Internal medicine multi-parameter real-time remote monitoring system based on Internet of Things

Abstract

The invention provides an internal medicine multi-parameter real-time remote monitoring system based on the Internet of Things, and belongs to the technical field of remote monitoring, the internal medicine multi-parameter real-time remote monitoring system comprises an integrated data acquisition terminal, a heterogeneous network communication module, an intelligent remote analysis server and a user interaction terminal, and the integrated data acquisition terminal adopts a medical grade flexible wearable design. According to the internal medicine multi-parameter real-time remote monitoring system based on the Internet of Things provided by the invention, the efficiency of remote monitoring of internal medicine diseases is remarkably improved through multi-dimensional technical innovation. On the data processing level, efficient fusion and stable transmission of multi-source data are achieved, and the real-time performance and integrity of monitoring data are guaranteed; in an analysis level, a specialized deep neural network model is combined with environmental factors, so that the accuracy of health state assessment and risk prediction is greatly improved; according to the application level, through a personalized rehabilitation scheme, intelligent early warning and remote equipment management, the medical service process is optimized, and patient compliance and medical cooperation efficiency are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of remote monitoring, in particular to a multi-parameter real-time remote monitoring system for internal medicine based on Internet of Things. BACKGROUND

[0002] Traditional internal medicine patient health monitoring relies on in-hospital equipment or single-parameter portable devices, which has many limitations. On the one hand, multi-source data (physiological parameters and environmental parameters) are collected in a scattered manner, and there is a lack of efficient fusion means, resulting in low data utilization and difficulty in fully reflecting the patient's health status. On the other hand, remote transmission is restricted by network environment, and there are problems of poor stability and high power consumption, which cannot meet the real-time monitoring requirements. In addition, existing analysis models are mostly general-purpose, without considering the characteristics of internal medicine diseases and environmental factors, resulting in insufficient diagnostic accuracy and prediction ability. Moreover, the data security and equipment maintenance system are not perfect, which makes it difficult to support long-term application in medical scenarios. These defects make it difficult for traditional monitoring systems to play an effective role in chronic disease management, home rehabilitation and other scenarios, and there is an urgent need for more intelligent and integrated solutions.

[0003] Therefore, a multi-parameter real-time remote monitoring system for internal medicine based on Internet of Things is proposed. SUMMARY

[0004] The present application provides a multi-parameter real-time remote monitoring system for internal medicine based on Internet of Things to solve the problems raised in the background art.

[0005] The specific technical solution is as follows:

[0006] A multi-parameter real-time remote monitoring system for internal medicine based on Internet of Things, comprising:

[0007] An integrated data acquisition terminal: adopting a medical-grade flexible wearable design, containing a multi-modal physiological parameter sensor group (heart rate, blood pressure, blood oxygen saturation, body temperature sensor), an environmental parameter sensor (temperature and humidity sensor and multi-gas detection sensor array), and a micro control unit (MCU); the MCU is configured to perform real-time timestamp synchronization, filtering and noise reduction, and format standardization preprocessing on heterogeneous physiological data streams, and to integrate the preprocessed physiological data and environmental parameters in space-time correlation to form a comprehensive monitoring data package; a heterogeneous network communication module: connected with the MCU, adopting an adaptive transmission strategy, dynamically selecting the optimal transmission path of 5G, NB-IoT or Wi-Fi network according to network signal strength and bandwidth, and transmitting the comprehensive monitoring data package with low power consumption;

[0008] The intelligent remote analysis server comprises a multi-parameter fusion analysis model based on a deep neural network (such as LSTM, Transformer), which is trained based on a specific internal disease historical data set and incorporates environmental parameters as influencing factors, can analyze the dynamic correlation between physiological parameters and environmental parameters, real-time infer the patient health status trend and potential risk level, and generate a structured health report; the intelligent remote analysis server realizes real-time bidirectional data interaction with a hospital information system (HIS) and an electronic medical record system (EMR) through a standardized API interface;

[0009] The user interaction terminal comprises a medical staff terminal and a patient family member terminal, is configured with a specialized customized monitoring application program, and is used for visualizing display of real-time data, a health report and graded early warning information; the medical staff terminal is provided with a parameterized remote control interface and can dynamically adjust a sensor sampling frequency, precision or switch state.

[0010] As a preferred scheme of the present application, the integrated data acquisition terminal comprises a high-precision fusion positioning module, adopts a multi-source data fusion algorithm of GPS, Beidou and base station positioning, and realizes real-time acquisition of a patient position and a motion trajectory; the MCU performs spatiotemporal fusion marking on the position information and physiological and environmental data, and the intelligent remote analysis server uses the information to perform activity mode analysis and emergency position auxiliary judgment.

