Wearable health scheme recommendation system based on Internet of Things gateway

Abstract

The invention discloses a wearable health scheme recommendation system based on an Internet of Things gateway, and belongs to the field of wearable medical equipment. Comprising an MCU processor, a touch screen display module used for displaying human body characteristic data, a voice module used for interacting with wearable medical equipment, a communication module used for uploading obtained human body monitoring indexes to a big data cloud platform, a data obtaining module used for obtaining detailed vital sign data of a wearer, and an AI module, and the module is used for analyzing the health data recorded by the wearer, is located on a cloud network platform, accesses the AI model, analyzes the health index data acquired by the sensor, compares a historical medical case database, and formulates a set of personalized medical health scheme for the wearer by adopting a health scheme recommendation algorithm based on cases. The invention aims to become an innovative customized platform, so that wearers can integrate health tracking into personal health habits of the wearers, and a sexual health scheme can be timely and accurately provided.

Description

Technical Field

[0001] The present invention belongs to the field of wearable medical equipment and relates to a wearable health solution recommendation system based on an Internet of Things gateway. Background Art

[0002] Existing wearable medical devices already have similar intelligent monitoring devices, but they are lacking in functions. For example, they are disconnected from the medical system, or equipped with many auxiliary devices, and do not take into account the patient group. Many monitoring indicators (heart rate, electrocardiogram, blood oxygen, body temperature, blood pressure) require a large number of medical equipment to detect. In addition, in today's society where medical resources are extremely tight, medical staff may not be able to conduct consultations and medical guidance in the first place. As the workload of medical staff increases, it is also very likely that misjudgments and wrong medical diagnoses will occur, resulting in medical disputes. Summary of the invention

[0003] In order to overcome the shortcomings of the existing technology, the present invention provides a wearable health plan recommendation system based on the Internet of Things gateway, which is connected to the AI ​​model, matches the medical case database according to the various health indicator data obtained by the sensor, and finally provides a case-based health plan recommendation algorithm to formulate a set of personalized and accurate medical and health plans for the wearer to solve the medical problems existing in the existing technology.

[0004] The technical solution adopted by the present invention to solve its technical problem is: A wearable health solution recommendation system based on the Internet of Things gateway, comprising: MCU processors for controlling wearable medical systems that can recommend personalized health plans; A touch screen display module, used for displaying human body characteristic data, connected to the MCU processor; Voice module, used to connect to the MCU processor and interact with wearable medical devices; A communication module is used to connect to the MCU processor and upload the acquired human body monitoring indicators to the big data cloud platform; The data acquisition module is used to obtain detailed vital signs data of the wearer and is connected to the MCU processor; it integrates auxiliary detection equipment to obtain a number of monitoring indicators of the human body; The AI ​​module is used to analyze the health data recorded by the wearer. This module is located on the cloud network platform and connected to the AI ​​model. It analyzes the health indicator data obtained by the sensor, compares it with the historical medical case database, and uses a case-based health plan recommendation algorithm to develop a personalized medical and health plan for the wearer.

[0005] Furthermore, in the AI ​​module, the case-based health plan recommendation algorithm is a comparison of the similarities between two cases. ; For two given cases a and b, Sim ( x (a) , x (b) ) represents the similarity between them, Sim i (x i (a) , x i (b) ) represents the similarity between a and b on the ith attribute, i = 1, 2, · · · L, L is the number of samples in the entire data set, and the comprehensive similarity between two samples is determined by the similarity between each attribute of the two samples; according to the above formula, for two cases, each attribute is compared one by one to determine the similarity of the two cases on each attribute, and then these similarities are combined to determine the similarity between the two cases, so as to recommend the best medical plan based on the diagnosis plan of the historical case.

[0006] Preferably, in the case-based health plan recommendation algorithm, if the values ​​of a and b on the i-th attribute are the same, then the value is 1, and if they are different, the value is 0. If the value of an attribute is missing, the similarity is set to 0.5.

