Intelligent oral medicine box system for hospitalized patients

Abstract

The invention discloses an intelligent oral medicine box system for hospitalized patients, and relates to the technical field of intelligent oral medicine boxes, and the system comprises an equipment distribution module which is used for distributing intelligent medicine boxes for patients; the medicine distribution and reminding module is used for automatically distributing medicine according to the medicine information of the patient and reminding the patient to take medicine through sound and visual signals; the medicine taking education module is used for providing medicine information for the patient by using the intelligent medicine box; the real-time monitoring and adjusting module is used for the intelligent medicine box to collect health data of a patient through health monitoring equipment, a multiple regression model is constructed to analyze the health data, and a support vector machine (SVM) model is constructed based on the analyzed data to adjust medicine use; and the storage module is used for storing the used data. According to the method, health data are analyzed and predicted through the multiple regression model and the SVM model of the intelligent medicine box system, and personalized medication guidance and dynamic adjustment strategies are provided.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent oral medicine boxes, in particular to an intelligent oral medicine box system for inpatients in a hospital. Background Art

[0002] In recent years, with the rapid development of medical information and intelligent technology, the application of intelligent medical equipment in clinical practice has become more and more extensive. The popularization of technologies such as hospital information system HIS and electronic medical record system EMR has made the management and use of medical data more efficient and convenient. Intelligent drug management systems have emerged, especially in the drug management of inpatients in hospitals. The smart medicine box system has received widespread attention and research due to its convenience and accuracy. Such systems usually integrate advanced technologies such as high-resolution cameras, weight sensors, and wireless communication modules, aiming to improve the accuracy of drug management and patient compliance, reduce human errors, and ensure that patients take medication on time and in the right amount.

[0003] Although the existing smart medicine box systems have solved some problems in traditional drug management to a certain extent, there are still many shortcomings. Many existing systems have relatively single functions and lack comprehensive monitoring and analysis of patient health data, and are unable to achieve personalized medication guidance and dynamic adjustment. The current smart medicine box systems lag behind in data synchronization and real-time communication, making it difficult to update patients' medication information and health status in a timely manner. The existing systems lack interactivity and content richness in medication education, and have failed to give full play to the advantages of multimedia technology, resulting in limited effects on improving patients' drug knowledge and medication skills. Summary of the invention

[0004] In view of the above problems existing in the existing intelligent oral medicine box system for hospitalized patients, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is that the existing systems generally lack consideration for the personalized needs of patients, the drug distribution and medication reminder methods are relatively simple, and cannot be flexibly adjusted according to the actual situation of the patients. The existing systems are not intelligent enough in data analysis and health monitoring, and cannot fully utilize the patients' health data for accurate analysis and dynamic adjustment, resulting in the accuracy and effectiveness of medication management being limited.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: an intelligent oral medicine box system for hospitalized patients, comprising: an equipment allocation module, used to allocate patients with intelligent medicine boxes; a medicine allocation and reminder module, used to automatically allocate medicines according to the patient's medicine information, and remind the patient to take the medicine through sound and visual signals; a medication education module, used to use the intelligent medicine box to provide patients with medicine information; a real-time monitoring and adjustment module, used for the intelligent medicine box to collect the patient's health data through health monitoring equipment, build a multivariate regression model to analyze the health data, and build a support vector machine (SVM) model based on the analyzed data to adjust the medication; a storage module, used to store the used data.

[0007] As a preferred solution of the intelligent oral medicine box system for hospitalized patients of the present invention, wherein: the allocation of the patient intelligent medicine box means that the reception nurse will allocate an intelligent medicine box when registering in the hospital, and the compartment of the medicine box is equipped with a high-resolution camera and a weight sensor;

[0008] The nurse enters the serial number of the medicine box into the hospital information system (NIS) by scanning the patient's ID card, linking the medicine box to the patient's electronic medical record (EMR);

[0009] The patient wears a health monitoring device to check his health status. The device is connected to the smart medicine box via Wi-Fi. The system automatically detects the device connection status and displays a successful connection prompt on the screen of the smart medicine box.

