Home postoperative nursing training risk identification platform based on internet of things

Abstract

The application discloses a home postoperative nursing training risk identification platform based on Internet of Things and relates to the technical field of rehabilitation nursing.The platform comprises a home postoperative nursing bed, a motion sensing module, a historical motion sensing data set, an Internet of Things-based transmission of the historical motion sensing data set to a risk identification cloud platform, a training risk identification fusion model, embedding of the risk identification fusion model in the risk identification cloud platform, a smart early warning module for risk identification of real-time motion sensing data sets and output of a first risk index, and a first reminder information generated based on the smart early warning module if the first risk index is greater than a preset risk index.The technical problem of the existing home postoperative nursing training risk identification, i.e., difficulty in real-time, accurate and comprehensive monitoring of the training rehabilitation state and thus insufficient accuracy and reliability of the training risk identification and inability to provide instant early warning, is solved, and the technical effect of improving the safety of postoperative nursing training and the nursing rehabilitation effect is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rehabilitation nursing, and particularly relates to a home postoperative nursing training risk identification platform based on Internet of Things. BACKGROUND

[0002] With the continuous progress of medical technology and the intensification of population aging trend, home postoperative nursing has become an important medical care mode, however, postoperative nursing training in the home environment faces many challenges, the most critical of which is how to timely and accurately identify the risks that patients may encounter in the rehabilitation process to ensure the safety and rehabilitation effect of patients, traditional postoperative nursing often relies on the self-management of patients and the care of family members, but in the actual postoperative nursing training, due to the lack of professional knowledge and insufficient monitoring means, it is difficult to achieve real-time and comprehensive monitoring of the rehabilitation state of patients, thereby increasing the risk in the postoperative rehabilitation training process of patients, and the inability to identify and warn risks leads to accidents.

[0003] Therefore, in the present home postoperative nursing training risk identification related technology, there is a technical problem that it is difficult to monitor the training rehabilitation state in real time, accurately and comprehensively, thereby leading to insufficient accuracy and reliability of training risk identification, and inability to timely early warning. SUMMARY

[0004] The present application provides a home postoperative nursing training risk identification platform based on Internet of Things, adopts real-time sensing, data transmission, model training and early warning feedback technical means, solves the technical problem that the existing home postoperative nursing training risk identification is difficult to monitor the training rehabilitation state in real time, accurately and comprehensively, thereby leading to insufficient accuracy and reliability of training risk identification, and inability to timely early warning, and achieves the technical effect of improving the safety of postoperative nursing training and the nursing rehabilitation effect.

[0005] The application provides a home postoperative care training risk identification platform based on the Internet of Things, which comprises a home postoperative care bed connection module for connecting a home postoperative care bed, wherein the home postoperative care bed is provided with a motion sensing module, the motion sensing module performs real-time sensing on the bed exercise of a user and acquires a historical motion sensing data set; a historical motion sensing data set transmission module for transmitting the historical motion sensing data set to a risk identification cloud platform based on the Internet of Things, wherein the risk identification cloud platform is connected with a home postoperative care bed management system; a risk identification fusion model training module for training a risk identification fusion model, embedding the risk identification fusion model in the risk identification cloud platform, wherein the risk identification fusion model is acquired by training data, and the training data comprises motion sensing data samples of authorized users and identification information training acquisition of identifying risk levels; a first risk index output module for downloading the risk identification fusion model to an intelligent early warning module of the home postoperative care bed, wherein the intelligent early warning module performs risk identification on real-time motion sensing data sets and outputs a first risk index; and a first reminder information generation module for generating first reminder information based on the intelligent early warning module if the first risk index is greater than a preset risk index.

[0006] In a possible implementation, the home postoperative care bed is further provided with a physiological sensing module; the physiological sensing module performs real-time sensing on the bed exercise of the user and acquires a historical physiological sensing data set; the physiological sensing module and the motion sensing module are synchronized, the historical physiological sensing data set and the historical motion sensing data set are subjected to time sequence synchronization processing, and the synchronized historical physiological sensing data set and historical motion sensing data set are output; and the synchronized historical physiological sensing data set and historical motion sensing data set are transmitted to the risk identification cloud platform based on the Internet of Things.

