Remote monitoring system for medical patients

Abstract

The invention relates to the technical field of remote medical monitoring, in particular to a medical patient remote monitoring system which comprises a health monitoring module, a disease course prediction module, a physiological regulation module, a medicine management module, a behavior analysis module, a therapy customization module, a dose adjustment module and a system comprehensive management module. According to the invention, accurate physiological monitoring is realized through a data fusion algorithm and a noise filtering technology, the accuracy and reliability of data are ensured, the recurrent neural network model can provide more accurate disease development trend analysis, early diagnosis and timely intervention are facilitated, and the accuracy and reliability of the system are improved. A self-adaptive control algorithm and a biological feedback technology enable a patient to more effectively adjust the physiological state by themselves, the treatment initiative is improved, and the safety and effectiveness of drug treatment are improved through a mode recognition algorithm and drug interaction analysis of drug management. Through deep data mining and personalized treatment strategies of the behavior analysis module and the therapy customization module, a more comprehensive and customized treatment scheme is provided for the patient.

Description

Technical Field

[0001] The present invention relates to the technical field of remote medical monitoring, and in particular to a remote monitoring system for medical patients. Background Art

[0002] The field of telemedicine monitoring technology is an important branch of the modern medical system. It uses communication technology to remotely monitor the health status of patients. It integrates multiple disciplines such as medicine, computer science, network technology, and data analysis, aiming to improve the efficiency and quality of patient care, especially in chronic disease management and elderly care. Telemedicine monitoring technology enables doctors to obtain patients' physiological data in real time, such as heart rate, blood pressure, blood sugar level, etc., and conduct effective health management without the need for patients to go to medical institutions in person.

[0003] Among them, the medical patient remote monitoring system is a remote medical monitoring system designed for medical patients. The main purpose is to monitor the health status of medical patients in real time through remote technical means, so as to timely discover and deal with potential health problems. The system aims to improve the management efficiency of medical diseases such as cardiovascular disease and diabetes, reduce the frequency of patients visiting the hospital, and provide doctors with real-time data to better formulate and adjust treatment plans.

[0004] Traditional medical patient monitoring systems rely more on experience to predict the course of disease, lack efficient data analysis and prediction models, and are difficult to accurately predict the progression of the disease, increasing treatment uncertainty. In terms of drug management and patient behavior analysis, traditional systems lack intelligent data processing and personalized treatment plans, resulting in unsatisfactory drug treatment effects and safety, and affecting the quality of life of patients. The lack of an effective system comprehensive management mechanism and insufficient coordination between modules have reduced overall treatment efficiency and effectiveness, resulting in poor treatment results, reduced patient satisfaction, and even increased risks of medical accidents. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a remote monitoring system for medical patients.

[0006] In order to achieve the above-mentioned object, the present invention adopts the following technical scheme: a remote monitoring system for medical patients, the system comprising a health monitoring module, a disease course prediction module, a physiological regulation module, a drug management module, a behavior analysis module, a therapy customization module, a dosage adjustment module, and a system comprehensive management module; The health monitoring module uses data fusion algorithm and noise filtering technology based on the patient's physiological data to continuously monitor heart rate and blood pressure, converts physiological data into tracking data in a unified format through real-time data standardization and outlier detection, and stores it in a database to generate physiological indicator tracking data; The disease course prediction module uses a recurrent neural network model to perform time series analysis of disease progression based on physiological indicator tracking data, predicts disease development trends through model training and parameter adjustment, and performs probabilistic analysis of disease progression in combination with historical disease course information to generate a disease course development prediction model; The physiological regulation module uses biofeedback technology and adaptive control algorithms based on physiological index tracking data to guide patients to self-regulate heart rate and blood pressure, optimize training plans through regular training and real-time feedback adjustment, achieve stable control of physiological state, and generate physiological regulation plans; The drug management module uses pattern recognition algorithms and drug interaction analysis technology based on the patient's drug taking history and physiological indicator tracking data to analyze the effects and side effects of drugs, dynamically adjust the drug plan, optimize the drug plan through drug adaptability evaluation and adjustment strategy, and generate an optimized drug management plan; The behavior analysis module uses cluster analysis algorithms and association rule mining technology based on the patient's living habits and behavior pattern data to conduct in-depth analysis of the patient's daily behavior. Through pattern mining and trend prediction, it puts forward targeted lifestyle adjustment suggestions, assists in the formulation of treatment plans, and generates behavior pattern analysis results; The therapy customization module uses genetic algorithms and multi-objective optimization strategies based on the disease course prediction model and behavior pattern analysis results, and combines the patient's disease course characteristics and living habits to iteratively search and optimize the treatment plan combination, while evaluating the effectiveness and feasibility of the plan, continuously updating and optimizing the treatment strategy, optimizing the management of the treatment plan, and generating a customized treatment plan; The dosage adjustment module uses Bayesian network model and pharmacokinetics simulation based on drug response data and physiological index tracking data to accurately adjust drug dosage and type, dynamically optimize dosage through continuous monitoring and data analysis, ensure the safety and effectiveness of drug treatment, and generate accurate drug dosage adjustment plan; The system comprehensive management module is based on physiological indicator tracking data, disease course prediction models, physiological regulation plans, optimized drug management plans, behavioral pattern analysis results, customized treatment plans and accurate drug dosage adjustment plans. It uses system dynamic analysis and performance evaluation technology to monitor the operating status of the system, continuously optimize the system performance through a data-driven decision-making process, and generate system function evaluation results.

[0007] The improvements of the present invention are that the physiological indicator tracking data includes heart rate changes and blood pressure fluctuations, the disease course prediction model is specifically a development trend analysis of various chronic diseases, the physiological adjustment plan includes heart rate adjustment strategies and blood pressure control techniques, the optimized drug management plan includes drug type selection and medication time arrangement, the behavior pattern analysis results include daily activity habits and behavioral regularities, the customized treatment plan is specifically a comprehensive treatment strategy for individuals, the accurate drug dosage adjustment plan specifically refers to dosage optimization and drug type selection, and the system function evaluation results are specifically overall operating efficiency and synergy between modules.

[0008] The present invention is improved in that the health monitoring module includes a heart rate monitoring submodule, a blood pressure monitoring submodule, and a comprehensive status monitoring submodule; The heart rate monitoring submodule uses a fast Fourier transform algorithm based on the patient's physiological data to convert the time domain signal into a frequency domain signal, separate the effective information and noise in the heart rate signal, and use a high-pass filter to remove the low-frequency noise in the data, extract the accurate heart rate signal, and generate purified heart rate data; The blood pressure monitoring submodule uses a logistic regression algorithm based on purified heart rate data to predict blood pressure trends by analyzing the relationship between heart rate data and blood pressure. It also uses a pulse wave velocity analysis method to calculate blood pressure by measuring the propagation speed of blood in blood vessels, monitor blood pressure changes, and generate stable blood pressure data. The comprehensive status monitoring submodule is based on purified heart rate data and stabilized blood pressure data, adopts a multivariate time series analysis method, constructs a time series model of multiple physiological indicators, analyzes the correlation and mutual influence between multiple indicators, combines a health risk assessment model, integrates multiple physiological indicators to evaluate the patient's comprehensive health status, and generates physiological indicator tracking data.