[0010] After successful connection, the smart medicine box displays the patient's prescription, medical personnel, and designated contacts on the screen;

[0011] The smart medicine box synchronizes data frequently with the hospital information system NIS via Wi-Fi.

[0012] As a preferred solution of the intelligent oral medicine box system for hospitalized patients of the present invention, wherein: the automatic distribution of medicines according to the patient's medicine information means that the pharmacist first registers each batch of newly arrived medicines into the warehouse, including scanning the barcode of the medicine and entering the name, batch number and expiration date of the medicine;

[0013] The robotic arm grabs the specified medicine through the prescription provided by the smart medicine box and dispenses it into the patient's medicine box according to the doctor's instructions;

[0014] Once the medication is dispensed, the smart pill box automatically records the dispensing and gives it to the patient after the pharmacist checks and confirms it is correct;

[0015] The data of the allocated smart medicine box is synchronously updated to the hospital information system NIS, including information on drug type, dosage and medication time.

[0016] As a preferred solution of the intelligent oral medicine box system for hospitalized patients of the present invention, wherein: the audio prompting of the patient to take medicine refers to automatically setting the reminder time for taking medicine according to the prescription information of the patient's intelligent medicine box;

[0017] Use audible and visual signals for reminders based on patient preference;

[0018] If the patient does not take out the medicine within 5 minutes after the set time, the smart medicine box will activate a vibration reminder. If it is still not taken away within 30 minutes, the smart medicine box will automatically notify medical staff and designated contacts.

[0019] As a preferred solution of the intelligent oral medicine box system for hospitalized patients of the present invention, wherein: the use of the intelligent medicine box to provide patients with medication education materials means that when the patient opens the intelligent medicine box and takes medicine, the medicine box touch screen automatically displays relevant medication education content, and the medication education content includes drug instructions, usage guidelines, precautions and side effect information, and the medicine box touch screen displays the materials in a multimedia form;

[0020] After presenting medication education content, the smart pill box provides interactive Q&A, where patients can ask questions and view answers via the touch screen;

[0021] After the patient takes the medicine, the smart medicine box prompts the patient to enter medication feedback. The preset options are displayed on the touch screen of the medicine box. The patient provides feedback by selecting preset options and manually entering text, and the smart medicine box automatically records the feedback information.

[0022] As a preferred solution of the intelligent oral medicine box system for hospitalized patients of the present invention, wherein: the collecting of health monitoring device data worn by the patient through the intelligent medicine box and constructing a multivariate regression model to analyze the monitoring data means that the intelligent medicine box collects the health data of the patient and performs standardized processing;

[0023] Collect public health data, standardize it, and generate MR training sets;

[0024] Construct a multivariate regression model to analyze monitoring data;

[0025] Use the MR training set to train the multivariate regression model, and use the loss function and Adam optimizer to iteratively optimize the model parameters;

[0026] The smart medicine box uses the trained multivariate regression model to analyze the standardized collected patient health data to obtain the health index F.

[0027] 8. As a preferred solution of the intelligent oral medicine box system for hospitalized patients of the present invention, wherein: the construction of a support vector machine (SVM) model based on the analyzed data for medication adjustment refers to collecting the patient's historical health data for preprocessing to generate an SVM training set;

[0028] Construct a support vector machine SVM model to predict the target health parameter value F′:

[0029] Build the SVM model;

[0030] Gaussian kernel function;

[0031] Use the training set to train the SVM model;

[0032] Use the trained support vector machine SVM model to calculate the current target health parameter value F′;

[0033] Dynamic adjustments are made based on the monitoring data analysis results F and the target health parameter value F'.

[0034] As a preferred solution of the intelligent oral medicine box system for hospitalized patients described in the present invention, the storage of the used data refers to storing the medicine box data, network connection data, hospital information system NIS data, drug information data, distribution records, medication time calculation data, reminder adjustment data, medication education, patient feedback data, health detection equipment data, health assessment data, health detection data and treatment plan adjustment data in the database, and setting security access measures. The database will back up the stored data in the cloud, and regularly perform integrity checks on the stored data and backup data. After the test is completed, an integrity test record is generated and stored synchronously in the database.

[0035] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of the above-mentioned intelligent oral medicine box system for hospitalized patients when executing the computer program.