[0007] In a possible implementation, the historical motion sensing data set transmission module further performs the following processing: the risk identification cloud platform respectively performs abnormality identification on the synchronized historical physiological sensing data set and historical motion sensing data set, and outputs an abnormal physiological sensing data set and an abnormal motion sensing data set; the time sequences of the abnormal physiological sensing data set and the abnormal motion sensing data set are aligned, and a synchronized abnormal sensing data group is extracted; and the training data for training the risk identification fusion model is processed according to the synchronized abnormal sensing data group.

[0008] In a possible implementation, the historical motion sensing data set transmission module further performs the following processing: an Internet of Things communication technology is adopted to establish a first transmission channel between the motion sensing module and a home gateway and a second transmission channel between the home gateway and the risk identification cloud platform; and the historical motion sensing data set is subjected to encrypted transmission according to the first transmission channel and the second transmission channel.

[0009] In a possible implementation, the motion sensing module includes an acceleration sensor, a displacement sensor, and a pressure sensor; the user is sensed in real time according to the acceleration sensor, and motion intensity data and frequency data are acquired; the user is sensed in real time according to the displacement sensor, and user position data and motion displacement data are acquired; the user is sensed in real time according to the pressure sensor, and user body position data and pressure distribution data are acquired; the motion intensity data, the frequency data, the user position data, the motion displacement data, the user body position data, and the pressure distribution data are used as parameter items for acquiring a historical motion sensing data set.

[0010] In a possible implementation, the physiological sensing module includes a heart rate sensor, a blood oxygen sensor, and a body temperature sensor, the heart rate sensor is configured to detect heart rate changes of the user, the blood oxygen sensor is configured to detect blood oxygen levels of the user, and the body temperature sensor is configured to detect body temperature changes of the user; the heart rate change data, the blood oxygen level data, and the body temperature change data are used as parameter items for acquiring a historical physiological sensing data set.

[0011] In a possible implementation, the risk identification fusion model training module further performs the following processing: collecting electronic medical record information and user basic information of the user; performing feature extraction on the electronic medical record information and the user basic information to acquire feature keywords; performing user analogy in the risk identification cloud platform according to a word frequency feature, and outputting an analogy permission user group; and generating the motion sensing data sample according to the analogy permission user group.

[0012] In a possible implementation, the risk identification fusion model training module further performs the following processing: collecting a motion sensing data health sample of the user; performing model optimization training by taking the motion sensing data health sample as comparison data of the motion sensing data sample, and outputting the risk identification fusion model.

[0013] The home postoperative care training risk identification platform based on the Internet of Things provided in the application, a home postoperative care bed connection module for connecting a home postoperative care bed, the home postoperative care bed is provided with a motion sensing module, the motion sensing module senses the bed movement of a user in real time and obtains a historical motion sensing data set; a historical motion sensing data set transmission module for transmitting the historical motion sensing data set to a risk identification cloud platform based on the Internet of Things, wherein the risk identification cloud platform is connected with a home postoperative care bed management system; a risk identification fusion model training module for training a risk identification fusion model, embedding the risk identification fusion model in the risk identification cloud platform, the risk identification fusion model is obtained by training data, and the training data includes motion sensing data samples of authorized users and identification information training acquisition of identifying risk levels; a first risk index output module for downloading the risk identification fusion model to an intelligent early warning module of the home postoperative care bed, the intelligent early warning module performs risk identification on a real-time motion sensing data set and outputs a first risk index; a first reminder information generation module for generating first reminder information based on the intelligent early warning module if the first risk index is greater than a preset risk index. The technical problems that the existing home postoperative care training risk identification cannot monitor the training rehabilitation state in real time, accurately and comprehensively, and thus leads to insufficient accuracy and reliability of training risk identification and inability to immediately warn are solved, and the technical effects of improving the safety of postoperative care training and the nursing rehabilitation effect are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. In the present application, a flowchart is used to illustrate the operations performed by the platform according to the embodiments of the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0015] Figure 1 The structure schematic diagram of the home postoperative care training risk identification platform based on the Internet of Things provided for the embodiments of the present application;

[0016] Figure 2 The execution process schematic diagram of the historical motion sensing data set transmission module in the home postoperative care training risk identification platform based on the Internet of Things provided for the embodiments of the present application.

