 1 risk, the display frame of the smart wearable device flashes white light; if the analysis result shows a level 2 risk, the display frame of the smart wearable device flashes yellow light; if the analysis result shows a level 3 risk, the display frame of the smart wearable device flashes red light. This setting allows users to quickly and intuitively understand the risk level by viewing the display of the smart wearable device.

[0036] In this embodiment, the specific process of the early warning notification module deciding whether to send an early warning notification to the user's guardian according to the analysis results is as follows: assigning weight values ​​to the analysis results. If a first-level risk appears in the analysis results, a weight value V=1 corresponding to the analysis results is assigned; if a second-level risk appears in the analysis results, a weight value V=2 corresponding to the analysis results is assigned; if a third-level risk appears in the analysis results, a weight value V=3 corresponding to the analysis results is assigned; if multiple risks appear in the analysis results at the same time, the weight values ​​corresponding to each risk are added together; the sum of the weight values ​​V corresponding to all analysis results in the past three days is counted and marked as V total; V total is compared with the preset value v. If V total>v, the early warning notification module sends an early warning notification to the user's guardian. If V total≤v, the early warning notification module does not send an early warning notification to the user's guardian. This setting can remind the user's guardian in time when the user has multiple risks and does not control himself, so that the user's guardian can intervene in time.

[0037] In this embodiment, the warning notification is sent to the user's guardian in the form of a text message.

[0038] The present invention can not only monitor the user's health data in real time to prevent diseases, but also monitor and analyze the user's daily exercise and diet to prevent diseases. Among them, the present application uses the camera function of the smart wearable device to take pictures of the food that the user eats, and then analyzes the food's calories, processing procedures, and online health reviews from multiple angles to determine whether the food to be consumed is risky and promptly remind the user. In addition, for the user's running exercise, the user's posture, stride, and step frequency are analyzed during running, and the user is reminded in time when the user's running is abnormal, to prevent the negative effects of running from causing damage to the user's body. In general, the present application conducts a comprehensive monitoring of the user's health from the user's diet, exercise, and physiological indicators, effectively preventing the occurrence of diseases and improving the safety and health of the user.

[0039] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A disease prevention system based on AI technology and smart wearables, It is characterized in that It includes a smart wearable device and an AI analysis platform, wherein the smart wearable device includes a data acquisition module, a vibration notification module and a display notification module, and the AI ​​analysis platform includes a data receiving module, a data analysis module, an early warning notification module and a data import module; The data acquisition module is used to collect the user's running monitoring data, health monitoring data and food pictures taken by the user with a mobile phone before each meal, and send the collected data to the data receiving module; the data receiving module is used to receive all the data for integration, and send the integrated data to the data analysis module; the data import module is used to import preset data to the data analysis module; the data analysis module is used to receive all the data for analysis, and output the analysis results to the early warning notification module, the vibration notification module and the display notification module; the early warning notification module decides whether to send an early warning notification to the user's guardian based on the analysis results; the vibration notification module notifies the user through the vibration of the smart wearable device, and adjusts the vibration frequency of the smart wearable device based on the analysis results; the display notification module is used to display the analysis results to the user and provide a flashing reminder.

2. According to claim 1, a disease prevention system based on AI technology and smart wearables, It is characterized in that The specific process of the data analysis module analyzing the collected food pictures is as follows: Perform image recognition on food pictures to obtain the type of food identified, marked as A; Search all reviews of food type A on the public food safety platform through big data, screen the searched food reviews, and then perform text recognition on the screened food reviews, and plan the positive reviews into the first sequence, and the negative reviews into the second sequence; Mark the number of reviews in the first sequence as B1, and the number of reviews in the second sequence as B2, and calculate the proportion of negative reviews of the identified food b1=B2 / (B1+B2); Search all the production processes of type A food through big data, analyze the production process that matches the identified food, and then mark the total number of steps in the analyzed production process as C; Match the identified food with the food database, calculate the calories of the identified food and mark it as D; Compare the negative review ratio b1 of the identified food with the preset value b2, the average number of processes C with the preset value c, and the calorie D with the preset value d; If any of b1>b2, C>c, D>d appears in the comparison result, it means that the food that the current user is about to eat has a first-level risk; If any two of b1>b2, C>c, and D>d appear in the comparison results, it means that the food that the current user is about to eat has a secondary risk; If all items of b1>b2, C>c, and D>d appear in the comparison results, it means that the food that the current user is about to eat has a third-level risk; If all items of b1≤b2, C≤c, and D≤d appear in the comparison results, it means that the food that the current user is about to eat does not pose a risk.

3. According to claim 2, a disease prevention system based on AI technology and smart wearables, It is characterized in that The specific process of screening the searched food reviews is as follows: All food evaluations of food type A found through big data search on the food safety public platform are marked as Pi, i=1···n, where n is a positive integer; Mark the city where the review corresponding to Pi is posted as pi, and then mark the city where the user is located as puser; If pi and p are in the same city, pi is planned to the first cluster. If pi and p are not in the same city, proceed to the next step. If pi and p are from the same city in the same province, pi will be planned to the second cluster. If pi and p are not from the same city in the same province, proceed to the next step. If pi and p are cities in adjacent provinces, pi is planned to the third cluster. If pi and p are cities in different provinces, pi is planned to the fourth cluster. All the food evaluations Pi corresponding to each pi in the first cluster are selected, half of the food evaluations Pi corresponding to each pi in the second cluster are randomly selected, one third of the food evaluations Pi corresponding to each pi in the third cluster are randomly selected, and one quarter of the food evaluations Pi corresponding to each pi in the fourth cluster are randomly selected; Output the filtered food reviews.