[0009] The present invention is improved in that the disease course prediction module includes a trend analysis submodule, a deep learning modeling submodule, and a disease progression prediction submodule; The trend analysis submodule is based on the physiological index tracking data, adopts the autoregressive moving average model, and analyzes the intrinsic characteristics of the time series data by combining the autoregressive and sliding average techniques, identifies the time series patterns and trend changes of the heart rate and blood pressure physiological indicators, and generates the time series trend analysis results; The deep learning modeling submodule uses a long short-term memory network to process time series data based on the time series trend analysis results, avoids long-term dependency problems by maintaining a long-term memory unit, learns deep patterns and relationships in the time series, predicts future trends of heart rate and blood pressure physiological indicators, and generates a deep learning disease prediction model; The disease progression prediction submodule is based on a deep learning disease prediction model and adopts the Bayesian estimation method to analyze the disease progression. By combining prior knowledge and new observation data, it provides a probabilistic prediction of disease progression, reveals the uncertainty of disease development, makes accurate predictions and decisions, and generates a disease progression prediction model.

[0010] The present invention is improved in that the physiological regulation module includes a heart rate regulation submodule, a blood pressure regulation submodule, and a relaxation training submodule; The heart rate regulation submodule uses frequency domain analysis method based on physiological index tracking data to analyze the main frequency components of the heart rate signal, including signal decomposition and frequency component extraction, identifies and analyzes key patterns in the heart rate signal, applies adaptive filtering technology to adjust and optimize the heart rate signal, and generates a heart rate regulation solution; The blood pressure regulation submodule is based on the heart rate regulation scheme and uses pressure feedback adjustment technology to analyze the correlation between heart rate and blood pressure and the time series characteristics of blood pressure data, implement dynamic adjustment, and use an adaptive control algorithm to automatically adjust parameters according to the real-time monitored blood pressure data to generate a blood pressure regulation scheme; The relaxation training submodule implements psychological relaxation techniques based on the blood pressure regulation program, combines deep breathing and meditation exercises, guides patients to perform specific relaxation exercises, continuously pays attention to the patient's physiological reactions, including heart rate and breathing rate, optimizes the relaxation training process based on the data, reduces psychological and physiological stress, and generates a physiological regulation program.

[0011] The present invention is improved in that the drug management module includes a drug record submodule, a medication plan management submodule, and a drug reaction monitoring submodule; The drug record submodule uses a cluster analysis algorithm based on the patient's drug use history to group drug use patterns, identify the use patterns and trends of multiple types of drugs, calculate the similarity between drug use data, classify similar data points into the same category, reveal implicit patterns and frequent combinations of drug use, understand the patient's drug use behavior, and generate drug use pattern analysis; The medication plan management submodule is based on the analysis of drug use patterns and uses association rule learning algorithms to mine frequent item sets and strong rules in drug use data, identify interactions and effects between drugs, predict the potential effects of multiple drug combinations on patient health, adjust and optimize medication plans, and generate adjusted medication plans; The drug response monitoring submodule uses a time series analysis algorithm based on the adjusted medication plan to observe the changing trend of drug response at multiple time points and analyze the time dependence of drug effects and side effects. The drug response time series is analyzed to explore the changes in drug effects and potential side effects, adjust the medication plan, and generate an optimized drug management plan.

[0012] The present invention is improved in that the behavior analysis module includes a life habit analysis submodule, a daily behavior analysis submodule, and a behavior pattern analysis submodule; The lifestyle analysis submodule uses the K-means clustering algorithm based on the patient's lifestyle and behavior pattern data to calculate the distance between each data point and the cluster center, assign the data point to the nearest cluster center, implement data classification and preliminary analysis, use data normalization to eliminate differences in data dimensions and ranges, and generate lifestyle analysis results; The daily behavior analysis submodule uses the Apriori association rule mining method based on the results of the life habit analysis to identify frequent item sets and construct association rules between item sets, analyze the correlation and regularity between the daily behavior patterns of patients, and predict and analyze the behavior patterns to generate daily behavior pattern analysis results; The behavior pattern analysis submodule is based on the daily behavior pattern analysis results, uses time series analysis methods and probability statistical models to analyze the time series of historical behavior data, predict future behavior trends, evaluate the occurrence probability of multiple behavior patterns, and generate behavior pattern analysis results.

[0013] The present invention is improved in that the therapy customization module includes a treatment plan design submodule, an efficacy evaluation submodule, and a treatment plan update submodule; The treatment plan design submodule uses a genetic algorithm based on the disease course prediction model and behavior pattern analysis results to encode the treatment plan through random generation of the initial population, evaluates each plan through the fitness function, and uses selection, crossover and mutation operations to iteratively optimize the plan. Combined with the patient's disease course characteristics and living habits, it finds the optimal treatment plan and generates a preliminary treatment plan design; The efficacy evaluation submodule is based on the design of the preliminary treatment plan, adopts a multi-objective optimization strategy, combines the effects of the treatment plan on efficacy, side effects, and cost, sets multiple evaluation indicators, conducts a comprehensive evaluation of each treatment plan, compares the comprehensive performance of multiple plans, optimizes and adjusts the plan, and generates efficacy evaluation results; The treatment plan update submodule is based on the efficacy evaluation results and adopts a genetic algorithm to adjust the treatment plan according to the changes in the patient's disease course and individual differences through continuous iterative optimization, update the fitness function and iterative parameters, continuously optimize and personalize the plan, and generate a customized treatment plan.

[0014] The present invention is improved in that the dosage adjustment module includes a dosage assessment submodule, a dosage adjustment strategy submodule, and a drug type selection submodule; The dosage assessment submodule uses a Bayesian network model to perform dosage assessment based on drug response data and physiological index tracking data. By defining the probabilistic dependency relationship between variables and using the Bayesian theorem to update and calculate conditional probabilities, the interaction between drug dosage and physiological response is analyzed, the optimal dosage range is determined, and a drug dosage assessment plan is generated. The dosage adjustment strategy submodule is based on the drug dosage evaluation scheme, uses pharmacokinetics simulation to adjust the dosage, establishes a mathematical model to analyze the dynamic process of the drug in the body, including absorption, distribution, metabolism and excretion, adjusts the drug dosage, optimizes the treatment effect, and generates a drug dosage adjustment scheme; The drug type selection submodule selects the drug type based on the drug dosage adjustment plan, analyzes the pharmacodynamic properties and interactions of the drugs in combination with the patient's condition and historical drug reactions, selects the drug type that best suits the patient's current condition, optimizes the treatment plan, and generates an accurate drug dosage adjustment plan.