[0017] Explanation of reference signs: home postoperative care bed connection module 10, historical motion sensing data set transmission module 20, risk identification fusion model training module 30, first risk index output module 40, first reminder information generation module 50. DETAILED DESCRIPTION

[0018] The above description is only a summary of the technical solutions of the present application. In order to make the technical means of the present application more clearly understood, the present application can be implemented according to the content of the description, and in order to make the above and other purposes, characteristics and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0019] In order to make the purposes, technical solutions and advantages of the present application more clear, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0020] In the following description, "some embodiments" are related to a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subset of all possible embodiments, and can be combined with each other without conflict. The term "first\second" is only to distinguish similar objects, and does not represent the specific order of the objects. The terms "include" and "have" and any variations, are intended to cover non-exclusive inclusion, for example, a process, platform, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0021] The embodiments of the present application provide a home postoperative care training risk identification platform based on Internet of Things, as shown in Figure 1 The platform includes:

[0022] The home postoperative care bed connection module 10 is used for connecting the home postoperative care bed, and the home postoperative care bed is provided with a motion sensing module, which senses the bed motion of the user in real time and obtains a historical motion sensing data set. The home postoperative care bed is a medical device specially designed for patients who have undergone surgery or need long-term home care, aiming to provide a safe, comfortable and convenient care environment and promote the recovery of patients. The home postoperative care bed is provided with a motion sensing module, specifically, the motion sensing module is a device module that can detect and record the motion state of the human body, which is usually integrated in the interior or periphery of the home postoperative care bed, and realizes real-time monitoring and sensing of the motion state of the user through specific sensor devices, which may include acceleration sensors, gyroscopes, pressure sensors, etc. Real-time capture of user motion, direction change and body pressure distribution on the bed surface, such as turning over, lifting legs, lifting hands, etc. Among them, real-time sensing refers to the motion sensing module that can collect user motion data in real time and continuously. These motion data are converted into electrical signals by sensors and displayed to medical staff or the user in digital or graphical form. Real-time sensing helps to discover abnormal conditions in user motion in a timely manner, such as sudden violent movements or long periods of inactivity. These user motion data collected by the motion sensing module over a period of time constitute a historical motion sensing data set, which is arranged, stored and classified to form a historical record of user motion, which can be used for recovery assessment. By comparing motion data at different time points, the user's recovery progress can be assessed. By analyzing the pressure distribution information in the data set, it can be determined whether the user is in a comfortable lying state, so as to adjust the hardness or angle of the mattress and other parameters. For postoperative patients, long-term bed rest may lead to complications such as pressure ulcers and deep vein thrombosis. By monitoring the user's motion in real time, the motion sensing module can discover potential safety risks in a timely manner and issue warning signals.

[0023] The historical motion sensing data set transmission module 20 is configured to transmit the historical motion sensing data set to a risk identification cloud platform based on the Internet of Things, wherein the risk identification cloud platform is connected to the home postoperative care bed management system. The home postoperative care bed is connected to the risk identification cloud platform in communication through the Internet of Things technology, so that real-time transmission and processing of data are realized. The historical motion sensing data set obtained through the Internet of Things technology is transmitted to the risk identification cloud platform by using Bluetooth, Wi-Fi, ZigBee and other technologies. During the transmission process, the data is subjected to encryption processing to ensure the security of data transmission and the protection of privacy. After the risk identification cloud platform receives the data, the data is stored, processed and analyzed. The cloud platform usually has strong data processing capability and can quickly and accurately process a large amount of data and extract useful information. Specifically, a stable connection is established between the risk identification cloud platform and the home postoperative care bed management system, such as direct network connection or connection through a third-party platform (middleware). Through the connection, the cloud platform can obtain the data uploaded by the care bed in real time and provide risk identification results and early warning information to the management system. The main functions of the risk identification cloud platform include data processing, risk identification, early warning generation and result feedback, that is, the risk identification cloud platform deeply mines and analyzes the received historical motion sensing data set, uses advanced algorithms and models to identify potential risks in the rehabilitation process of the user, and generates early warning information immediately if a risk is identified. Early warning is notified through the management system or a mobile phone APP and the like.