[0036] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned intelligent oral medicine box system for hospitalized patients.

[0037] The beneficial effects of the present invention are as follows: the present invention analyzes and predicts health data through the multivariate regression model and support vector machine (SVM) model of the smart medicine box system, provides personalized medication guidance and dynamic adjustment strategies, as well as multimedia medication education content display and interactive Q&A, significantly improving patients' medication compliance and knowledge level, ensuring that patients can receive prompts and interventions in time when they fail to take medication on time, and effectively reducing medication risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0039] Figure 1 This is a flow chart of the intelligent oral medicine box system for hospitalized patients.

[0040] Figure 2 This is a schematic diagram of the structure of an intelligent oral medicine box system for hospitalized patients. DETAILED DESCRIPTION

[0041] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0042] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0043] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0044] Example 1

[0045] Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and this embodiment provides an intelligent oral medicine box system for hospitalized patients. The intelligent oral medicine box system for hospitalized patients includes, S1, a device allocation module for allocating patient intelligent medicine boxes;

[0046] Specifically, the distribution of smart medicine boxes to patients means that when registering at the hospital, the reception nurse will distribute a smart medicine box, in which the compartments are equipped with a high-resolution camera and a weight sensor;

[0047] The nurse enters the serial number of the medicine box into the hospital information system (NIS) by scanning the patient's ID card, linking the medicine box to the patient's electronic medical record (EMR);

[0048] Patients wear health monitoring devices to check their health conditions, including heart rate monitors, blood pressure monitors, and blood glucose meters. The devices are connected to the smart medicine box via Wi-Fi. The system automatically detects the device connection status and displays a successful connection prompt on the screen of the smart medicine box.

[0049] After successful connection, the smart medicine box displays the patient's prescription, medical personnel, and designated contacts on the screen;

[0050] The smart medicine box synchronizes data frequently with the hospital information system NIS via Wi-Fi.

[0051] Through high-resolution cameras and weight sensors, the medicine box can accurately monitor the storage and access of medicines, prevent missed or wrong medications, and improve the safety and compliance of medication. The distribution process of the medicine box is closely integrated with the hospital information system to ensure that the usage information of the medicine box can be recorded and updated in real time. By associating the smart medicine box with the patient's EMR, the seamless connection between drug management and patient medical history and treatment plan is achieved. The distribution and use of drugs can be personalized based on the patient's specific medical needs to ensure the scientificity and rationality of medication. By transmitting the data of health monitoring equipment to the smart medicine box in real time, the patient's health status can be monitored in real time, providing dynamic data support for drug management. The real-time monitoring and data transmission mechanism helps medical personnel adjust the treatment plan according to the patient's health status, optimize the effect of medication, and improve the patient's overall health level. Through frequent data synchronization, it is ensured that the drug usage information in the smart medicine box and the patient's health data can be updated to the hospital information system in real time. The real-time data synchronization mechanism not only improves the accuracy and reliability of drug management, but also provides medical personnel with comprehensive and timely patient medication information, which is convenient for dynamic adjustment of the patient's treatment plan and further improves the quality and effect of medical services.

[0052] S2, a drug dispensing and reminder module, which is used to automatically dispense drugs according to the patient's drug information and remind the patient to take the drug through sound and visual signals;

[0053] Furthermore, automatically dispensing drugs based on the patient's drug information means that the pharmacist first registers each batch of newly arrived drugs into the warehouse, including scanning the barcode of the drug and entering the name, batch number and expiration date of the drug;

[0054] The robotic arm grabs the specified medicine through the prescription provided by the smart medicine box and dispenses it into the patient's medicine box according to the doctor's instructions;

[0055] Once the medication is dispensed, the smart pill box automatically records the dispensing and gives it to the patient after the pharmacist checks and confirms it is correct;

[0056] The data of the allocated smart medicine box is synchronously updated to the hospital information system NIS, including information on drug type, dosage and medication time.