[0011] As a preferred scheme of the present application, detection threshold values of the multi-gas detection sensor array refer to clinical environmental guidelines of internal diseases such as chronic obstructive pulmonary disease (COPD) and asthma; a server data analysis engine performs coupled analysis on harmful gas concentration data and patient parameters such as blood oxygen saturation and respiratory frequency, and evaluates an instant influence of the environment on a disease condition.

[0012] As a preferred scheme of the present application, the multi-parameter fusion analysis model adopts a transfer learning technology, fine tunes small sample data of specific internal medicine specialities such as diabetes management in an endocrinology department on the basis of pre-training on a general physiological data set, and improves accuracy of individualized health status judgment and early abnormal prediction ability.

[0013] As a preferred scheme of the present application, the monitoring application program of the user interaction terminal provides a multi-dimensional data visualization dashboard, and uses interactive superimposed curve graphs and heat maps to display time correlation of physiological and environmental parameters; when a composite early warning rule (such as physiological parameter abnormality + environmental parameter exceeding a standard) is triggered, vibration, sound and pop-up window graded reminders are started according to a risk level, and associated data snapshots and preliminary processing suggestions are pushed.

[0014] As a preferred scheme of the present application, the shell of the integrated data acquisition terminal is of split structure (a stretchable fabric bracelet main body and a replaceable sensor chest patch module), is made of antibacterial medical silica gel or breathable textile, the internal sensor layout is optimized by electromagnetic compatibility (EMC), and the wearing comfort and signal stability are ensured.

[0015] As a preferred scheme of the present application, the remote server is provided with a distributed disaster recovery backup module, adopts an incremental backup and version control strategy, synchronizes encrypted data to a physically isolated hard disk array and a medical compliance cloud storage platform, and guarantees data integrity and business continuity.

[0016] As a preferred scheme of the present application, the internal medicine multi-parameter real-time remote monitoring system based on the Internet of Things adopts an end-to-end data security protection system: the MCU hardware-level encrypts original data through a lightweight AES encryption chip, the communication module is additionally provided with a dynamic message authentication code (MAC), the server stores data by using a national encryption algorithm (SM2 / SM4) hybrid encryption, and implements role-based permission management (RBAC) and operation audit logs, so that the requirements of medical data privacy protection regulations such as HIPAA and GDPR are met.

[0017] As a preferred scheme of the present application, the remote control interface of the medical staff terminal integrates a device diagnosis subsystem, can trigger a deep self-checking process of the data acquisition terminal, the MCU injects a calibration signal into the sensor according to a diagnosis script, detects an output response and an internal power supply and memory state, generates a structured diagnosis report containing a fault code, and assists remote maintenance decision-making.

[0018] As a preferred scheme of the present application, the remote server data analysis engine is connected to an individualized health management knowledge base, dynamically generates a digital rehabilitation scheme based on patient historical data trends, current health status, environmental influences and medical record information through a rule engine and a recommendation algorithm, including: dynamic dietary suggestions based on real-time data, individualized exercise prescriptions within a safety threshold, drug-environment interaction warnings, predictive re-consultation time windows and environmental improvement guidance.

[0019] The present application has the following beneficial effects:

[0020] The internal medicine multi-parameter real-time remote monitoring system based on the Internet of Things provided by the application significantly improves the efficiency of remote monitoring of internal medicine diseases through multi-dimensional technical innovation. At the data processing level, efficient fusion and stable transmission of multi-source data are realized, ensuring the real-time and integrity of the monitoring data; at the analysis level, the specialized deep neural network model combined with environmental factors significantly improves the accuracy of health status assessment and risk prediction; at the application level, through personalized rehabilitation programs, intelligent early warning and remote device management, the medical service process is optimized, and patient compliance and medical collaboration efficiency are improved. In addition, the end-to-end data security protection system and distributed disaster recovery backup mechanism constructed by the system ensure the security and compliance of medical data, meet the long-term clinical application requirements, and provide a reliable intelligent solution for internal medicine chronic disease management and home rehabilitation scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The composition block diagram of the internal medicine multi-parameter real-time remote monitoring system based on the Internet of Things provided by the embodiment of the application is shown in the figure.

[0022] Figure 2 The workflow diagram of the dynamic space-time fusion evaluation equation provided by the embodiment of the application is shown in the figure.