[0007] For a case to be recommended, first find the k most similar samples in the data set D according to the algorithm, and then make a decision using the majority voting method: use the classification that appears most frequently among the k most similar cases as the classification of the case to be predicted.

[0008] Or: for a case x to be predicted, take the similarity of each adjacent case in k similar samples as its weight. For each category in the classification set, add the weights of the consistent neighboring cases. The category with the largest value is taken as the final category of the case to be predicted. Then, based on a series of historical cases, the AI ​​model recommends the best medical and health plan.

[0009] Furthermore, the system also includes a communication server, which serves as a gateway connecting the AI ​​module and the communication module of the wearable medical device. The communication module of the wearable medical device uploads medical data in MQTT format to the communication server, and the communication server uploads it to the AI ​​cloud platform using TCP / IP protocol.

[0010] The data acquisition module includes a body temperature sensor for acquiring the wearer's temperature, and the body temperature sensor is connected to the MCU processor.

[0011] Preferably, the data acquisition module also includes a biopotential and bioimpedance module for acquiring the wearer's ECG signal, and the biopotential and bioimpedance module is connected to the MCU processor.

[0012] More preferably, the data acquisition module also includes a high-sensitivity pulse oximeter and a heart rate sensor, which are used to detect the wearer's pulse oxygen saturation (SpO2) and heart rate and observe the wearer's vital signs in real time. The pulse oximeter and the heart rate sensor are both connected to the MCU processor.

[0013] More preferably, the data acquisition module also includes a high-precision accelerometer and a gyroscope, which are used to obtain the wearer's usual movements and daily living habits and observe the wearer's movement trajectory. The high-precision accelerometer and gyroscope are both connected to the MCU processor.

[0014] The communication module uses WiFi / 5G / BLE and other transmission modes to upload the acquired human monitoring indicators (heart rate, electrocardiogram, blood oxygen, body temperature, blood pressure) to the communication server.

[0015] The beneficial effects of the present invention are mainly manifested in: becoming an innovative customization platform that enables wearers to integrate health tracking into their personal health habits and can provide a personalized health plan in a timely and accurate manner; wearers can ask questions at any time to fully meet the needs of daily health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a block diagram of the wearable health solution recommendation system based on the Internet of Things gateway.

[0017] Figure 2 Flowchart of a wearable medical system that can recommend personalized health plans. DETAILED DESCRIPTION

[0018] The present invention will be further described below in conjunction with the accompanying drawings.

[0019] Reference Figure 1 and Figure 2 , a wearable health solution recommendation system based on the Internet of Things gateway, including: MCU processors for controlling wearable medical systems that can recommend personalized health plans; A touch screen display module, used for displaying human body characteristic data, connected to the MCU processor; Voice module, used to connect to the MCU processor and interact with wearable medical devices; A communication module is used to connect to the MCU processor and upload the acquired human body monitoring indicators to the big data cloud platform; The data acquisition module is used to obtain detailed vital signs data of the wearer and is connected to the MCU processor; it integrates auxiliary detection equipment to obtain a number of monitoring indicators of the human body; The AI ​​module is used to analyze the health data recorded by the wearer. This module is located on the cloud network platform and connected to the AI ​​model. It analyzes the health indicator data obtained by the sensor, compares it with the historical medical case database, and uses a case-based health plan recommendation algorithm to develop a personalized medical and health plan for the wearer.

[0020] Furthermore, in the AI ​​module, the case-based health plan recommendation algorithm is a comparison of the similarities between two cases. ; For two given cases a and b, Sim ( x (a) , x (b) ) represents the similarity between them, Sim i (x i (a) , x i (b) ) represents the similarity between a and b on the ith attribute, i = 1, 2, · · · L, L is the number of samples in the entire data set, and the comprehensive similarity between two samples is determined by the similarity between each attribute of the two samples; according to the above formula, for two cases, each attribute is compared one by one to determine the similarity of the two cases on each attribute, and then these similarities are combined to determine the similarity between the two cases, so as to recommend the best medical plan based on the diagnosis plan of the historical case.