[0015] The present invention is improved in that the system comprehensive management module includes a system monitoring submodule, a performance optimization submodule, and a function evaluation submodule; The system monitoring submodule uses real-time data monitoring algorithms to analyze system data flow and interaction, track data changes, identify anomalies and efficiency issues in system operation, and generate system operation monitoring data based on physiological indicator tracking data, disease course prediction models, physiological regulation plans, optimized drug management plans, behavioral pattern analysis results, customized treatment plans, and accurate drug dosage adjustment plans; The performance optimization submodule applies machine learning optimization algorithms based on system operation monitoring data to conduct in-depth analysis of system performance, predict system performance bottlenecks by learning historical data, automatically adjust system parameters and operation strategies, optimize system response speed and processing capacity, and generate performance optimization solutions; The functional evaluation submodule is based on the performance optimization solution and uses a comprehensive performance evaluation model to comprehensively evaluate multiple dimensions of system operation, including service quality, operational stability and treatment effect. By analyzing and comparing multiple indicators, it provides a comprehensive quantitative and qualitative evaluation of system operation and generates system functional evaluation results.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, accurate physiological monitoring is achieved through data fusion algorithm and noise filtering technology to ensure the accuracy and reliability of data. The recurrent neural network model can provide more accurate disease development trend analysis in disease course prediction, which is helpful for early diagnosis and timely intervention. The application of adaptive control algorithm and biofeedback technology enables patients to self-regulate their physiological state more effectively, and improves the initiative and personalization of treatment. The pattern recognition algorithm and drug interaction analysis of drug management improve the safety and effectiveness of drug treatment. The deep data mining and personalized treatment strategies of the behavior analysis module and the therapy customization module provide patients with a more comprehensive and customized treatment plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The module diagram of the remote monitoring system for medical patients is proposed for the present invention; Figure 2 The present invention proposes a system framework diagram of a remote monitoring system for medical patients; Figure 3 A schematic diagram of a health monitoring module in a remote monitoring system for medical patients is provided for the present invention; Figure 4 The present invention proposes a schematic diagram of a disease course prediction module in a remote monitoring system for medical patients; Figure 5 A schematic diagram of a physiological regulation module in a remote monitoring system for medical patients is provided for the present invention; Figure 6 The present invention proposes a schematic diagram of a drug management module in a remote monitoring system for medical patients; Figure 7 The present invention proposes a schematic diagram of a behavior analysis module in a remote monitoring system for medical patients; Figure 8 A schematic diagram of a therapy customization module in a remote monitoring system for medical patients is provided for the present invention; Fig. 9 The present invention proposes a schematic diagram of a dosage adjustment module in a remote monitoring system for medical patients; Fig.10 The present invention provides a schematic diagram of a system integrated management module in a remote monitoring system for internal medicine patients. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0020] Example: See Figure 1 , the present invention provides a technical solution: a remote monitoring system for medical patients, the system includes a health monitoring module, a disease course prediction module, a physiological regulation module, a drug management module, a behavior analysis module, a therapy customization module, a dosage adjustment module, and a system comprehensive management module; The health monitoring module uses data fusion algorithm and noise filtering technology based on the patient's physiological data to continuously monitor heart rate and blood pressure. Through real-time data standardization and outlier detection, the physiological data is converted into tracking data in a unified format, and stored in the database to generate physiological indicator tracking data; The disease course prediction module uses a recurrent neural network model to perform time series analysis of disease progression based on physiological indicator tracking data. Through model training and parameter adjustment, it predicts the disease development trend and combines historical disease course information to perform probabilistic analysis of disease progression and generate a disease course development prediction model. The physiological regulation module uses biofeedback technology and adaptive control algorithms based on physiological index tracking data to guide patients to self-regulate their heart rate and blood pressure. Through regular training and real-time feedback adjustment, it optimizes the training program, achieves stable control of the physiological state, and generates a physiological regulation program. The drug management module uses pattern recognition algorithms and drug interaction analysis technology to analyze the effects and side effects of drugs based on the patient's drug taking history and physiological indicator tracking data, dynamically adjust the drug plan, optimize the drug plan through drug adaptability evaluation and adjustment strategy, and generate an optimized drug management plan; The behavior analysis module uses cluster analysis algorithms and association rule mining technology to conduct in-depth analysis of patients' daily behaviors based on their living habits and behavior pattern data. Through pattern mining and trend prediction, it proposes targeted lifestyle adjustment suggestions, assists in the formulation of treatment plans, and generates behavior pattern analysis results. The therapy customization module uses genetic algorithms and multi-objective optimization strategies based on the disease course prediction model and behavior pattern analysis results, and combines the patient's disease course characteristics and living habits to iteratively search and optimize the treatment plan combination, while evaluating the effectiveness and feasibility of the plan, continuously updating and optimizing the treatment strategy, optimizing the management of the treatment plan, and generating a customized treatment plan; The dosage adjustment module uses Bayesian network model and pharmacokinetics simulation to accurately adjust the dosage and type of drugs based on drug response data and physiological indicator tracking data. Through continuous monitoring and data analysis, it dynamically optimizes the dosage to ensure the safety and effectiveness of drug treatment and generate accurate drug dosage adjustment plans. The system comprehensive management module is based on physiological indicator tracking data, disease course prediction models, physiological regulation plans, optimized drug management plans, behavioral pattern analysis results, customized treatment plans and accurate drug dosage adjustment plans. It uses system dynamic analysis and performance evaluation technology to monitor the system's operating status, continuously optimize system performance through a data-driven decision-making process, and generate system function evaluation results.

[0021] Physiological indicator tracking data include heart rate changes and blood pressure fluctuations. The disease course prediction model specifically analyzes the development trends of various chronic diseases. The physiological adjustment plan includes heart rate adjustment strategies and blood pressure control techniques. The optimized drug management plan includes drug type selection and medication schedule. The behavioral pattern analysis results include daily activity habits and behavioral patterns. The customized treatment plan is a comprehensive treatment strategy for individuals. The accurate drug dosage adjustment plan specifically refers to dosage optimization and drug type selection. The system function evaluation results specifically include overall operating efficiency and synergy between modules.

[0022] See also Figure 2 and Figure 3 ,The health monitoring module includes a heart rate monitoring submodule, a blood pressure monitoring submodule, and a comprehensive state monitoring submodule; The heart rate monitoring submodule is based on the patient's physiological data and uses the fast Fourier transform algorithm to convert the time domain signal into a frequency domain signal, separate the effective information and noise in the heart rate signal, and use a high-pass filter to remove the low-frequency noise in the data, extract the accurate heart rate signal, and generate purified heart rate data; The heart rate monitoring submodule is based on the patient's physiological data. It converts the time domain signal into the frequency domain signal to separate the effective information and noise in the heart rate signal. , evaluates the frequency content of the signal, where the heart rate frequency is usually between 0.5 and 4 Hz. By comparing the amplitude of each frequency component with the threshold , will be lower than The frequency components are regarded as noise and removed. This threshold It is determined by the average frequency amplitude of the heart rate signal in the resting state. Next, a high-pass filter is applied with a cutoff frequency of 0.5 Hz to remove all noise below this frequency, ensuring that the signal retains the effective frequency of the heart rate. The processing process not only clearly removes the noise, but also strengthens the quality of the heart rate signal to generate purified heart rate data.

[0023] The blood pressure monitoring submodule is based on purified heart rate data and uses a logistic regression algorithm to analyze the relationship between heart rate data and blood pressure to predict blood pressure trends. Combined with the pulse wave velocity analysis method, it calculates blood pressure by measuring the speed at which blood propagates in blood vessels, monitors blood pressure changes, and generates stable blood pressure data. The blood pressure monitoring submodule is based on purified heart rate data and predicts blood pressure trends by analyzing the relationship between heart rate data and blood pressure. The module first sets a baseline heart rate value to standardize heart rate changes. The baseline value is calculated by averaging the historical heart rate data. Based on the purified heart rate data, the module calculates the heart rate change rate. , It is the ratio of the difference between the current heart rate and the baseline heart rate to the baseline heart rate. Then, the pulse wave velocity analysis method is used to calculate blood pressure by measuring the speed of blood propagation in blood vessels. and blood pressure The relationship between To describe, It is a constant determined empirically based on the elastic coefficient of blood vessels. In this way, the module can monitor the instantaneous changes in blood pressure and generate stable blood pressure data.