[0057] By scanning the barcode, pharmacists can quickly and accurately enter detailed information about the drug, reducing the risk of manual entry errors. Recording the batch number and expiration date of the drug helps with drug traceability and inventory management, prevents the use of expired drugs, and improves the safety and efficiency of drug management. The use of robotic arms for drug grabbing and distribution significantly improves the accuracy and efficiency of drug distribution. The robotic arms can accurately identify prescription information, ensure that each patient receives the correct drug and dosage, and reduce the possibility of human error. The automated drug distribution process shortens drug preparation time, improves the work efficiency of hospital pharmacies, and ensures that patients can use their medications in a timely manner. The smart medicine box automatically Record the drug distribution process and ensure that each step is recorded in detail for easy traceability and management. Pharmacists will check and confirm after the distribution is completed to further ensure the accuracy and safety of drug distribution, enhance the transparency of drug management, and provide patients with higher security. By synchronizing the drug information after distribution to the hospital information system in real time, it is ensured that the patient's medication information can be updated in time for doctors and pharmacists to review at any time. The real-time data synchronization mechanism helps medical staff to fully understand the patient's medication situation, adjust treatment plans in time, and optimize the quality of medical services. Data synchronization can also avoid information lags and errors, and improve the overall drug management level of the hospital.

[0058] Further, reminding the patient to take medicine by audio means automatically setting a reminder time for taking medicine according to the prescription information of the patient's smart medicine box;

[0059] Use audible and visual signals for reminders based on patient preference;

[0060] If the patient does not take out the medicine within 5 minutes after the set time, the smart medicine box will activate a vibration reminder. If it is still not taken away within 30 minutes, the smart medicine box will automatically notify medical staff and designated contacts.

[0061] The system automatically sets the reminder time according to the patient's specific prescription information, avoiding errors or omissions that may be caused by manual settings. The automated reminder setting reduces the workload of medical staff, improves work efficiency, and ensures that each patient can take the medicine on time and in the right amount. The combination of sound reminders and visual signals can ensure that patients with different preferences can receive reminders in time. Especially for patients with hearing or visual impairments, the reminder method can be adjusted to meet their needs. The personalized reminder method improves patient compliance and ensures that the medicine can be taken on time. Vibration reminders, as a supplement to sound and visual signals, enhance the diversity and effectiveness of reminders. Vibration reminders can provide additional reminders when patients do not respond to initial reminders in time, further ensuring that patients take medicine on time. This is especially effective for patients who may not notice the sound or visual reminders in time due to environmental noise or other reasons. If the patient does not take the medicine on time, the system will automatically notify medical staff and designated contacts so that they can take timely measures. This automated notification mechanism not only improves the safety of medication management, but also enhances the monitoring and management of patients' medication by medical staff and family members, reducing the health risks caused by missed medications.

[0062] S3, medication education module, used to provide patients with drug information using smart medicine boxes;

[0063] Furthermore, using the smart medicine box to provide patients with medication education materials means that when the patient opens the smart medicine box and takes medicine, the medicine box touch screen automatically displays relevant medication education content, the medication education content includes medication instructions, usage guidelines, precautions and side effect information, and the medicine box touch screen displays the materials in a multimedia form;

[0064] After presenting medication education content, the smart pill box provides interactive Q&A, where patients can ask questions and view answers via the touch screen;

[0065] After the patient takes the medicine, the smart medicine box prompts the patient to enter medication feedback. The preset options are displayed on the touch screen of the medicine box. The patient provides feedback by selecting preset options and manually entering text, and the smart medicine box automatically records the feedback information.

[0066] The presentation of medication education content in multimedia form makes the information more vivid and easy to understand, which increases patients' understanding and memory. The drug instructions and user guides provide specific medication methods and dosages. Precautions and side effect information help patients understand potential risks and countermeasures, thereby improving the safety and compliance of medication. The interactive question-and-answer function enhances the interactivity between patients and smart medicine boxes, enabling patients to actively participate in the medication education process. Patients can ask specific questions according to their own needs and get instant answers, helping them better understand drug information and usage methods. This interaction not only improves patients' medication knowledge level, but also enhances their confidence in medication and reduces medication risks caused by misunderstandings or uncertainties. The medication feedback recording function enables patients' medication experience and effects to be systematically collected and analyzed. Through preset options and manual input, patients can record their feelings and any adverse reactions after taking the medication in detail. These feedback information not only helps medical staff understand the actual effects and side effects of the drugs and adjust treatment plans in a timely manner, but also provides valuable data support for future drug management and research.