[0023] Figure 3 The data processing efficiency curve of the internal medicine multi-parameter real-time remote monitoring system based on the Internet of Things provided by the embodiment of the application is shown in the figure.

[0024] Figure 4 The accuracy rate curve of the multi-parameter fusion analysis model provided by the embodiment of the application with the change of the number of training iterations is shown in the figure.

[0025] Figure 5 The curve of patient compliance with time is shown in the figure. DETAILED DESCRIPTION

[0026] The technical solutions of the application will be further described below in conjunction with the drawings and through specific embodiments.

[0027] Among them, the drawings are only used for illustrative explanation, and the representation is only a schematic diagram, not a physical diagram, and cannot be understood as a limitation of the patent; in order to better illustrate the embodiments of the application, some components of the drawings will be omitted, enlarged or reduced, and do not represent the size of the actual product; for those skilled in the art, it is understandable that some known structures and their descriptions in the drawings can be omitted.

[0028] The same or similar reference numerals in the drawings of the embodiments of the present application correspond to the same or similar components; in the description of the present application, it should be understood that, if the terms "upper", "lower", "left", "right", "inner", "outer" and the like indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore the terms describing the positional relationship in the drawings are only used for exemplary illustration, and cannot be understood as a limitation on the present patent, and for those skilled in the art, the specific meanings of the above terms can be understood according to the specific circumstances.

[0029] In the description of the present application, unless otherwise explicitly specified and limited, if the terms "connection" and the like indicating the connection relationship between components appear, the term should be understood in a broad sense, for example, it can be a fixed connection, or a detachable connection, or an integral; it can be a mechanical connection, or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium; it can be the communication inside two components or the interaction relationship between two components. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0030] Embodiment

[0031] Reference Figures 1-5 As shown in the figure, Figure 1 The composition architecture of the internal medicine multi-parameter real-time remote monitoring system based on the Internet of Things is shown; Figure 2 The working process of the dynamic space-time fusion evaluation equation is shown; Figure 3 The change of data transmission efficiency with time is shown, the X-axis represents time (hours), and the Y-axis represents efficiency, through the curve, the trend of the change of data transmission efficiency with time can be directly seen; Figure 4 The change of model accuracy with training iteration number is shown, the X-axis represents training iteration number, and the Y-axis represents accuracy, through the curve, the trend of the increase of model accuracy with training iteration number can be directly seen; Figure 5 The change of patient compliance with time is shown, the X-axis represents time (days), and the Y-axis represents compliance, through the curve, the trend of the change of patient compliance with time can be directly seen, so as to evaluate the influence of the individualized rehabilitation scheme on patient behavior.

[0032] The internal medicine multi-parameter real-time remote monitoring system based on the Internet of Things provided by the embodiment comprises an integrated data acquisition terminal, a heterogeneous network communication module, an intelligent remote analysis server and a user interaction terminal, wherein:

[0033] The integrated data acquisition terminal adopts a medical-grade flexible wearable design, includes a multi-modal physiological parameter sensor group (heart rate, blood pressure, blood oxygen saturation, body temperature sensor), an environmental parameter sensor (temperature and humidity sensor and multi-gas detection sensor array), and a micro control unit (MCU); the MCU is configured to perform real-time timestamp synchronization, filtering and noise reduction, and format standardization preprocessing on the heterogeneous physiological data stream, and to integrate the preprocessed physiological data and environmental parameters in space-time correlation to form a comprehensive monitoring data package;

[0034] The heterogeneous network communication module is connected with the MCU, adopts an adaptive transmission strategy, dynamically selects the optimal transmission path of 5G, NB-IoT or Wi-Fi network according to network signal strength and bandwidth, and transmits the comprehensive monitoring data package with low power consumption;

[0035] The intelligent remote analysis server has a multi-parameter fusion analysis model based on a deep neural network (such as LSTM, Transformer) built in, which is trained by a specific internal medicine disease historical data set and incorporates environmental parameters as influencing factors, can analyze the dynamic correlation between physiological parameters and environmental parameters, real-time infer the patient's health status trend and potential risk level, and generate a structured health report; the intelligent remote analysis server interacts with the hospital information system (HIS) and the electronic medical record system (EMR) in real time through a standardized API interface;

[0036] The user interaction terminal includes a medical staff terminal and a patient family terminal, and is configured with a specialized customized monitoring application program for visual display of real-time data, health reports and graded warning information; the medical staff terminal has a parameterized remote control interface that can dynamically adjust the sensor sampling frequency, accuracy or switch state.