[0024] The comprehensive status monitoring submodule is based on purified heart rate data and stable blood pressure data. It uses multivariate time series analysis methods to build time series models of multiple physiological indicators, analyze the correlation and mutual influence between multiple indicators, and integrate multiple physiological indicators to evaluate the patient's comprehensive health status and generate physiological indicator tracking data in combination with the health risk assessment model. The comprehensive status monitoring submodule is based on purified heart rate data and stable blood pressure data. It builds a time series model of multiple physiological indicators to analyze the correlation and mutual influence between multiple indicators. First, define the health indicator array , including but not limited to heart rate, blood pressure, and body temperature. Set a window period for each physiological indicator, such as the past 7 days and 30 days, to calculate the short-term and long-term averages of these indicators and The mean value is used to construct a linear regression model to predict possible changes in health status in the short term. The objective function of the model is set as ,in, and is the regression coefficient, representing the weight of short-term and long-term trends. is a constant term, which represents the baseline health status. Through this model, multiple physiological indicators can be effectively integrated to evaluate the comprehensive health status of patients and generate physiological indicator tracking data.

[0025] See also Figure 2 and Figure 4 ,The disease course prediction module includes a trend analysis submodule, a deep learning modeling submodule, and a disease progression prediction submodule; The trend analysis submodule is based on the tracking data of physiological indicators and adopts the autoregressive moving average model. By combining autoregressive and sliding average techniques, it analyzes the intrinsic characteristics of time series data, identifies the time series patterns and trend changes of heart rate and blood pressure physiological indicators, and generates time series trend analysis results; The trend analysis submodule is based on the physiological index tracking data. By combining autoregression and sliding average technology, it analyzes the intrinsic characteristics of time series data and identifies the time series patterns and trend changes of heart rate and blood pressure physiological indicators. First, determine the order of the autoregressive term and the order of the sliding average term , set to and , based on these two parameters, an autoregressive moving average model is established The model is specifically ,in is a constant term is the error term, is the autoregressive coefficient, is the moving average coefficient. In addition, the least squares method is used to estimate the above coefficients to ensure that the obtained model reflects the statistical characteristics of the time series most accurately. In this way, the submodule can effectively extract and predict the future trends of heart rate and blood pressure from historical data and generate time series trend analysis results.

[0026] The deep learning modeling submodule uses a long short-term memory network to process time series data based on the results of time series trend analysis. It avoids long-term dependency problems by maintaining a long-term memory unit, learns deep patterns and relationships in time series, predicts future trends of heart rate and blood pressure physiological indicators, and generates a deep learning disease prediction model. The deep learning modeling submodule uses a long short-term memory network to process time series data based on the results of time series trend analysis. It avoids long-term dependency problems by maintaining a long-term memory unit and learns deep patterns and relationships in time series. During the implementation process, an LSTM network with multiple hidden layers is constructed. Each hidden layer has several LSTM units, which can control the preservation and forgetting of information through a gating mechanism. Specifically, the network controls the flow of information through three types of gates (input gate, forget gate, and output gate). The function of each gate is controlled by data-driven weights and bias parameters, so that the network can add new observation data while maintaining historical information. Through this structure, the submodule can predict future trends of physiological indicators such as heart rate and blood pressure, and generate a deep learning disease prediction model; The disease progression prediction submodule is based on a deep learning disease prediction model and uses the Bayesian estimation method to analyze disease progression. By combining prior knowledge and new observation data, it provides a probabilistic prediction of disease progression, reveals the uncertainty of disease development, makes accurate predictions and decisions, and generates a disease progression prediction model. The disease progression prediction submodule is based on a deep learning disease prediction model and uses the Bayesian estimation method to analyze the disease progression by combining prior knowledge and new observation data. First, the prior distribution is defined, usually based on historical case data to set the probability model of disease development, and then the model is updated in combination with new patient data to form a posterior distribution. This process can be expressed by the formula posterior probability α = likelihood probability × prior probability, where the likelihood probability is the probability based on the new observation data and the prior probability is the probability based on historical knowledge. In this way, the submodule can not only reveal the uncertainty of disease development, but also make more accurate predictions and decisions based on the patient's specific situation, and generate a disease course prediction model.

[0027] See also Figure 2 and Figure 5 ,The physiological regulation module includes a heart rate regulation submodule, a blood pressure regulation submodule, and a relaxation training submodule; The heart rate regulation submodule uses frequency domain analysis method based on physiological index tracking data to analyze the main frequency components of the heart rate signal, including signal decomposition and frequency component extraction, identify and analyze key patterns in the heart rate signal, apply adaptive filtering technology to adjust and optimize the heart rate signal, and generate a heart rate regulation solution; The heart rate regulation submodule analyzes the main frequency components of the heart rate signal based on the physiological indicator tracking data, including signal decomposition and frequency component extraction. In this process, the main frequency components are first identified by calculating the power spectrum density of the heart rate signal. The power spectrum density reflects the energy distribution of the signal at different frequencies. Through this analysis, the dominant frequency range in the heart rate signal can be clearly identified, such as the frequency range of 0.8 to 2.5 Hz, which is usually associated with heart rate fluctuations in adults. After determining the main frequency, a bandpass filter is applied to allow only signals within this frequency range to pass through to reduce interference caused by other frequencies. The filtering process is finely controlled by adjusting the center frequency and bandwidth of the filter to ensure that the most valuable information is retained in the heart rate signal, thereby generating a heart rate regulation solution.

[0028] The blood pressure regulation submodule is based on the heart rate regulation scheme and uses pressure feedback adjustment technology to analyze the correlation between heart rate and blood pressure, as well as the time series characteristics of blood pressure data, implement dynamic adjustment, and use adaptive control algorithms to automatically adjust parameters according to real-time monitored blood pressure data to generate a blood pressure regulation scheme; The blood pressure regulation submodule uses pressure feedback adjustment technology based on the heart rate regulation scheme to analyze the correlation between heart rate and blood pressure, as well as the time series characteristics of blood pressure data. By establishing a mathematical model between heart rate and blood pressure, the module can predict the dynamic adjustment of blood pressure based on heart rate changes. For example, if the heart rate increases, the model predicts that blood pressure may also increase; at this time, the submodule automatically adjusts the pressure feedback control parameters to stabilize blood pressure. Dynamic adjustment is achieved through a closed-loop control system that adjusts feedback parameters based on real-time monitored blood pressure data, such as adjusting the drug infusion rate or adjusting physical intervention measures (such as the tightness of the compression belt). Through adaptive control strategies, the blood pressure regulation submodule can effectively respond to rapid changes in patient conditions and generate blood pressure regulation schemes.