[0067] S4, real-time monitoring and adjustment module, used for the smart medicine box to collect the patient's health data through the health monitoring device, build a multivariate regression model to analyze the health data, and build a support vector machine SVM model based on the analyzed data to adjust the medication;

[0068] Specifically, the smart medicine box collects the data of the health monitoring device worn by the patient, and constructs a multivariate regression model to analyze the monitoring data. The smart medicine box collects the patient's health data and performs standardized processing. The formula is:

[0069]

[0070] Among them, H i is the i-th standardized health data, X i is the i-th health data collected, μ i is the mean of the i-th health data, σ i is the standard deviation of the ith health data;

[0071] Collect public health data, standardize it, and generate MR training sets;

[0072] Construct a multivariate regression model to analyze the monitoring data:

[0073] Use the MR training set to train the multivariate regression model to obtain the regression coefficient β, and use the least squares method to obtain the coefficient value β that minimizes the prediction error. n ;

[0074] The standardized health data were analyzed using a multiple regression model. The existing multiple regression model formula is:

[0075] F0=β0+β1H1+β2H2+…+β n H n ;

[0076] Where F0 is the health status index calculated by the existing technology, H n is the nth standardized health monitoring data, β n is the regression coefficient, n is the total number of health data;

[0077] The improved multiple regression model formula is:

[0078]

[0079] Where F is the health status indicator, n is the total number of health data, ρ i is the dynamic weight of health data, τ i is the time decay term, tanh(H i ) is the nonlinear hyperbolic tangent function, m is the total number of health data processed by the composite logarithmic function, γ j is the regression coefficient in the composite logarithmic function, which indicates the weight of the jth health data in the logarithmic function and controls the impact of the data on the health status prediction. j ) is a compound logarithmic function used to capture nonlinear relationships in health data;

[0080] Dynamic weights are designed to meet the different needs of each health data on health status. The importance of health data is not fixed, but changes dynamically with the change of data. Dynamic weights can automatically adjust the impact of each data on the model results according to the different values ​​of each data. The Sigmoid function is used to dynamically adjust the weights in order to nonlinearly allocate weights according to different data values. This nonlinear function allows the impact of data on the results to be smoothly adjusted as the value changes, avoiding the limitations of linear weights. The simple regression coefficient of the linear weight allocation method is relatively rigid when processing health data and cannot flexibly adapt to the dynamic changes of health data. Using linear regression or fixed weight methods will cause the model to be unable to adapt to different data dimensions. Due to the different impacts on health conditions, the flexibility and predictive accuracy of the model are greatly reduced. The effectiveness of health data often gradually weakens over time. This is especially important in dynamic medical monitoring scenarios. The introduction of time decay terms can ensure that the impact of data gradually weakens over time, ensuring that the model is more sensitive to the latest data. The exponential decay function is a very effective means to smoothly reduce the impact of historical data on the results without suddenly abandoning historical data. If the time decay term is not used, or a linear decreasing function is used, it cannot effectively reflect the changes in data over time. The linear decreasing function cannot naturally handle the exponential changes over time, and directly abandoning historical data will also lead to data loss and reduced model accuracy. The time decay term enhances the model. The timeliness and accuracy of the model ensure that the model makes the best prediction based on the latest data. Health data often exhibits nonlinear characteristics. The hyperbolic tangent function can handle this nonlinear relationship well. It can effectively capture extreme value changes in the data and smooth the nonlinear fluctuations of the data. The nonlinear characteristics of the hyperbolic tangent function enable it to limit abnormally high or low data to a certain range, avoiding excessive impact of extreme data on the model, while retaining important trends in the data. When using linear functions or polynomial functions to process nonlinear data, extreme values ​​may be over-amplified, resulting in model instability, and high-order terms of polynomials may cause data oscillation or overfitting. In contrast, the smooth characteristics and limited output range of the hyperbolic tangent function are more suitable for processing health data. Compared with traditional linear or polynomial regression models, the hyperbolic tangent function can capture more complex health data patterns and effectively handle outliers and extreme changes. The nonlinear processing method brings higher prediction accuracy to the model. Sometimes extreme values ​​will appear in health data, and directly using these extreme values ​​may lead to unstable model predictions. The compound logarithmic function is used to smooth extreme data and capture nonlinear trends. The compound logarithmic function avoids numerical instability caused by direct logarithmic processing of extreme data by introducing a fractional structure. At the same time, it can compress excessively large data, making the model more robust when dealing with extreme values. Simple logarithmic functions or polynomial processing methods may produce excessive fluctuations when facing extreme values.The compound logarithmic function can smooth extreme values ​​and capture their nonlinear characteristics, which prevents the model from being overly sensitive to large data values. Compared with traditional logarithmic functions or other processing methods, the compound logarithmic function has a stronger smoothing effect and can effectively handle extreme fluctuations in health data. If dynamic weights, time decay and nonlinear processing are not combined, traditional linear regression models will not be able to handle the complex patterns and changing trends in health data, resulting in insufficient prediction accuracy. By combining dynamic weights, time decay and nonlinear processing, this compound multivariate regression model significantly improves the existing linear regression model and can better capture the complex relationships, nonlinear changes and time effects in health data.