[0037] By using the above technical scheme, the integrated data acquisition terminal realizes the synchronous acquisition of multi-modal physiological parameters and environmental parameters, and the MCU realizes the space-time correlation integration and adaptive transmission of heterogeneous networks, solving the problems of low multi-source data fusion efficiency and poor transmission stability, ensuring the real-time and integrity of internal medicine patient monitoring data; the intelligent remote analysis server integrates the specialized multi-parameter fusion model with the medical system to realize a closed loop from data acquisition to clinical decision-making, improving the accuracy of internal medicine disease remote monitoring and the efficiency of medical collaboration.

[0038] The multi-parameter fusion analysis model uses a dynamic space-time fusion evaluation equation, and the expression is:

[0039]

[0040] Wherein:

[0041] S(t) is the comprehensive score of the patient's health status at time t;

[0042] f_i(t) represents the standardized feature vector of the i-th physiological parameter (such as heart rate, blood oxygen saturation);

[0043] e_j(t) represents the standardized influence factor of the j-th environmental parameter (such as harmful gas concentration, temperature and humidity);

[0044] w_i(t) is the dynamic spatio-temporal weight function of the physiological parameter, determined by the correlation between the patient's position trajectory and historical data, dynamically adjusted by the spatio-temporal correlation between the patient's position trajectory and disease events, breaking through the limitations of traditional static weights;

[0045] g_j(t) is the specialty sensitive coefficient of the environmental parameter, preset according to the clinical guidelines of COPD, asthma and other diseases, realizing the quantitative mapping of the medical significance of the environmental parameter;

[0046] α, β, γ are adaptive weight parameters optimized by model training;

[0047] σ is an activation function (such as ReLU) for non-linear mapping;

[0048] h(s) is a time series memory kernel function, using dynamic features extracted by LSTM network to capture the time series dependence of physiological parameters and improve early abnormal prediction ability;

[0049] τ is the historical data window length, which is dynamically adjusted by transfer learning for internal medicine and specialty scenes, and is optimized for analysis accuracy under small sample data by transfer learning for diabetes, cardiovascular and other specialty scenes;

[0050] Example: Taking a diabetic patient as an example, when the system collects real-time blood glucose value f_1(t) = 8.5 mmol / L, heart rate f_2(t) = 92 bpm, and environmental sensor detects indoor formaldehyde concentration e_1(t) = 0.08 mg / m 3 , the system calculates the health status score by the following steps:

[0051] 1. Parameter standardization: convert blood glucose and heart rate into deviation feature vectors relative to individual baseline values;

[0052] 2. Weight calculation: according to the spatio-temporal correlation between the patient's current position (such as kitchen) and the position of historical hypoglycemia events, generate w_1(t) = 0.6, w_2(t) = 0.4;

[0053] 3. Environmental sensitivity coefficient: according to the guidelines for diabetes combined with respiratory diseases, set the g_1(t) of formaldehyde concentration to 0.3;

[0054] 4. Time series memory: extract the past 2 hours blood glucose fluctuation trend h(s) by LSTM, calculate the influence of the integral term on the current score;

[0055] 5. Comprehensive calculation: substitution into the equation

[0056] S(t) = 0.7 x σ(0.6 x 1.2 + 0.4 x 0.8 + 0.3 x 1.5) + 0.2 x historical score integral = 0.82 (full score 1 point), triggering moderate early warning.

[0057] Technical effects:

[0058] 1. Multi-dimensional fusion innovation: by integrating physiological parameters, environmental factors, spatio-temporal trajectory and time series memory into a unified equation, the problem of ignoring the dynamic correlation between environment and physiology in traditional models is solved, and the accuracy of health status assessment is significantly improved;

[0059] 2. Specialized adaptive: g_j(t) linkage mechanism with clinical guidelines, realizing personalized assessment of COPD, diabetes and other diseases, significantly reducing misdiagnosis rate compared with general model;

[0060] 3. Dynamic weight optimization: w_i(t) is calculated based on real-time location trajectory, which greatly improves the accuracy of heart rate anomaly recognition in sports scenes, solving the defect that traditional static weight cannot adapt to the activity state of patients;

[0061] 4. Small sample learning advantage: through the combination of τ and transfer learning, in small sample scenes such as diabetes management in endocrinology department, the model convergence speed is greatly improved, filling the analysis gap under the condition of scarce special data.