[0029] The relaxation training submodule implements psychological relaxation techniques based on the blood pressure regulation program, combines deep breathing and meditation exercises, guides patients to perform specific relaxation exercises, continuously pays attention to the patient's physiological reactions, including heart rate and breathing rate, optimizes the relaxation training process based on the data, reduces psychological and physiological stress, and generates a physiological regulation program; The relaxation training submodule implements psychological relaxation techniques based on the blood pressure regulation program, combined with deep breathing and meditation exercises. This module first determines the patient's stress level and psychological state through psychological assessment, and then develops a personalized relaxation training plan. During the training process, the module continuously monitors the patient's heart rate and respiratory rate, and the physiological data is used to evaluate the effectiveness of the training. For example, if the heart rate and respiratory rate are significantly reduced when deep breathing exercises are implemented, it means that the relaxation effect is good. In addition, the module will adjust the intensity and duration of the training according to real-time data, optimize the relaxation exercises, and enable the patient to achieve the best state of relaxation during the exercises. This dynamic adjustment is based on physiological data feedback to ensure that the relaxation training is both safe and effective, thereby reducing the patient's psychological and physiological stress and generating a physiological regulation program.

[0030] See also Figure 2 and Figure 6 ,The drug management module includes a drug record submodule, a medication plan management submodule, and a drug reaction monitoring submodule; The drug record submodule uses a cluster analysis algorithm based on the patient's drug use history to group drug use patterns, identify the use patterns and trends of multiple types of drugs, calculate the similarity between drug use data, classify similar data points into the same category, reveal the implicit patterns and frequent combinations of drug use, understand the patient's drug use behavior, and generate drug use pattern analysis; The medication record submodule groups medication usage patterns based on the patient's medication history and identifies the patterns and trends of medications of various types. First, by sorting and classifying the patient's medication records, including information such as the type of medication, time of administration, and dosage, the data is preprocessed, such as removing outliers and supplementing missing data. Then, a distance-based clustering method is used to calculate the similarity between different medication usage records. The similarity can be measured by the Euclidean distance of the frequency of administration and dosage. By setting an appropriate distance threshold, medication records with high similarity are grouped together to form several groups of medication use. Each group represents a specific medication use pattern, revealing the implicit patterns and frequent combinations of patients' medication use, helping doctors better understand patients' medication use behaviors, and generating medication use pattern analysis results.

[0031] The medication plan management submodule is based on the analysis of drug use patterns and uses association rule learning algorithms to mine frequent item sets and strong rules in drug use data, identify the interactions and effects between drugs, predict the potential effects of multiple drug combinations on patient health, adjust and optimize medication plans, and generate adjusted medication plans; The medication plan management submodule uses association rule learning based on drug use pattern analysis to mine frequent item sets and strong rules in drug use data. This submodule first uses the Apriori algorithm, which iteratively searches for frequent item sets to identify co-occurring drug combinations. Through these data, the interactions and effects between drugs can be discovered. For example, when two drugs are often used at the same time, they may lead to increased side effects or improved efficacy. Based on these association rules, the submodule further predicts the potential impact of different drug combinations on patient health. These predictions are based on the intensity and frequency of drug interactions. Finally, based on these analysis results, the medication plan is adjusted and optimized to ensure the safety and maximum effectiveness of the drug combination, and an adjusted medication plan is generated.

[0032] The drug response monitoring submodule uses a time series analysis algorithm based on the adjusted medication plan to observe the changing trend of drug response at multiple time points and analyze the time dependence of drug effects and side effects. The drug response time series is analyzed to explore changes in drug effects and potential side effects, adjust the medication plan, and generate an optimized drug management plan; The drug response monitoring submodule observes the changing trends of drug responses at multiple time points based on the adjusted medication plan. First, a time series of drug response data is constructed, including the effects of the drug such as the speed of pain relief and the time of occurrence of side effects. Using methods such as moving average and exponential smoothing, the time series data is smoothed to reduce the interference of random fluctuations. In this way, the module can clearly observe how drug effects and side effects change over time, which is particularly important for long-term drug treatment. Based on the results of these time series analyses, the module further explores the duration of drug effects and the frequency of side effects, as well as the correlation. Finally, the drug plan is adjusted according to the analysis results, such as adjusting the drug dosage or changing the type of drug, to optimize the treatment effect and reduce adverse reactions, and generate an optimized drug management plan.

[0033] See also Figure 2 and 7 ,The behavior analysis module includes the life habit analysis submodule, the daily behavior analysis submodule, and the behavior pattern analysis submodule; The lifestyle analysis submodule uses the K-means clustering algorithm based on the patient's lifestyle and behavior pattern data to calculate the distance between each data point and the cluster center, assign the data point to the nearest cluster center, implement data classification and preliminary analysis, use data normalization to eliminate differences in data dimensions and ranges, and generate lifestyle analysis results; The lifestyle analysis submodule uses the K-means clustering algorithm based on the patient's lifestyle and behavior pattern data to calculate the distance between each data point and the cluster center, assign the data point to the nearest cluster center, and achieve data classification and preliminary analysis. In the specific operation, all lifestyle data are first normalized to ensure that the dimensions of each dimension of data are consistent. This is usually achieved by calculating the difference between each attribute value and its average value and dividing it by the standard deviation, that is, ,in is the original data, and are the mean and standard deviation of the data respectively. Next, set the number of clusters , randomly selected Cluster centers are then calculated, and the distance from each data point to these centers is classified into the nearest center. The distance calculation usually uses the Euclidean distance formula. ,in is an attribute of a data point, is the corresponding attribute of the cluster center. Through the iterative optimization process, until the position of the cluster center no longer changes, the life habit analysis results are generated.

[0034] The daily behavior analysis submodule uses the Apriori association rule mining method based on the results of life habit analysis to identify frequent item sets and construct association rules between item sets, analyze the correlation and regularity between patients' daily behavior patterns, predict and analyze behavior patterns, and generate daily behavior pattern analysis results; The daily behavior analysis submodule uses the Apriori association rule mining method based on the results of life habit analysis to identify frequent item sets and construct association rules between item sets. First, the minimum support threshold and the minimum confidence threshold are set. These two parameters determine the conditions for an item set to be considered a frequent item set. The identification of frequent item sets can be described as follows: an item set is considered a frequent item set if its frequency of occurrence in all transactions is not less than , then the item set is a frequent item set. The association rule form is , the confidence calculation formula is Conf(A , where supp(X) represents the support of the item set, that is, the proportion of item sets. Through this method, the correlation and regularity between patients' daily behavior patterns can be found, and then the behavior patterns can be predicted and analyzed to generate daily behavior pattern analysis results.

[0035] The behavior pattern analysis submodule uses time series analysis methods and probability statistical models based on the results of daily behavior pattern analysis to analyze the time series of historical behavior data, predict future behavior trends, evaluate the probability of occurrence of multiple behavior patterns, and generate behavior pattern analysis results; The behavior pattern analysis submodule uses time series analysis methods and probability statistical models based on the results of daily behavior pattern analysis to analyze the time series of historical behavior data and predict future behavior trends. In the specific method, the change of behavior pattern over time is defined as time series. , and apply ARIMA model for analysis and prediction. The formula of ARIMA model can be expressed as , are the orders of autoregression and moving average, respectively. represents the parameters of the moving average part, is the error term. By fitting this model, the probability of occurrence of multiple behavior patterns can be evaluated, and the behavior pattern analysis results can be further generated.