[0024] The risk identification fusion model training module 30 is configured to train a risk identification fusion model, embed the risk identification fusion model in the risk identification cloud platform, and obtain the risk identification fusion model through training data. The training data includes motion sensor data samples of authorized users and identification information identifying the risk level. The motion sensor data samples of authorized users are derived from the motion sensor module of the home postoperative care bed. These data samples contain various motion information of the user on the bed, such as acceleration, direction change, pressure distribution, etc., which can reflect the motion state and habit of the user. Strict permission control and anonymization processing are required to protect the privacy of the user. The identification information identifying the risk level refers to the identification information corresponding to the motion sensor data of the authorized user, which is used to identify the risk level of each data sample. For example, it is derived from the assessment of the nursing staff or the self-feedback of the user. The identification information should be accurate, reliable, and able to objectively reflect the risk situation in the data sample. Specifically, a risk identification model is constructed based on a machine learning model (such as a neural network, a random forest, a support vector machine, etc.). Before training the risk identification model, the collected data is preprocessed, including data cleaning (removing noise, outliers, etc.), data normalization (making different features have the same scale), and data augmentation (increasing the diversity and quantity of data samples), etc. Then, the preprocessed data is input into the risk identification fusion model as training data. Through iterative optimization algorithms (such as gradient descent method), the parameters of the model are continuously adjusted so that the model can accurately identify the risk information in the data. The risk identification fusion model is trained, and then the trained risk identification fusion model is embedded in the risk identification cloud platform, which may include serialization, packaging, etc. of the model, so that users can conveniently call the model for risk identification.

[0025] The first risk indicator output module 40 is used to download the risk identification fusion model to the intelligent early warning module of the home postoperative care bed. The intelligent early warning module performs risk identification on the real-time motion sensing data set and outputs the first risk indicator. Through a secure file transfer protocol (such as FTP, SFTP, etc.), a trained risk identification fusion model is selected from the risk identification cloud platform and downloaded to the intelligent early warning module of the home postoperative care bed, ensuring the safety and integrity of the model file during transmission. Then, the downloaded model file is deployed to the local storage of the intelligent early warning module and necessary configuration and initialization operations are performed, including setting the input and output interfaces of the risk identification fusion model, loading model parameters, etc., to ensure that the model can operate normally and receive real-time motion sensing data. When the motion sensing module on the home postoperative care bed collects user motion data in real time and sends these data to the intelligent early warning module, the intelligent early warning module performs preprocessing operations on the real-time motion sensing data set to ensure the quality of the data input into the risk identification fusion model. The risk identification fusion model immediately processes and analyzes these data, identifies the risk information therein, and outputs the first risk indicator according to the preset risk assessment standard and threshold. The first risk indicator, usually a numerical value or a classification label, represents the risk level or category corresponding to the current motion sensing data set. For example, it can be a numerical value between 0 and 1, where 0 represents no risk or low risk, and 1 represents high risk. Or it can be a classification label such as "normal", "slight risk", "moderate risk", "high risk", etc.

[0026] The first reminder information generation module 50 is used to generate first reminder information based on the intelligent early warning module if the first risk indicator is greater than the preset risk indicator. If the first risk indicator is greater than the preset risk indicator, indicating that the current risk level has exceeded the safe range, the intelligent early warning module will generate first reminder information according to the preset rules, such as timely understanding of the patient's rehabilitation status, usually including risk level ("high risk", "emergency situation", etc.), occurrence time (risk identification timestamp), risk description (such as "long time without movement may cause pressure ulcer risk"), and recommended measures (such as "please adjust the patient's body position immediately and increase the number of turning over"). The intelligent early warning module performs early warning notification through wireless network transmission, Bluetooth communication, mobile phone APP push, etc., realizing real-time monitoring and risk assessment of the user's motion state and improving the accuracy and efficiency of nursing.

[0027] The home postoperative care training risk identification platform based on the Internet of Things according to the embodiment of the present application is used to solve the technical problems that the existing home postoperative care training risk identification cannot monitor the training rehabilitation state in real time, accurately and comprehensively, and thus cannot achieve the accuracy and reliability of training risk identification and cannot achieve instant early warning, and achieves the technical effect of improving the safety of postoperative care training and the nursing rehabilitation effect. The home postoperative care training risk identification platform based on the Internet of Things comprises a home postoperative care bed connection module 10, a historical motion sensing data set transmission module 20, a risk identification fusion model training module 30, a first risk index output module 40 and a first reminder information generation module 50.