[0021] Furthermore, the system also includes a communication server, which serves as a gateway connecting the AI ​​module and the wearable medical device communication module. The communication module of the wearable medical device uploads medical data in MQTT format to the communication server, and the communication server uploads it to the AI ​​cloud platform using TCP / IP protocol.

[0022] The data acquisition module includes a body temperature sensor for acquiring the wearer's temperature, and the body temperature sensor is connected to the MCU processor. The data acquisition module also includes a biopotential and bioimpedance module for acquiring the wearer's ECG signal, and the biopotential and bioimpedance module is connected to the MCU processor. The data acquisition module also includes a high-sensitivity pulse oximeter and a heart rate sensor for detecting the wearer's pulse oxygen saturation (SpO2) and heart rate, and observing the wearer's vital signs in real time. The pulse oximeter and heart rate sensor are both connected to the MCU processor. The data acquisition module also includes a high-precision accelerometer and a gyroscope for acquiring the wearer's usual exercise and daily living habits, and observing the wearer's movement trajectory. The high-precision accelerometer and gyroscope are both connected to the MCU processor.

[0023] Figure 2The following is a flowchart of an embodiment of the present invention. The data acquisition module obtains various important human health data, and presents the data on the touch screen display module through the MCU processor. The touch screen display module can view and operate various data. The voice module can also broadcast health data. The acquired health data is uploaded to the communication server (gateway) through the communication module. The gateway stores and uploads it to the AI ​​module network big data platform, accesses the AI ​​model, analyzes the acquired human index data, and gives a health plan with targeted opinions. It is resent to the gateway, and then sent to each communication module by the gateway. After passing through the MCU processor, it is displayed on the touch screen and voice broadcast.

[0024] In the data acquisition module sensor wearing model of the embodiment of the present invention, various important biological characteristic signals of the human body such as ECG, heart rate, HRV, PPG, SpO 2 , blood pressure trends and breathing, through a highly integrated sensor detection device, worn on the human body, connected to the device, solving many complex problems of health data collection equipment. The collected data of various human health indicators are processed by the MCU processor. In an embodiment of the present invention, the data acquisition module integrates a body temperature sensor to obtain the wearer's body temperature. The data acquisition module also integrates a biopotential and bioimpedance module, which can detect the potential resistance encountered when the current passes through the tissues and organs in the human body, and obtain some information about human tissues and organs through resistance measurement technology, such as body composition, water content, blood flow, cardiopulmonary function and muscle health status. The embodiment of the present invention is used to obtain the ECG signal of the wearer. The data acquisition module also integrates a high-sensitivity pulse oximeter and a heart rate sensor to obtain the wearer's pulse oxygen saturation (SpO2) and heart rate data. The data acquisition module also integrates a high-precision accelerometer and a gyroscope. The accelerometer is used to measure the acceleration change of an object in three axes and can detect the linear acceleration of the object. By analyzing the output data of the accelerometer, the motion state of the object, such as acceleration, velocity and displacement, can be obtained. The gyroscope is used to measure the angular velocity of an object around three axes. It can detect the rotation state of an object and derive the rotational angular velocity and direction of the object. Through high-precision accelerometers and gyroscopes, the wearer's daily exercise and daily living habits can be obtained and the wearer's movement trajectory can be observed. Accurate exercise suggestions and health guidance can be given through AI analysis.

[0025] The wearable health solution recommendation system based on the Internet of Things gateway of this embodiment integrates a data acquisition chip, an MCU processor, a communication module, a voice module and a touch screen display module. The voice module is connected to the MCU processor to realize human-computer interaction. When the patient is inconvenient to operate, he can give voice commands, and the voice module recognizes the voice to realize voice operation. When the MCU receives health guidance, the voice module can realize the real-time broadcast reminder function. The dial is a touch screen display module, which can be adjusted to view various human body data, view health guidance, and send instructions.