[0036] See also Figure 2 and Figure 8 ,The therapy customization module includes the treatment plan design submodule, the ,efficacy evaluation submodule, and the treatment plan update submodule; The treatment plan design submodule uses a genetic algorithm to encode treatment plans through random generation of the initial population based on the disease course prediction model and behavior pattern analysis results, evaluates each plan through the fitness function, and uses selection, crossover and mutation operations to iteratively optimize the plan. Combined with the patient's disease course characteristics and lifestyle habits, it finds the best treatment plan and generates a preliminary treatment plan design; The treatment plan design submodule uses a genetic algorithm to encode the treatment plan by randomly generating an initial population based on the disease course prediction model and behavior pattern analysis results. In this process, each treatment plan is regarded as an individual and encoded using a series of parameters (such as drug dosage, treatment frequency, etc.) to form a specific coding sequence. The effectiveness of each plan is evaluated using a fitness function, which is calculated based on the treatment effect prediction and patient adaptability. The fitness function can be expressed as ,in represents the predicted treatment effect, Represents potential side effects, and is a weight factor used to balance the effects of treatment and side effects. Next, the scheme is optimized through selection, crossover and mutation operations. The selection operation selects excellent individuals according to their fitness, and the crossover and mutation operations introduce new genetic information to promote the diversity of solutions and the depth of optimization. Finally, the optimal treatment plan is found based on the patient's disease course characteristics and lifestyle habits, and a preliminary treatment plan design is generated.

[0037] The efficacy evaluation submodule is based on the design of the preliminary treatment plan, adopts a multi-objective optimization strategy, combines the impact of the treatment plan on efficacy, side effects, and cost, sets multiple evaluation indicators, conducts a comprehensive evaluation of each treatment plan, compares the comprehensive performance of multiple plans, optimizes and adjusts the plan, and generates efficacy evaluation results; The efficacy evaluation submodule is based on the preliminary treatment plan design, adopts a multi-objective optimization strategy, and combines the impact of the treatment plan on efficacy, side effects, and cost. This module sets multiple evaluation indicators, such as the treatment effect indicator , side effect indicators , and cost indicators The optimization of these indicators can be achieved by the formula to carry out, among which , and It is the weight of each indicator, which expresses the relative importance of treatment effect, side effects and cost. By evaluating the comprehensive performance of each treatment plan, comparing the plans, and optimizing and adjusting the plans based on this comprehensive evaluation, the efficacy evaluation results are finally generated.

[0038] The treatment plan update submodule uses genetic algorithms based on the results of efficacy evaluation to adjust the treatment plan according to the changes in the patient's disease course and individual differences through continuous iterative optimization, update the fitness function and iterative parameters, continuously optimize and personalize the plan, and generate a customized treatment plan; The treatment plan update submodule uses genetic algorithms again based on the efficacy evaluation results to adjust the treatment plan according to the changes in the patient's disease course and individual differences through continuous iterative optimization. In this process, the fitness function is updated to reflect the patient's current condition and treatment response, and iterative parameters such as mutation rate and crossover rate are adjusted to adapt to the dynamic changes in the disease course. The update of the fitness function can be expressed as ,in and Demonstrate new treatment effects and evaluate side effects and is the adjusted weight factor. The continuous optimization and personalized adjustment process ultimately generates a customized treatment plan, ensuring that the treatment plan evolves over time and with changes in the patient's condition.

[0039] See also Figure 2 and 9 ,The dose adjustment module includes a dose assessment submodule, a dose adjustment strategy submodule, and a drug type selection submodule; The dose assessment submodule uses the Bayesian network model to perform dose assessment based on drug response data and physiological index tracking data. By defining the probabilistic dependency relationship between variables and using the Bayesian theorem to update and calculate conditional probabilities, it analyzes the mutual influence between drug dose and physiological response, determines the optimal dose range, and generates a drug dose assessment plan. The dose assessment submodule uses a Bayesian network model to perform dose assessment based on drug response data and physiological indicator eye tracking data. First, define the drug dose , physiological response , and other relevant influencing factors As nodes in the network, and establish probabilistic dependencies between them. Using Bayes' theorem to update and calculate conditional probabilities, the formula can be expressed as ,in is the conditional probability of a physiological response given the drug dose and other factors It is the joint probability of drug dose and other factors under the condition of given physiological response. Through this method, the mutual influence between drug dose and physiological response can be accurately analyzed, and the optimal dose range can be determined based on the probability model to generate a drug dose evaluation plan.

[0040] The dose adjustment strategy submodule uses pharmacokinetics simulation to adjust the dose based on the drug dose evaluation plan, establishes a mathematical model to analyze the dynamic process of drugs in the body, including absorption, distribution, metabolism and excretion, adjusts the drug dose, optimizes the treatment effect, and generates a drug dose adjustment plan; The dose adjustment strategy submodule uses pharmacokinetics simulation to adjust the dose based on the drug dose evaluation scheme. A mathematical model is constructed to analyze the dynamic process of the drug in the body, which usually includes the absorption rate of the drug. , volume of distribution , clearance rate , and half-life The specific model can be expressed as ,in Represents at time The model optimizes the drug dose by adjusting the absorption rate, distribution volume and clearance rate so that the drug concentration is maintained within the therapeutic window, thereby optimizing the therapeutic effect and generating a drug dose adjustment plan.

[0041] The drug type selection submodule selects drug types based on the drug dosage adjustment plan, analyzes the pharmacodynamic properties and interactions of drugs in combination with the patient's condition and historical drug response, selects the drug type that best suits the patient's current condition, optimizes the treatment plan, and generates an accurate drug dosage adjustment plan; The drug type selection submodule selects drug types based on the drug dosage adjustment scheme. Analyzes the pharmacodynamic properties of drugs, such as the mechanism of action of drugs , efficacy , and possible side effects . Combined with the patient's condition and history of drug reactions , using the evaluation function To determine which drug is best for the patient. This evaluation function weighs the drug choices based on the drug's efficacy and the patient's history of drug response, thereby selecting the type of drug that best suits the patient's current condition, optimizing the treatment plan, and generating an accurate drug dosage adjustment plan.

[0042] See also Figure 2 and Fig.10 ,The system comprehensive management module includes the system monitoring submodule, the ,performance optimization submodule, and the function evaluation submodule; The system monitoring submodule uses real-time data monitoring algorithms to analyze system data flow and interaction, track data changes, identify anomalies and efficiency issues in system operation, and generate system operation monitoring data based on physiological indicator tracking data, disease course prediction models, physiological regulation plans, optimized drug management plans, behavioral pattern analysis results, customized treatment plans, and accurate drug dosage adjustment plans. The system monitoring submodule uses real-time data monitoring algorithms to analyze system data flow and interaction based on physiological indicator tracking data, disease progression prediction models, physiological regulation plans, optimized drug management plans, behavioral pattern analysis results, customized treatment plans, and accurate drug dosage adjustment plans. This process uses an event-driven architecture to capture system status and activities in real time by collecting and analyzing various data input and output streams. Specific methods include setting up listeners to track key data indicators such as data throughput, response time, and error rate. The indicator uses the formula: throughput ,The response time is the time difference from the initiation of a request to the completion of a response.,Through this monitoring, the system can identify anomalies and efficiency problems in,operation in real time.,For example, through anomaly point detection, an alarm is triggered when a data indicator suddenly deviates from its,normal range, helping to identify system performance bottlenecks or potential failure points,and generate system operation monitoring data.