[0028] Next, the specific configuration of the home postoperative care bed connection module 10 will be described in detail. As shown in Figure 2 The home postoperative care bed connection module 10 can further comprise a physiological sensing module. The physiological sensing module is mainly used for real-time monitoring of physiological parameters of the user, which can include body temperature, heart rate, blood pressure, respiratory rate, blood oxygen saturation and the like. The physiological sensing module collects physiological data of the user in a non-invasive or minimally invasive manner (such as wearable devices, patch sensors, etc.) and converts it into a processable digital signal. Further, the physiological sensing module is used to sense the user's bed movement in real time to obtain a historical physiological sensing data set. The physiological sensing module is used to sense the user's bed movement in real time, i.e., to monitor the physiological parameter changes (such as heart rate acceleration, respiratory acceleration, etc.) caused by movement, and thus to obtain a historical physiological sensing data set. For example, when the user turns over or moves, these actions can cause changes in physiological parameters such as heart rate and respiration, which are captured by the physiological sensing module.

[0029] Further, the physiological sensing module and the motion sensing module are synchronized, and the historical physiological sensing data set and the historical motion sensing data set are time-synchronized, and the synchronized historical physiological sensing data set and historical motion sensing data set are output. The physiological sensing module and the motion sensing module are synchronized, and the historical physiological sensing data set and the historical motion sensing data set are time-synchronized, to ensure the correlation and accuracy of physiological data and motion data, for example, by time stamp matching, data interpolation and the like. The physiological data and the corresponding motion data at each moment can be one-to-one corresponding, and the historical physiological sensing data set and the historical motion sensing data set after time-synchronized processing can more accurately reflect the health status and motion state of the user.

[0030] Further comprising, transmitting the synchronized historical physiological sensor data set and the historical motion sensor data set to a risk identification cloud platform based on the Internet of Things. The use of Internet of Things technology enables the home postoperative care bed to realize remote data transmission and real-time monitoring. By uploading the synchronized historical physiological sensor data set and the historical motion sensor data set to the risk identification cloud platform, the nursing staff or the user can access these data at any time and anywhere, and understand the rehabilitation training status and real-time motion situation.

[0031] In the following, the specific configuration of the historical motion sensor data set transmission module 20 will be described in detail. The historical motion sensor data set transmission module 20 can further comprise: the risk identification cloud platform performs abnormal identification on the synchronized historical physiological sensor data set and the historical motion sensor data set respectively, and outputs an abnormal physiological sensor data set and an abnormal motion sensor data set. The risk identification cloud platform first receives the synchronized historical physiological sensor data set and the historical motion sensor data set, applies a preset abnormal detection algorithm (such as pattern recognition, time series analysis, etc.), and performs abnormal identification on the historical physiological sensor data set and the historical motion sensor data set respectively. Abnormal records that are significantly different from normal patterns are identified. For the physiological sensor data set, the abnormality may be manifested as a sudden change in physiological parameters, a value exceeding the normal range, a persistent abnormal trend, etc. For the motion sensor data set, the abnormality may be manifested as a sudden decrease in user activity, an abnormal pattern of body position change, an abnormal motion trajectory, etc. Then, the abnormal physiological sensor data set and the abnormal motion sensor data set are output.

[0032] Further comprising, aligning the time sequence of the abnormal physiological sensor data set and the abnormal motion sensor data set, and extracting a synchronized abnormal sensor data group. The risk identification cloud platform uses methods such as timestamp matching and interpolation to align the time sequence of the abnormal physiological sensor data set and the abnormal motion sensor data set, eliminates the slight differences in timestamps between the two sensor data sets, and ensures that the abnormal records in the two data sets can be one-to-one corresponding in time. The aligned data will form a synchronized abnormal sensor data group, each data group containing physiological parameters and motion parameters identified as abnormal at the same time point.

[0033] Further comprising, processing training data for training the risk identification fusion model according to the synchronized abnormal sensor data group. According to the synchronized abnormal sensor data group, matching data is selected from the training data for the risk identification fusion model, and then the matching successful data is used as new training data for the training of the risk identification fusion model.