[0062] Referring to Figure 2 , the working principle process of the dynamic spatio-temporal fusion evaluation equation is as follows:

[0063] 1. Data preprocessing: collect physiological and environmental parameters and standardize to generate f_i(t), e_j(t);

[0064] 2. Spatio-temporal feature extraction: calculate dynamic weight w_i(t) according to patient GPS trajectory and historical data, and generate g_j(t) combined with clinical guidelines;

[0065] 3. Time series modeling: analyze historical health score S(s) through LSTM network to generate memory kernel function h(s); 4. Equation solving: substitute into the equation to calculate real-time health score S(t), triggering corresponding early warning or intervention strategy.

[0066] The dynamic space-time fusion evaluation equation mathematically couples physiological parameters, environmental factors, and space-time trajectories to form a "physiology-environment-space-time" three-dimensional evaluation system; through g_j(t), the environmental exposure threshold of diseases such as COPD and asthma is converted into a calculable parameter, realizing the deep integration of medical knowledge and data models; the LSTM memory kernel function h(s) captures the time dependence of physiological parameters, and compared with the traditional sliding window model, it can more accurately identify trend anomalies; the dynamic adjustment mechanism of w_i(t) and τ enables the model to maintain high resolution in different scenarios such as at home and during exercise, breaking through the scene limitations of static models; through mathematical modeling, the technical details of the multi-parameter fusion analysis model are strengthened, and quantitative support is provided for precise remote monitoring of internal diseases.

[0067] Specifically, in the present embodiment: the integrated data acquisition terminal includes a high-precision fusion positioning module, which uses a multi-source data fusion algorithm of GPS, Beidou, and base station positioning to obtain the patient's position and motion trajectory in real time; the MCU marks the position information with physiological and environmental data for space-time fusion, and the intelligent remote analysis server uses this information for activity pattern analysis and emergency location auxiliary judgment.

[0068] The use of high-precision fusion positioning modules and space-time fusion marking technology enables the system to track the patient's location and activity trajectory in real time, and combined with physiological data, it can analyze the health status in dynamic scenarios (such as changes in heart rate during exercise); in emergency situations (such as sudden illness), the location information can be used to quickly respond and improve emergency handling capabilities, while providing space-time dimensional data support for activity pattern evaluation of chronic disease patients.

[0069] Specifically, in the present embodiment: the detection threshold of the multi-gas detection sensor array is referenced to the clinical environmental guidelines for internal diseases such as chronic obstructive pulmonary disease (COPD) and asthma; the server data analysis engine couples harmful gas concentration data with patient parameters such as oxygen saturation and respiratory rate for analysis, evaluating the immediate impact of the environment on the disease.

[0070] Using an air quality sensor array combined with clinical guidelines for setting thresholds for internal diseases, the environmental factors (such as formaldehyde concentration) and patient physiological parameters (such as respiratory rate) are coupled for analysis, which can quantitatively evaluate the immediate impact of the environment on specific internal diseases (such as COPD), assisting medical personnel in developing environmental intervention programs (such as recommending ventilation time), and filling the gap in traditional monitoring systems that ignore the environmental-health correlation.

[0071] Specifically, in the present embodiment: the multi-parameter fusion analysis model uses transfer learning technology, which fine-tunes small sample data for specific internal medicine specialties such as diabetes management in the endocrinology department based on pre-training on general physiological data sets, improving the accuracy of individualized health status judgment and early abnormal prediction ability.

[0072] By using deep neural network architectures such as LSTM / Transformer, environmental parameters are included as influencing factors in model training, and through transfer learning, the model is fine-tuned for small sample data of internal medicine specialty, significantly improving the model's ability to analyze complex physiological-environmental data, enabling individualized prediction of early abnormalities in internal medicine diseases (such as cardiovascular diseases), and having stronger dynamic trend analysis and potential risk identification capabilities compared to traditional machine learning models.

[0073] Specifically, in this embodiment: the monitoring application of the user interaction terminal provides a multi-dimensional data visualization dashboard to interactively superimpose curve graphs and heat maps to show the time correlation of physiological and environmental parameters; when a composite early warning rule (such as physiological parameter abnormalities + environmental parameter exceedance) is triggered, vibration, sound, and pop-up window alerts are started according to risk levels, and associated data snapshots and preliminary processing suggestions are pushed.

[0074] By using interactive data visualization dashboards and intelligent hierarchical early warning systems, multi-parameter data is presented in intuitive charts, making it easy for medical staff to quickly capture abnormalities; through composite early warning rules combined with hierarchical alert mechanisms, processing suggestions can be automatically pushed according to risk levels, shortening clinical response time; the patient compliance feedback module reports subjective symptoms and calibrates the model, forming a "data collection - analysis - feedback" closed loop, improving the clinical practicality of the monitoring system.