[0081] Use the Sigmoid function to calculate the standardized dynamic weight ρ of health data i , the formula is:

[0082]

[0083] Among them, α is a parameter that controls the speed of weight adjustment and the degree of nonlinearity of dynamic weight;

[0084] Calculate the time decay term τ using the time decay formula i , the formula is:

[0085] τ i =e -λt ;

[0086] Where λ is the decay rate, which is used to control the decay speed of data impact, and t is the time of collecting health data, which represents the time difference from the current time;

[0087] Nonlinear hyperbolic tangent function, the formula is:

[0088]

[0089] The compound logarithmic function is:

[0090]

[0091] Among them, H j is the jth standardized health data;

[0092] During the data analysis process, the system will compare the patient’s current health data with historical health data and calculate the trend of health status changes;

[0093] The smart medicine box uses the trained multivariate regression model to analyze the standardized collected patient health data to obtain the health index F.

[0094] The smart medicine box standardizes the collected health data to make data from different sources comparable, eliminates the data dimension differences between different devices, and thus improves the accuracy of data analysis. A multivariate regression model is constructed to analyze the standardized health data. The data of multiple health monitoring devices can be comprehensively considered to form an overall health status indicator. The smart medicine box uses the trained multivariate regression model to analyze the patient's health data in real time to obtain a health indicator. This indicator can reflect the patient's health change trend in real time, help doctors to detect abnormal conditions in time, intervene and adjust treatment plans. This real-time monitoring and evaluation method can improve the patient's health management level and reduce the risk of emergencies. Based on the analysis results of the multivariate regression model, the smart medicine box can provide patients with personalized health management suggestions. By analyzing the changing trend of the health indicator F, the smart medicine box can remind patients to make necessary lifestyle adjustments or drug adjustments to improve the personalization and refinement of health management. The health data collected and analyzed by the smart medicine box can not only provide patients with real-time health monitoring and management services, but also provide doctors with data-based decision support. By analyzing the health data of a large number of patients, doctors can discover potential health problems, optimize treatment plans, and improve the quality and efficiency of medical services.

[0095] Furthermore, constructing a support vector machine (SVM) model based on the analyzed data for medication adjustment refers to collecting the patient's historical health data for preprocessing and generating an SVM training set;

[0096] Construct a support vector machine SVM model to predict the target health parameter value F′:

[0097] Construct the SVM model, the formula is:

[0098]

[0099] where ω i is the model parameter, n is the number of health data, K(H i ,H j ) is the Gaussian kernel function, H j is the jth standardized health data, b is the bias term;

[0100] Gaussian kernel function, the formula is:

[0101]

[0102] Where θ is the bandwidth parameter of the Gaussian kernel function, which controls the scale of the kernel function, r is the adjustment factor, which controls the influence of the cosine function in capturing high-order similarities, and ||H i -H j || 2 For health data iand H j The Euclidean distance between