[0034] In the following, the specific configuration of the historical motion sensing data set transmission module 20 will be described in detail. The historical motion sensing data set transmission module 20 can further include: establishing a first transmission channel between the motion sensing module and the home gateway, and a second transmission channel between the home gateway and the risk identification cloud platform by using Internet of Things communication technology; and performing encrypted transmission of the historical motion sensing data set according to the first transmission channel and the second transmission channel. The motion sensing module (including acceleration sensor, displacement sensor, pressure sensor, etc.) is connected with the home gateway through Internet of Things communication technology (such as wired communication such as Ethernet, wireless communication such as Wi-Fi, Bluetooth, etc.). The acceleration, displacement, pressure, etc. data collected by the motion sensing module in real time is transmitted to the home gateway after being encrypted through the first transmission channel; the home gateway is connected with the risk identification cloud platform through the Internet or a specific communication protocol (such as cellular communication 4G / 5G, VPN tunnel, etc.). The encrypted data received from the motion sensing module is further encrypted and transmitted to the risk identification cloud platform through the second transmission channel for processing and analysis; wherein, in order to ensure the security of data transmission, symmetric encryption (such as AES), asymmetric encryption (such as RSA) or hybrid encryption technology can be used, and the transmitted data is encrypted using the SSL / TLS protocol to ensure the confidentiality, integrity and identity authentication of the data in the transmission process. The encrypted data packet is transmitted through the first transmission channel and the second transmission channel, and after reaching the receiving end (home gateway or risk identification cloud platform), it is decrypted for processing.

[0035] In the following, the specific configuration of the home postoperative care bed connection module 10 will be described in detail. The home postoperative care bed connection module 10 can further include: the motion sensing module includes an acceleration sensor, a displacement sensor and a pressure sensor; real-time sensing of the user according to the acceleration sensor to obtain motion intensity data and frequency data; real-time sensing of the user according to the displacement sensor to obtain user position data and motion displacement data; real-time sensing of the user according to the pressure sensor to obtain user body position data and pressure distribution data. The acceleration sensor can measure the acceleration change of the user in motion in real time, and then evaluate the motion intensity data and frequency data of the user; the displacement sensor is used to monitor the position data and motion displacement data of the user in real time, i.e. the moving track and distance of the user on the home postoperative care bed, and then obtain the motion pattern of the user; the pressure sensor is used to measure the pressure distribution and change between the user and the contact surface of the care bed, and to infer the body position (such as sitting posture, lying posture, standing posture, etc.) and posture change of the user, and to obtain the user body position data and pressure distribution data (to evaluate the comfort, support demand and potential health risks of the user, such as pain or injury caused by pressure points), which helps to understand the body position and comfort of the user.

[0036] Further, based on the motion intensity data, frequency data, user position data, motion displacement data, user body position data, and pressure distribution data as parameter items for obtaining the historical motion sensing data set. The motion intensity data, frequency data, user position data, motion displacement data, user body position data, and pressure distribution data and corresponding time stamps as parameter items for the historical motion sensing data set provide important basis for subsequent data analysis, health assessment, and exercise guidance.

[0037] In the following, the specific configuration of the historical motion sensing data set transmission module 20 will be described in detail. The historical motion sensing data set transmission module 20 can further include that the physiological sensing module includes a heart rate sensor, a blood oxygen sensor, and a body temperature sensor, the heart rate sensor is used to detect the heart rate change of the user, the blood oxygen sensor is used to detect the blood oxygen level of the user, and the body temperature sensor is used to detect the body temperature change of the user. The heart rate sensor is used to continuously monitor the number of heartbeats and rhythm of the user, i.e. the heart rate change, through the heart rate sensor, the heart rate data of the user is obtained in real time, which reflects the health status of the heart and the load condition during exercise; the blood oxygen sensor is used to measure the content of oxygen in the blood of the user, i.e. the blood oxygen saturation, the blood oxygen sensor calculates the blood oxygen level of the user by detecting the proportion of oxygenated hemoglobin and reduced hemoglobin in the blood, which is used to evaluate the function of the respiratory system and judge whether there is hypoxia condition; the body temperature sensor is used to detect the body temperature change of the user, i.e. the temperature inside the body, which provides information about whether the body is in the normal temperature range.