[0075] Specifically, in this embodiment: the integrated data collection terminal has a split structure (a stretchable fabric bracelet main body and a replaceable sensor chest patch module), and is made of antibacterial medical silicone or breathable fabric, with internal sensor layout optimized by electromagnetic compatibility (EMC) design to ensure comfort and signal stability.

[0076] By using medical-grade flexible split structure and antibacterial material design, the problem of poor wearing comfort and easy allergy of traditional monitoring equipment is solved, making it suitable for long-term wear by internal medicine patients; electromagnetic compatibility optimization ensures the stability of multi-sensor signals, avoiding data distortion caused by equipment interference, and providing hardware support for continuous monitoring.

[0077] Specifically, in this embodiment: the remote server deploys a distributed disaster recovery backup module, uses incremental backup and version control strategies, and synchronizes encrypted data to physically isolated hard disk arrays and medical compliant cloud storage platforms, ensuring data integrity and business continuity.

[0078] By combining incremental backup and off-site disaster recovery strategies with medical compliant cloud storage, the integrity and recoverability of patient data can be ensured in the event of hardware failure, network attacks, and other unexpected situations, meeting the long-term storage and regulatory requirements of medical data, and avoiding the risk of diagnosis and treatment errors caused by data loss.

[0079] Specifically, in the embodiment: the medical multi-parameter real-time remote monitoring system based on Internet of Things adopts an end-to-end data security protection system: the MCU encrypts the original data at the hardware level through a lightweight AES encryption chip, the communication module is additionally provided with a dynamic message authentication code (MAC), the server stores data by using a national encryption algorithm (SM2 / SM4) hybrid encryption, and role-based permission management (RBAC) and operation audit logs are implemented, so as to meet the requirements of medical data privacy protection regulations such as HIPAA and GDPR.

[0080] From the full-link encryption from the hardware encryption chip to the national encryption algorithm, combined with the role-based permission management and operation audit, a medical data security barrier conforming to the regulations such as HIPAA and GDPR is constructed, the patient privacy leakage and data tampering are prevented, and the security pain points of sensitive information transmission and storage in the Internet of Things medical system are solved.

[0081] Specifically, in the embodiment: the remote control interface integrated device diagnosis subsystem of the medical staff terminal can trigger the deep self-checking process of the data acquisition terminal; the MCU injects a calibration signal to the sensor according to the diagnosis script, detects the output response and the internal power supply and memory state, generates a structured diagnosis report containing fault codes, and assists the remote maintenance decision.

[0082] The deep self-checking process through the calibration signal injection and fault code generation enables the medical staff to remotely locate the sensor faults, communication abnormalities and other problems, reduces the on-site maintenance cost and equipment downtime, ensures the continuous and reliable operation of the monitoring system, and is especially suitable for the equipment management of patients at home or in remote areas.

[0083] Specifically, in the embodiment: the remote server data analysis engine is connected to the personalized health management knowledge base, based on the patient historical data trend, current health status, environmental influence and medical record information, a digital rehabilitation scheme is dynamically generated through a rule engine and a recommendation algorithm, including: dynamic dietary suggestions based on real-time data, personalized exercise prescriptions within a safety threshold, drug-environment interaction warnings, predictive re-consultation time windows and environmental improvement guidance.

[0084] The dynamic rehabilitation scheme generated based on long-term historical data, environmental analysis and individual medical records upgrades passive monitoring to active intervention, and provides personalized guidance including diet, exercise and medication; the predictive re-consultation suggestion and environmental improvement guidance can reduce the readmission rate, improve the accuracy of internal medicine chronic disease management and patient compliance, and realize whole-cycle management from disease treatment to health maintenance.

[0085] In summary, the working principle of the medical multi-parameter real-time remote monitoring system based on Internet of Things provided in the embodiment is as follows:

[0086] 1. Data acquisition and preprocessing: The integrated data acquisition terminal synchronously collects physiological data and environmental data of the patient through multi-modal physiological parameter sensor groups (heart rate, blood pressure, blood oxygen saturation, body temperature sensors) and environmental parameter sensors (temperature and humidity sensors, multi-gas detection sensor array). The micro control unit (MCU) performs real-time timestamp synchronization, filtering and noise reduction, and format standardization preprocessing on the collected heterogeneous physiological data stream, and integrates it with environmental parameters in space-time correlation to form a comprehensive monitoring data package, providing a high-quality data basis for subsequent analysis.