[0026] The communication module of this embodiment integrates multiple communication transmission methods such as WiFi / 5G / BLE, and uploads them to the communication server (gateway). The embodiment of the present invention adopts the BLE low-power Bluetooth transmission mode to upload the original ECG data to the communication server (gateway). The communication server of the wearable medical system architecture that can recommend personalized health plans serves as an intermediate station to bridge the wearable medical device and the cloud AI module. The data upload adopts the mqtt format and TCP / IP protocol. It has powerful information processing capabilities and can receive data information from multiple wearable medical detection devices, receive a large amount of medical information and store and format it. The embodiment of the present invention uploads the heart rate data to the network cloud platform.

[0027] The AI ​​module of this embodiment, after the communication server (gateway) sends the data to the network cloud platform, is connected to the AI ​​big model. First, the health indicator data acquired by the sensor is extracted with characteristic values ​​to form the current case, and then compared with the historical medical case database to provide a case-based health plan recommendation algorithm to formulate a personalized medical and health plan for the wearer. The core of this optimization algorithm is to compare the similarity between two cases. ; For two given cases a and b, Sim ( x (a) , x (b) ) represents the similarity between them, Sim i (x i (a) , x i (b) ) represents the similarity between a and b on the i-th attribute, i = 1, 2, · · · L, L is the number of samples in the entire data set, and the comprehensive similarity between two samples is determined by the similarity between each attribute of the two samples; according to the above formula, for two cases, each attribute is compared one by one to determine the similarity between the two cases on each attribute. To compare the similarity, first map the values ​​of a and b on the i-th attribute to three values: if the values ​​of a and b on the i-th attribute are the same, it is 1, if they are different, it is 0, and if the value of an attribute is missing, the similarity is set to 0.5; If two cases are consistent in a certain data feature value, the similarity of the two cases increases; if two cases are inconsistent in a certain data feature value, the similarity of the two cases will not increase due to this feature value; in order to make the similarity measurement more accurate, if two cases have a missing value in a certain data feature value, let this data feature value contribute a relatively compromised similarity to the two cases.

[0028] According to the above similarity measurement method, for a case to be recommended, first find the k samples with the highest similarity in the data set D according to this method, and then make a decision. At this time, how to decide the target category becomes another key issue of the kNN classifier. A simple method is the majority voting method: use the category with the most occurrences in the k most similar cases as the category of the case to be predicted. Although this method is simple and easy to implement, it ignores the influence of case similarity when making decisions. The higher the similarity of the neighbor with the case to be predicted, the higher the influence on the decision process should be; and the lower the similarity of the neighbor with the case to be predicted, the weaker the influence on the decision process should be. For a case x to be predicted, the similarity of each adjacent case in the k similar samples is used as its weight, instead of simply using the simple majority voting method to make a decision on the target category. For each category in the classification set, add the weights of the neighboring cases that are consistent with it, and the category with the largest value is used as the final category of the case to be predicted. Then, based on a series of historical cases, the AI ​​model recommends the best medical and health plan.

[0029] In the embodiment of the present invention, the AI ​​module analyzes the health warning prompts based on the heart rate and blood oxygen data obtained by the sensor and the health plan recommendation algorithm based on the case. The AI ​​module obtains the number of exercise steps from the sensor and analyzes the health guidance plan based on the health plan recommendation algorithm based on the case.

[0030] The contents described in the embodiments of this specification are merely enumerations of implementation forms of the inventive concept and are for illustrative purposes only. The protection scope of the present invention should not be considered to be limited to the specific forms described in this embodiment, and the protection scope of the present invention also extends to equivalent technical means that can be thought of by ordinary technicians in this field based on the inventive concept.

Claims

1. A wearable health solution recommendation system based on the Internet of Things gateway, characterized in that: The recommendation system comprises: MCU processors for controlling wearable medical systems that can recommend personalized health plans; A touch screen display module, used for displaying human body characteristic data, connected to the MCU processor; Voice module, used to connect to the MCU processor and interact with wearable medical devices; A communication module is used to connect to the MCU processor and upload the acquired human body monitoring indicators to the big data cloud platform; The data acquisition module is used to obtain detailed vital signs data of the wearer and is connected to the MCU processor; it integrates auxiliary detection equipment to obtain a number of monitoring indicators of the human body; The AI ​​module is used to analyze the health data recorded by the wearer. This module is located on the cloud network platform and connected to the AI ​​model. It analyzes the health indicator data obtained by the sensor, compares it with the historical medical case database, and uses a case-based health plan recommendation algorithm to develop a personalized medical and health plan for the wearer.