4. According to claim 3, a disease prevention system based on AI technology and smart wearables, It is characterized in that The specific process of analyzing the production process matching the identified food is as follows: Mark all the production processes of type A food searched through big data as Wi, i=1···n, where n is a positive integer; Get the origin of the production process Wi, marked as wi; Identify wi and select wi that are in the same area as user p in the user's city, where the same area means that the straight-line distance between wi and user p in the user's city is within 100 kilometers; The finished food images of the production process Wi corresponding to the selected wi are selected from the public platform through big data; The screened images are compared with the food images collected by the data collection module for similarity, and the production process corresponding to the image with the highest similarity is used as the production process that matches the identified food.

5. According to claim 4, a disease prevention system based on AI technology and smart wearables, It is characterized in that The data acquisition module collects running monitoring data through the sensor component built into the smart wearable device, and the running monitoring data includes the user's running posture, stride, and step frequency data. The specific process of the data analysis module analyzing the collected running monitoring data is as follows: The collected user's running posture is marked as E, the running stride is marked as F, and the running frequency is marked as G; The posture E is compared with the preset posture e for similarity. If the similarity between E and e is less than the preset similarity, it means that the current user's running posture is abnormal; Compare the stride F with the preset stride f. If F>f, it means that the current user's running stride is abnormal. Compare the step frequency G with the preset step frequency g. If G>g, it means that the current user's running step frequency is abnormal. If any of the following items is found in the comparison result: abnormal running posture, abnormal running stride, or abnormal running cadence, it means that the current user's running exercise has a level 1 risk; If any two of the following are found in the comparison results: abnormal running posture, abnormal running stride, or abnormal running frequency, it means that the current user's running exercise has a secondary risk; If the comparison results show abnormal running posture, abnormal running stride, and abnormal running frequency, it means that the current user's running exercise has a third-level risk; If the comparison results do not show any of the abnormal running posture, abnormal running stride, or abnormal running frequency, it means that there is no risk in the current user's running exercise.

6. According to claim 5, a disease prevention system based on AI technology and smart wearables, It is characterized in that The health monitoring data includes the monitored user's heart rate data, blood pressure data and blood oxygen data. The specific process of the data analysis module analyzing the collected health monitoring data is as follows: The collected user's heart rate data is marked as X, blood pressure data is marked as Y, and blood oxygen data is marked as Z; Match X with a preset range value. If X is within the preset range value, it means the user's heart rate is normal. If X is not within the preset range value, it means the user's heart rate is abnormal. Match Y with the preset range value. If Y is within the preset range value, it means that the user's blood pressure is normal. If Y is not within the preset range value, it means that the user's blood pressure is abnormal. Match Z with the preset range value. If Z is within the preset range value, it means that the user's blood oxygen is normal. If Z is not within the preset range value, it means that the user's blood oxygen is abnormal. If any of abnormal heart rate, abnormal blood pressure, or abnormal blood oxygen is found in the analysis results, it means that the current user's physiological indicators are at level 1 risk; If any two of abnormal heart rate, abnormal blood pressure, and abnormal blood oxygen are found in the analysis results, it means that the current user's physiological indicators have secondary risks; If the analysis results show abnormal heart rate, abnormal blood pressure, and abnormal blood oxygen, it means that the current user's physiological indicators are at level 3 risk; If the analysis results do not show any of abnormal heart rate, abnormal blood pressure, or abnormal blood oxygen, it means that there is no risk in the current user's physiological indicators.

7. According to claim 6, a disease prevention system based on AI technology and smart wearables, It is characterized in that The specific process of the vibration notification module notifying the user is as follows: If the analysis results show a level 1 risk, the vibration frequency of the smart wearable device will be adjusted to 60 vibrations per minute; If a level 2 risk appears in the analysis results, the vibration frequency of the smart wearable device will be adjusted to 100 vibrations per minute; If a third-level risk appears in the analysis results, the vibration frequency of the smart wearable device will be adjusted to two hundred vibrations per minute.

8. According to claim 7, a disease prevention system based on AI technology and smart wearables, It is characterized in that The specific process of the display notification module performing the flashing reminder is as follows: If a level 1 risk appears in the analysis results, the border of the display screen of the smart wearable device will flash white light; If a secondary risk appears in the analysis results, the border of the display screen of the smart wearable device will flash yellow; If a third-level risk appears in the analysis results, the border of the display screen of the smart wearable device will flash red.

9. According to claim 8, a disease prevention system based on AI technology and smart wearables, It is characterized in that The specific process of the early warning notification module deciding whether to send an early warning notification to the user's guardian according to the analysis results is as follows: Assign weight values ​​to the analysis results. If a first-level risk appears in the analysis results, assign a weight value V=1 to the analysis results; if a second-level risk appears in the analysis results, assign a weight value V=2 to the analysis results; if a third-level risk appears in the analysis results, assign a weight value V=3 to the analysis results; if multiple risks appear in the analysis results at the same time, add up the weight values ​​corresponding to each risk; Count the sum of the weight values ​​V corresponding to all analysis results in the past three days, marked as Vtotal; Vtotal is compared with the preset value v. If Vtotal>v, the early warning notification module sends an early warning notification to the user's guardian. If Vtotal≤v, the early warning notification module does not send an early warning notification to the user's guardian.

10. A disease prevention system based on AI technology and smart wearables according to claim 9, It is characterized in that The warning notification is sent to the user's guardian in the form of a text message.

Images