[0038] Further, the heart rate change data, blood oxygen level data, and body temperature change data are included as parameter items for obtaining the historical physiological sensing data set. The heart rate change data, blood oxygen level data, and body temperature change data are included as parameter items for obtaining the historical physiological sensing data set, specifically, the heart rate change data records the heart rate values and their change trends of the user in different time periods, which is used to analyze the heart health status, exercise load, and possible abnormal heart rate events (such as tachycardia, bradycardia, etc.) of the user; the blood oxygen level data records the blood oxygen saturation values of the user in different time periods, which is used to evaluate the respiratory system function of the user, judge whether there is hypoxia condition, and monitor the progress of certain diseases (such as chronic obstructive pulmonary disease, sleep apnea syndrome, etc.); the body temperature change data records the body temperature values and their change trends of the user in different time periods, which helps to identify temperature abnormalities (such as fever), so that medical measures can be taken in time, and is used to evaluate the physiological state and environmental adaptability of the user.

[0039] Below, the specific configuration of the risk identification fusion model training module 30 will be described in detail. The risk identification fusion model training module 30 can further include: collecting electronic medical record information and user basic information of the user. Collecting electronic medical record information of the user refers to collecting medical records of the user, including disease diagnosis, examination results, treatment plan, medication history, operation record, etc., and collecting user basic information refers to collecting basic information of the user, such as age, gender, height, weight, occupation, lifestyle (such as diet, exercise habits), etc., to comprehensively understand the health status and lifestyle of the user.

[0040] Further including, performing feature extraction on the electronic medical record information and the user basic information to obtain feature keywords. Text processing is performed on the electronic medical record information, including steps such as word segmentation, stop word removal, and part-of-speech tagging, and then using natural language processing (NLP) technology to extract key feature words from the electronic medical record information and user basic information, i.e., obtaining feature keywords, which may include disease names, symptom descriptions, examination result abnormal items, lifestyle keywords, etc., as the basis for subsequent user analogy and data analysis.

[0041] Further including, performing user analogy in the risk identification cloud platform according to the word frequency characteristics, and outputting an analogy permission user group. The word frequency characteristics of the feature keywords, i.e., the frequency of a certain keyword appearing in a specific user group, are used to evaluate the relevance between the keyword and the user's health status. In the risk identification cloud platform, user analogy is performed according to the word frequency characteristics of the feature keywords, i.e., finding other user groups that have similar health characteristics (such as similar diseases, similar symptoms, similar lifestyles, etc.) as the current user, i.e., outputting an analogy permission user group, which may have similar disease development trends, treatment needs, or health improvement strategies.

[0042] Further including, generating the motion sensing data samples according to the analogy permission user group. According to the health characteristics and needs of the analogy permission user group, the specific requirements of the motion sensing data samples are determined. For example, it may be necessary to collect data such as exercise intensity, exercise frequency, exercise type, etc. within a certain period of time, and then use motion sensing modules (such as acceleration sensors, displacement sensors, etc.) to monitor the real-time motion of users in the analogy permission user group, and preprocess the collected motion sensing data, including data cleaning (removing noise, outliers, etc.), data conversion (such as converting acceleration data to speed, displacement, etc.), and data labeling (labeling or classifying data accordingly), etc. The preprocessed motion sensing data is sorted into motion sensing data samples, which are used for training of the risk identification fusion model.

[0043] Next, the specific configuration of the risk identification fusion model training module 30 will be described in detail. The risk identification fusion model training module 30 can further include: collecting motion sensing data health samples of the user; performing model optimization training with the motion sensing data health samples as comparison data of the motion sensing data samples, and outputting the risk identification fusion model. The motion sensing data of the user when the user is in good health, has no obvious disease or movement disorder, and is on the home postoperative care bed is collected as the motion sensing data health samples, which comprehensively reflect the movement habits and health status of the user, and then the motion sensing data health samples are used as comparison data of the motion sensing data samples to obtain the difference degree of the two data samples, and then the difference degree of the data samples is used as a parameter of the model for optimization training, thereby obtaining the risk identification fusion model. The risk identification fusion model can comprehensively consider information from multiple sources and improve the accuracy and efficiency of risk identification.

[0044] Although the present application makes various references to certain modules in the platform according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server, and the various units and modules are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for easy mutual differentiation, and do not limit the protection scope of the present application.