[0043] The performance optimization submodule uses machine learning optimization algorithms based on system operation monitoring data to conduct in-depth analysis of system performance. By learning historical data, it predicts system performance bottlenecks, automatically adjusts system parameters and operation strategies, optimizes system response speed and processing capabilities, and generates performance optimization solutions. The performance optimization submodule applies machine learning optimization algorithms based on system operation monitoring data to conduct in-depth analysis of system performance. This submodule uses regression analysis and clustering algorithms to identify performance patterns in historical data and predict future system bottlenecks through these patterns. The algorithm builds a prediction model based on historical performance data, such as using a linear regression model. ,in Represents system delay, Represents the load level, and is a model parameter. Based on this model, the response time of the system under different loads can be predicted. In addition, optimization algorithms such as gradient descent are used to automatically adjust system parameters such as memory allocation and number of concurrent threads to achieve the best performance configuration. Adaptive adjustment can optimize system response speed and processing power and generate performance optimization solutions.

[0044] The function evaluation submodule uses a comprehensive performance evaluation model based on the performance optimization solution to comprehensively evaluate multiple dimensions of system operation, including service quality, operation stability, and treatment effect. By analyzing and comparing multiple indicators, it provides a comprehensive quantitative and qualitative evaluation of system operation and generates system function evaluation results. The functional evaluation submodule uses a comprehensive performance evaluation model based on the performance optimization solution to comprehensively evaluate multiple dimensions of system operation, including service quality, operational stability, and treatment effect. This module defines key performance indicators (KPIs), such as system availability, by establishing an evaluation framework. ,reliability , and user satisfaction These indicators may be expressed as To conduct a comprehensive assessment, , , is the weight of each indicator. In addition, this submodule provides a comprehensive quantitative and qualitative evaluation of system operation by comparing the data before and after performance optimization, such as the reduction of error rate and the improvement of processing speed. Such a comprehensive evaluation helps identify the strengths and weaknesses of the system, thereby providing a basis for continuous improvement of the system and generating system function evaluation results.

[0045] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A remote monitoring system for medical patients, characterized in that: The system includes a health monitoring module, a disease course prediction module, a physiological regulation module, a drug management module, a behavior analysis module, a therapy customization module, a dosage adjustment module, and a system comprehensive management module; The health monitoring module uses data fusion algorithm and noise filtering technology based on the patient's physiological data to continuously monitor heart rate and blood pressure, converts physiological data into tracking data in a unified format through real-time data standardization and outlier detection, and stores it in a database to generate physiological indicator tracking data; The disease course prediction module uses a recurrent neural network model to perform time series analysis of disease progression based on physiological indicator tracking data, predicts disease development trends through model training and parameter adjustment, and performs probabilistic analysis of disease progression in combination with historical disease course information to generate a disease course development prediction model; The physiological regulation module uses biofeedback technology and adaptive control algorithms based on physiological index tracking data to guide patients to self-regulate heart rate and blood pressure, optimize training plans through regular training and real-time feedback adjustment, achieve stable control of physiological state, and generate physiological regulation plans; The drug management module uses pattern recognition algorithms and drug interaction analysis technology based on the patient's drug taking history and physiological indicator tracking data to analyze the effects and side effects of drugs, dynamically adjust the drug plan, optimize the drug plan through drug adaptability evaluation and adjustment strategy, and generate an optimized drug management plan; The behavior analysis module uses cluster analysis algorithms and association rule mining technology based on the patient's living habits and behavior pattern data to conduct in-depth analysis of the patient's daily behavior. Through pattern mining and trend prediction, it puts forward targeted lifestyle adjustment suggestions, assists in the formulation of treatment plans, and generates behavior pattern analysis results; The therapy customization module uses genetic algorithms and multi-objective optimization strategies based on the disease course prediction model and behavior pattern analysis results, and combines the patient's disease course characteristics and living habits to iteratively search and optimize the treatment plan combination, while evaluating the effectiveness and feasibility of the plan, continuously updating and optimizing the treatment strategy, optimizing the management of the treatment plan, and generating a customized treatment plan; The dosage adjustment module uses Bayesian network model and pharmacokinetics simulation based on drug response data and physiological index tracking data to accurately adjust drug dosage and type, dynamically optimize dosage through continuous monitoring and data analysis, ensure the safety and effectiveness of drug treatment, and generate accurate drug dosage adjustment plan; The system comprehensive management module is based on physiological indicator tracking data, disease course prediction models, physiological regulation plans, optimized drug management plans, behavioral pattern analysis results, customized treatment plans and accurate drug dosage adjustment plans. It uses system dynamic analysis and performance evaluation technology to monitor the operating status of the system, continuously optimize the system performance through a data-driven decision-making process, and generate system function evaluation results.

2. The medical patient remote monitoring system according to claim 1, characterized in that: The physiological indicator tracking data includes heart rate changes and blood pressure fluctuations. The disease course prediction model specifically analyzes the development trends of various chronic diseases. The physiological adjustment plan includes heart rate adjustment strategies and blood pressure control techniques. The optimized drug management plan includes drug type selection and medication schedule. The behavioral pattern analysis results include daily activity habits and behavioral patterns. The customized treatment plan is specifically a comprehensive treatment strategy for individuals. The accurate drug dosage adjustment plan specifically refers to dosage optimization and drug type selection. The system function evaluation results specifically include overall operating efficiency and synergy between modules.

3. The medical patient remote monitoring system according to claim 1, characterized in that: The health monitoring module includes a heart rate monitoring submodule, a blood pressure monitoring submodule, and a comprehensive status monitoring submodule; The heart rate monitoring submodule uses a fast Fourier transform algorithm based on the patient's physiological data to convert the time domain signal into a frequency domain signal, separate the effective information and noise in the heart rate signal, and use a high-pass filter to remove the low-frequency noise in the data, extract the accurate heart rate signal, and generate purified heart rate data; The blood pressure monitoring submodule uses a logistic regression algorithm based on purified heart rate data to predict blood pressure trends by analyzing the relationship between heart rate data and blood pressure. It also uses a pulse wave velocity analysis method to calculate blood pressure by measuring the propagation speed of blood in blood vessels, monitor blood pressure changes, and generate stable blood pressure data. The comprehensive status monitoring submodule is based on purified heart rate data and stabilized blood pressure data, adopts a multivariate time series analysis method, constructs a time series model of multiple physiological indicators, analyzes the correlation and mutual influence between multiple indicators, combines a health risk assessment model, integrates multiple physiological indicators to evaluate the patient's comprehensive health status, and generates physiological indicator tracking data.

4. The medical patient remote monitoring system according to claim 1, characterized in that: The disease course prediction module includes a trend analysis submodule, a deep learning modeling submodule, and a disease progression prediction submodule; The trend analysis submodule is based on the physiological index tracking data, adopts the autoregressive moving average model, and analyzes the intrinsic characteristics of the time series data by combining the autoregressive and sliding average techniques, identifies the time series patterns and trend changes of the heart rate and blood pressure physiological indicators, and generates the time series trend analysis results; The deep learning modeling submodule uses a long short-term memory network to process time series data based on the time series trend analysis results, avoids long-term dependency problems by maintaining a long-term memory unit, learns deep patterns and relationships in the time series, predicts future trends of heart rate and blood pressure physiological indicators, and generates a deep learning disease prediction model; The disease progression prediction submodule is based on a deep learning disease prediction model and adopts the Bayesian estimation method to analyze the disease progression. By combining prior knowledge and new observation data, it provides a probabilistic prediction of disease progression, reveals the uncertainty of disease development, makes accurate predictions and decisions, and generates a disease progression prediction model.