[0087] 2. Data transmission: The heterogeneous network communication module is connected with the MCU, which adopts adaptive transmission strategy, real-time monitors network signal strength and bandwidth, dynamically selects the optimal transmission path of 5G, NB-IoT or Wi-Fi network, and transmits the comprehensive monitoring data package to the intelligent remote analysis server in a low-power mode, ensuring the timeliness and reliability of data transmission.

[0088] 3. Data analysis and processing: The intelligent remote analysis server is built-in with a multi-parameter fusion analysis model based on deep neural network (such as LSTM, Transformer), which is trained by specific internal medicine disease historical data set and incorporates environmental parameters as influencing factors. After receiving the data package, the server analyzes the dynamic correlation between physiological parameters and environmental parameters using the model, and real-time infers the patient's health status trend and potential risk level, and generates a structured health report. At the same time, through standardized API interface, the server realizes real-time bidirectional data interaction with hospital information system (HIS) and electronic medical record system (EMR), and realizes the sharing and cooperation of medical data.

[0089] 4. Data display and interaction: User interaction terminal (medical staff terminal and patient family terminal) is configured with specialized customized monitoring application program, which receives data transmitted by the server and displays real-time data, health report and graded warning information through multi-dimensional data visualization dashboard (such as interactive superimposed curve graph, heat map). The medical staff terminal has a parameterized remote control interface, which can remotely and dynamically adjust the sampling frequency, accuracy or switch state of the sensor, and can also trigger device self-checking, realizing remote management and maintenance of the data acquisition terminal.

[0090] Method for use

[0091] 1. Patient end: The patient wears the integrated data acquisition terminal, and the device automatically and continuously collects physiological parameters and environmental parameters, and the MCU completes data preprocessing and integration. During the collection process, the patient or family member can report subjective symptoms or medication through the terminal application program to assist the system in calibrating the analysis model.

[0092] 2. Medical staff can view patient data, health reports and early warning information in real time through a dedicated terminal, and adjust the working state of the sensor through a parameterized remote control interface according to the patient's condition. When the system issues a warning or suspects a device failure, the device diagnosis subsystem can be triggered to perform a deep self-check on the data acquisition terminal, obtain a fault diagnosis report and develop a maintenance plan. In addition, medical staff can also provide further diagnosis and treatment recommendations for patients based on the rehabilitation plan generated by the server.

[0093] 3. System management end: The intelligent remote analysis server continuously receives, stores and analyzes data, and synchronizes key information to the clinical workflow through the interface with the hospital information system. At the same time, the server uses a distributed disaster recovery backup module to perform incremental backup and off-site storage of data to ensure data security; and generates personalized rehabilitation plans based on patient data to promote the transition of medical services from passive treatment to active health management.

[0094] Technical effects

[0095] The multi-parameter real-time remote monitoring system for internal medicine based on the Internet of Things significantly improves the efficiency of remote monitoring of internal medicine diseases through multi-dimensional technical innovation. At the data processing level, efficient fusion and stable transmission of multi-source data are achieved to ensure the real-time and integrity of monitoring data; at the analysis level, the specialized deep neural network model combined with environmental factors significantly improves the accuracy of health status assessment and risk prediction; at the application level, through personalized rehabilitation plans, intelligent early warning and remote device management, the medical service process is optimized, and patient compliance and medical collaboration efficiency are improved. In addition, the end-to-end data security protection system and distributed disaster recovery mechanism built by the system ensure the security and compliance of medical data, meet the long-term clinical application requirements, and provide a reliable intelligent solution for internal medicine chronic disease management and home rehabilitation scenarios.

[0096] The above is only the preferred embodiment of the present application, and does not limit the implementation and protection scope of the present application. For those skilled in the art, it should be realized that any equivalent replacement and obvious change made by applying the contents of the present application specification and drawings should be included in the protection scope of the present application.

Claims

1. An Internet of Things-based medical multi-parameter real-time remote monitoring system, characterized in that, Comprise: Integrated data acquisition terminal: adopts medical-grade flexible wearable design, contains multi-modal physiological parameter sensor group, environmental parameter sensor, and MCU; The MCU is configured to perform real-time timestamp synchronization, filtering and noise reduction, and format standardization preprocessing on heterogeneous physiological data streams, and to integrate the preprocessed physiological data and environmental parameters in space-time correlation to form comprehensive monitoring data packets; Heterogeneous network communication module: connected with the MCU, adopts adaptive transmission strategy, dynamically selects the optimal transmission path of 5G, NB-IoT or Wi-Fi network according to network signal strength and bandwidth, and transmits the comprehensive monitoring data packets with low power consumption; Intelligent remote analysis server: built-in multi-parameter fusion analysis model based on deep neural network, which is trained by specific internal medicine disease historical data set and includes environmental parameters as influencing factors, can analyze the dynamic correlation between physiological parameters and environmental parameters, real-time infer the patient's health status trend and potential risk level, and generate structured health report; the intelligent remote analysis server realizes real-time bidirectional data interaction with hospital information system and electronic medical record system through standardized API interface; User interaction terminal: including medical staff terminal and patient family terminal, configured with specialized monitoring application program, used for visual display of real-time data, health report and graded warning information; the medical staff terminal has parameterized remote control interface, which can dynamically adjust the sampling frequency, accuracy or switch state of the sensor.