2. A wearable health solution recommendation system based on the Internet of Things gateway as claimed in claim 1, characterized in that: In the AI ​​module, the case-based health plan recommendation algorithm compares the similarities between two cases. ； For two given cases a and b, Sim ( x (a) , x (b) ) represents the similarity between them, Sim i ( x i (a) , x i (b) ) represents the similarity between a and b on the ith attribute, i = 1, 2, · · · L, L is the number of samples in the entire data set, and the comprehensive similarity between two samples is determined by the similarity between each attribute of the two samples; according to the above formula, for two cases, each attribute is compared one by one to determine the similarity of the two cases on each attribute, and then these similarities are combined to determine the similarity between the two cases, so as to recommend the best medical plan based on the diagnosis plan of the historical case.

3. A wearable health solution recommendation system based on the Internet of Things gateway as claimed in claim 2, characterized in that: In the case-based health plan recommendation algorithm, if the values ​​of a and b on the i-th attribute are the same, then the value is 1, and if they are different, the value is 0. If the value of an attribute is missing, the similarity is set to 0.

5.

4. A wearable health solution recommendation system based on the Internet of Things gateway as claimed in claim 3, characterized in that: For a case to be recommended, first find the k most similar samples in the data set D according to the algorithm, and then make a decision using the majority voting method: use the classification that appears most frequently among the k most similar cases as the classification of the case to be predicted.

5. A wearable health solution recommendation system based on the Internet of Things gateway as claimed in claim 3, characterized in that: For a case x to be predicted, the similarity of each adjacent case in k similar samples is used as its weight. For each category in the classification set, the weights of the consistent neighboring cases are added. The category with the largest value is taken as the final category of the case to be predicted. Based on a series of historical cases, the AI ​​model recommends the best medical and health plan.

6. A wearable health solution recommendation system based on an Internet of Things gateway as described in any one of claims 1 to 5, characterized in that: The system also includes a communication server, which serves as a gateway connecting the AI ​​module and the communication module of the wearable medical device. The communication module of the wearable medical device uploads medical data in MQTT format to the communication server, and the communication server uploads it to the AI ​​cloud platform using TCP / IP protocol.

7. A wearable health solution recommendation system based on an Internet of Things gateway as described in any one of claims 1 to 5, characterized in that: The data acquisition module includes a body temperature sensor for acquiring the wearer's temperature, and the body temperature sensor is connected to the MCU processor.

8. A wearable health solution recommendation system based on the Internet of Things gateway as claimed in claim 7, characterized in that: The data acquisition module also includes a biopotential and bioimpedance module for acquiring the ECG signal of the wearer, and the biopotential and bioimpedance module is connected to the MCU processor.

9. A wearable health solution recommendation system based on the Internet of Things gateway as claimed in claim 8, characterized in that: The data acquisition module also includes a high-sensitivity pulse oximeter and a heart rate sensor, which are used to detect the wearer's pulse oxygen saturation and heart rate and observe the wearer's vital signs in real time. The pulse oximeter and the heart rate sensor are both connected to the MCU processor; The data acquisition module also includes a high-precision accelerometer and a gyroscope, which are used to obtain the wearer's usual movements and daily living habits and observe the wearer's movement trajectory. The high-precision accelerometer and gyroscope are both connected to the MCU processor.

10. A wearable health solution recommendation system based on an Internet of Things gateway as claimed in any one of claims 1 to 5, characterized in that: The communication module is in WiFi / 5G / BLE transmission mode, and the acquired human body monitoring indicators are uploaded to the communication server.

Images