5. The medical patient remote monitoring system according to claim 1, characterized in that: The physiological regulation module includes a heart rate regulation submodule, a blood pressure regulation submodule, and a relaxation training submodule; The heart rate regulation submodule uses frequency domain analysis method based on physiological index tracking data to analyze the main frequency components of the heart rate signal, including signal decomposition and frequency component extraction, identifies and analyzes key patterns in the heart rate signal, applies adaptive filtering technology to adjust and optimize the heart rate signal, and generates a heart rate regulation solution; The blood pressure regulation submodule is based on the heart rate regulation scheme and uses pressure feedback adjustment technology to analyze the correlation between heart rate and blood pressure and the time series characteristics of blood pressure data, implement dynamic adjustment, and use an adaptive control algorithm to automatically adjust parameters according to the real-time monitored blood pressure data to generate a blood pressure regulation scheme; The relaxation training submodule implements psychological relaxation techniques based on the blood pressure regulation program, combines deep breathing and meditation exercises, guides patients to perform specific relaxation exercises, continuously pays attention to the patient's physiological reactions, including heart rate and breathing rate, optimizes the relaxation training process based on the data, reduces psychological and physiological stress, and generates a physiological regulation program.

6. The medical patient remote monitoring system according to claim 1, characterized in that: The drug management module includes a drug record submodule, a medication plan management submodule, and a drug reaction monitoring submodule; The drug record submodule uses a cluster analysis algorithm based on the patient's drug use history to group drug use patterns, identify the use patterns and trends of multiple types of drugs, calculate the similarity between drug use data, classify similar data points into the same category, reveal implicit patterns and frequent combinations of drug use, understand the patient's drug use behavior, and generate drug use pattern analysis; The medication plan management submodule is based on the analysis of drug use patterns and uses association rule learning algorithms to mine frequent item sets and strong rules in drug use data, identify interactions and effects between drugs, predict the potential effects of multiple drug combinations on patient health, adjust and optimize medication plans, and generate adjusted medication plans; The drug response monitoring submodule is based on the adjusted medication plan and adopts a time series analysis algorithm to observe the changing trend of drug response at multiple time points, analyze the time dependence of drug effects and side effects, analyze the drug response time series, explore the changes in drug effects and potential side effects, adjust the medication plan, and generate an optimized drug management plan.

7. The medical patient remote monitoring system according to claim 1, characterized in that: The behavior analysis module includes a life habit analysis submodule, a daily behavior analysis submodule, and a behavior pattern analysis submodule; The lifestyle analysis submodule uses the K-means clustering algorithm based on the patient's lifestyle and behavior pattern data to calculate the distance between each data point and the cluster center, assign the data point to the nearest cluster center, implement data classification and preliminary analysis, use data normalization to eliminate differences in data dimensions and ranges, and generate lifestyle analysis results; The daily behavior analysis submodule uses the Apriori association rule mining method based on the results of the life habit analysis to identify frequent item sets and construct association rules between item sets, analyze the correlation and regularity between the daily behavior patterns of patients, and predict and analyze the behavior patterns to generate daily behavior pattern analysis results; The behavior pattern analysis submodule is based on the daily behavior pattern analysis results, uses time series analysis methods and probability statistical models to analyze the time series of historical behavior data, predict future behavior trends, evaluate the occurrence probability of multiple behavior patterns, and generate behavior pattern analysis results.

8. The medical patient remote monitoring system according to claim 1, characterized in that: The therapy customization module includes a treatment plan design submodule, an efficacy evaluation submodule, and a treatment plan update submodule; The treatment plan design submodule uses a genetic algorithm based on the disease course prediction model and behavior pattern analysis results to encode the treatment plan through random generation of the initial population, evaluates each plan through the fitness function, and uses selection, crossover and mutation operations to iteratively optimize the plan. Combined with the patient's disease course characteristics and living habits, it finds the optimal treatment plan and generates a preliminary treatment plan design; The efficacy evaluation submodule is based on the design of the preliminary treatment plan, adopts a multi-objective optimization strategy, combines the effects of the treatment plan on efficacy, side effects, and cost, sets multiple evaluation indicators, conducts a comprehensive evaluation of each treatment plan, compares the comprehensive performance of multiple plans, optimizes and adjusts the plan, and generates efficacy evaluation results; The treatment plan update submodule is based on the efficacy evaluation results and adopts a genetic algorithm to adjust the treatment plan according to the changes in the patient's disease course and individual differences through continuous iterative optimization, update the fitness function and iterative parameters, continuously optimize and personalize the plan, and generate a customized treatment plan.

9. The medical patient remote monitoring system according to claim 1, characterized in that: The dosage adjustment module includes a dosage assessment submodule, a dosage adjustment strategy submodule, and a drug type selection submodule; The dosage assessment submodule uses a Bayesian network model to perform dosage assessment based on drug response data and physiological index tracking data. By defining the probabilistic dependency relationship between variables and using the Bayesian theorem to update and calculate conditional probabilities, the interaction between drug dosage and physiological response is analyzed, the optimal dosage range is determined, and a drug dosage assessment plan is generated. The dosage adjustment strategy submodule is based on the drug dosage evaluation scheme, uses pharmacokinetics simulation to adjust the dosage, establishes a mathematical model to analyze the dynamic process of the drug in the body, including absorption, distribution, metabolism and excretion, adjusts the drug dosage, optimizes the treatment effect, and generates a drug dosage adjustment scheme; The drug type selection submodule selects the drug type based on the drug dosage adjustment plan, analyzes the pharmacodynamic properties and interactions of the drugs in combination with the patient's condition and historical drug reactions, selects the drug type that best suits the patient's current condition, optimizes the treatment plan, and generates an accurate drug dosage adjustment plan.

10. The medical patient remote monitoring system according to claim 1, characterized in that: The system comprehensive management module includes a system monitoring submodule, a performance optimization submodule, and a function evaluation submodule; The system monitoring submodule uses real-time data monitoring algorithms to analyze system data flow and interaction, track data changes, identify anomalies and efficiency issues in system operation, and generate system operation monitoring data based on physiological indicator tracking data, disease course prediction models, physiological regulation plans, optimized drug management plans, behavioral pattern analysis results, customized treatment plans, and accurate drug dosage adjustment plans; The performance optimization submodule applies machine learning optimization algorithms based on system operation monitoring data to conduct in-depth analysis of system performance, predict system performance bottlenecks by learning historical data, automatically adjust system parameters and operation strategies, optimize system response speed and processing capacity, and generate performance optimization solutions; The functional evaluation submodule is based on the performance optimization solution and uses a comprehensive performance evaluation model to comprehensively evaluate multiple dimensions of system operation, including service quality, operational stability and treatment effect. By analyzing and comparing multiple indicators, it provides a comprehensive quantitative and qualitative evaluation of system operation and generates system functional evaluation results.

Images