[0045] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A home-based postoperative care training risk identification platform based on the Internet of Things, characterized in that, The platform includes: A home postoperative care bed connection module is used to connect to a home postoperative care bed. The home postoperative care bed is equipped with a motion sensing module, which senses the user's movements in bed in real time and obtains historical motion sensing datasets. The historical motion sensor dataset transmission module is used to transmit the historical motion sensor dataset to the risk identification cloud platform based on the Internet of Things, wherein the risk identification cloud platform is connected to the home postoperative care bed management system. A risk identification fusion model training module is used to train a risk identification fusion model and embed the risk identification fusion model into the risk identification cloud platform. The risk identification fusion model is obtained through training data, which includes motion sensor data samples of authorized users and identification information that identifies the degree of risk. The first risk indicator output module is used to download the risk identification fusion model to the intelligent early warning module of the home postoperative care bed. The intelligent early warning module performs risk identification on the real-time motion sensor dataset and outputs the first risk indicator. The first reminder information generation module is used to generate a first reminder information based on the intelligent early warning module if the first risk indicator is greater than the preset risk indicator. The steps performed by the risk identification fusion model training module include: Collect the user's electronic medical record information and basic user information; Feature extraction is performed on the electronic medical record information and the user basic information to obtain feature keywords; Based on word frequency characteristics, user analogies are performed on the risk identification cloud platform to output the user group with analogy permissions; The motion sensing data sample is generated based on the analogous permission user group; The platform also includes: Collect the user's motion sensor data and health sample; The model is optimized and trained using the healthy samples of the motion sensing data as comparison data of the motion sensing data samples, and the risk identification fusion model is output.

2. The home-based postoperative care training risk identification platform based on the Internet of Things as described in claim 1, characterized in that, The home-based postoperative care bed is also equipped with a physiological sensing module; The physiological sensing module is used to sense the user’s movements in bed in real time and to obtain historical physiological sensing datasets. The physiological sensing module and the motion sensing module are synchronized, and the historical physiological sensing dataset and the historical motion sensing dataset are processed for time synchronization, and the synchronized historical physiological sensing dataset and historical motion sensing dataset are output. Based on the Internet of Things, the synchronized historical physiological sensor dataset and historical motion sensor dataset are transmitted to the risk identification cloud platform.

3. The IoT-based home-based postoperative care training risk identification platform as described in claim 2, characterized in that, The historical motion sensor dataset transmission module performs the following steps: The risk identification cloud platform performs anomaly identification on the synchronized historical physiological sensor dataset and historical motion sensor dataset, and outputs abnormal physiological sensor dataset and abnormal motion sensor dataset. Align the time series of the abnormal physiological sensor dataset and the abnormal motion sensor dataset, and extract the synchronous abnormal sensor data group. The training data for training the risk identification fusion model is processed based on the synchronous anomaly sensing data set.

4. The home-based postoperative care training risk identification platform based on the Internet of Things as described in claim 1, characterized in that, The historical motion sensor dataset transmission module performs the following steps: Using Internet of Things (IoT) communication technology, a first transmission channel is established between the motion sensing module and the home gateway, and a second transmission channel is established between the home gateway and the risk identification cloud platform; The historical motion sensor dataset is transmitted in encrypted form using the first transmission channel and the second transmission channel.

5. The home-based postoperative care training risk identification platform based on the Internet of Things as described in claim 1, characterized in that, The motion sensing module includes an acceleration sensor, a displacement sensor, and a pressure sensor; The user is sensed in real time using the accelerometer to obtain motion intensity data and frequency data. The displacement sensor performs real-time sensing of the user to obtain user position data and motion displacement data. The pressure sensor is used to sense the user in real time to obtain user position data and pressure distribution data. The historical motion sensing dataset is obtained using motion intensity data, frequency data, user location data, motion displacement data, user body position data, and pressure distribution data as parameters.

6. The IoT-based home-based postoperative care training risk identification platform as described in claim 2, characterized in that, The physiological sensing module includes a heart rate sensor, a blood oxygen sensor, and a body temperature sensor. The heart rate sensor is used to detect changes in the user's heart rate, the blood oxygen sensor is used to detect the user's blood oxygen level, and the body temperature sensor is used to detect changes in the user's body temperature. Heart rate variability data, blood oxygen level data, and body temperature variability data were used as parameters to obtain historical physiological sensor datasets.

Images