2. The IoT-based multi-parameter real-time remote monitoring system for internal medicine as claimed in claim 1, wherein: The integrated data acquisition terminal contains a high-precision fusion positioning module, which uses a multi-source data fusion algorithm of GPS, Beidou and base station positioning to obtain the patient's location and motion trajectory in real time; the MCU time-space fusion marks the location information with physiological and environmental data, and the intelligent remote analysis server uses this information for activity pattern analysis and emergency location auxiliary judgment. 3.The Internet of Things based medical multi-parameter real-time remote monitoring system according to claim 1, characterized in that: The detection threshold of the multi-gas detection sensor array refers to the clinical environmental guidelines of chronic obstructive pulmonary disease and asthma internal medicine diseases; the server data analysis engine couples the harmful gas concentration data with the patient's blood oxygen saturation and respiratory rate parameters for analysis, to evaluate the immediate impact of the environment on the patient's condition.

4. The IoT-based multi-parameter real-time remote monitoring system for internal medicine as claimed in claim 1, wherein: The multi-parameter fusion analysis model uses transfer learning technology, which fine-tunes the small sample data of endocrinology department diabetes management specific internal medicine specialty based on the pre-training of general physiological data set, to improve the accuracy of individualized health status judgment and early abnormal prediction ability.

5. The IoT-based multi-parameter real-time remote monitoring system for internal medicine as claimed in claim 1, wherein: The monitoring application program of the user interaction terminal provides a multi-dimensional data visualization dashboard to display the time correlation of physiological and environmental parameters through interactive superimposed curve graphs and heat maps; When the composite warning rule is triggered, start vibration, sound, and pop-up window graded reminders according to risk level, and push associated data snapshots and preliminary processing suggestions. 6.The IoT-based multi-parameter real-time remote monitoring system for internal medicine of claim 1, wherein: The shell of the integrated data acquisition terminal is a split structure, made of antibacterial medical silicone or breathable textile, and the internal sensor layout is designed with electromagnetic compatibility optimization to ensure comfort and signal stability. 7.The IoT-based multi-parameter real-time remote monitoring system for internal medicine of claim 1, wherein: The remote server deploys a distributed disaster recovery backup module, adopts an incremental backup and version control strategy, synchronizes encrypted data to a physically isolated hard disk array and a medical compliance cloud storage platform, and guarantees data integrity and business continuity. 8.The IoT-based multi-parameter real-time remote monitoring system for internal medicine of claim 1, wherein: The end-to-end data security protection system is adopted in the internal medicine multi-parameter real-time remote monitoring system based on the Internet of Things: the MCU encrypts the original data at the hardware level through a lightweight AES encryption chip, the communication module is additionally provided with a dynamic message authentication code, the server stores data by using a national encryption algorithm, and role-based permission management and operation audit logs are implemented, so that the HIPAA and GDPR medical data privacy protection regulations are met. 9.The IoT-based multi-parameter real-time remote monitoring system for internal medicine of claim 1, wherein: The remote control interface of the medical staff terminal is integrated with a device diagnosis subsystem, which can trigger a deep self-checking process of the data acquisition terminal; the MCU injects calibration signals into the sensor according to a diagnosis script, detects the output response and the internal power supply and memory state, generates a structured diagnosis report containing fault codes, and assists in remote maintenance decision-making. 10.The IoT-based multi-parameter real-time remote monitoring system for internal medicine of claim 1, wherein: The remote server data analysis engine is connected with a personalized health management knowledge base, dynamically generates a digital rehabilitation scheme based on the patient's historical data trend, current health status, environmental influence and medical record information through a rule engine and a recommendation algorithm, including dynamic dietary suggestions based on real-time data, personalized exercise prescriptions within a safety threshold, drug-environment interaction warnings, predictive re-consultation time windows and environmental improvement guidance.

Images