[0103] The Gaussian kernel function can effectively process nonlinear data. The core is to measure similarity by calculating the Euclidean distance between data points. It is especially suitable for high-dimensional data space, but a simple Gaussian kernel may not be enough to capture more subtle nonlinear patterns when dealing with very complex data relationships. The interaction between multidimensional health parameters and the simple similarity measurement of the Gaussian kernel may not fully reflect the deep-level associations of the data. It is necessary to enhance the adaptability of the model to health data through complex kernel function design. By introducing the cosine function, more complex relationships between data can be captured. Multidimensional health data analysis provides more powerful tools that can effectively handle more complex nonlinear patterns and similarities. Compared with the existing simple Gaussian kernel, this scheme has stronger distinguishing ability and flexibility when processing health data. The kernel function depends on the distance between health data, and the most common distance metric is the Euclidean distance. Its advantage lies in its simplicity and stability. Especially in health data, the shorter the distance, the higher the similarity of the data, and the longer the distance, the greater the difference. When processing multidimensional data, directly using the Euclidean distance may not fully capture the nonlinear changes of the data, especially some dimensions in health data may be more decisive than other dimensions. Simple Euclidean distance can effectively capture the nonlinear changes of the data. The Euclidean distance metric may cause the model to ignore important health data features. The Euclidean distance must be further processed nonlinearly and combined with the cosine function to capture the complex patterns in the health data. The expanded formula processes the Euclidean distance as a fraction, so that when the data differences are large, the kernel function can still smoothly handle these changes, avoiding the oversensitivity or instability of the model caused by the simple use of distance values. The similarity in health data is not only determined by linear distance, but also includes complex nonlinear interactions. The introduction of nonlinear modulation of the cosine function can capture the high-order nonlinear relationship of health data. This improvement improves the expressive power of the kernel function, so that the model can have better adaptability and distinguishing ability to the nonlinear characteristics of the data. The use of fractional structure can avoid data distortion caused by simple multiplication or addition, and ensure that the model can balance changes in different dimensions. The combination of Gaussian kernel and cosine function provides a multi-level similarity measure and enhances the expressive power of the kernel function. This improvement not only improves the model's ability to capture local patterns, but also can handle high-order nonlinear interactions. The complexity of this kernel function greatly improves the accuracy of health data analysis, especially in the application scenarios of multidimensional and nonlinear data.

[0104] Use the SVM training set to train the support vector machine (SVM) model, and use the loss function and Adam optimizer to iteratively optimize the model parameters;

[0105] Use the trained support vector machine SVM model to calculate the current target health parameter value F′, the formula is:

[0106] F′=f(H1,H2,…,H n );

[0107] Parameter δ is set through historical data and clinical experience;

[0108] According to the monitoring data analysis result F and the target health parameter value F', the formula is:

[0109] D h =D l +δ(F′-F);

[0110] Where D h is the adjusted drug dose, D l is the current drug dosage, δ is the drug dosage adjustment coefficient, and F' is the target health parameter value.

[0111] The data preprocessing step includes noise removal, standardization and feature extraction to ensure the quality and consistency of the data. This step can improve the training effect of the SVM model and enhance the generalization ability of the model. The preprocessed training set is used to build the SVM model to predict the target health parameter value. After the model training is completed, the trained SVM model is used to analyze the current health data and calculate the current target health parameter value. According to the monitoring data analysis result F and the target health parameter value, the drug dosage is dynamically adjusted. Based on the prediction results of the SVM model, the smart medicine box can provide each patient with a personalized treatment plan. By analyzing historical data and real-time monitoring data, the SVM model can capture individual differences and provide accurate drug dosage adjustment recommendations, thereby improving the effectiveness and safety of treatment. The support vector machine model can avoid the problem of over-medication or under-medication through accurate prediction and analysis. The method of dynamically adjusting drug dosage can respond to the patient's health changes in a timely manner, reduce the risk of drug side effects, and improve the safety of medication. The smart medicine box collects and analyzes the patient's health data and builds a support vector machine model, which can provide doctors with data-based decision support. Doctors can optimize treatment plans and improve the quality and efficiency of medical services based on the model's prediction results. The smart medicine box uses the SVM model for real-time monitoring and feedback, which can promptly detect patients' health abnormalities and make interventions and adjustments. This real-time monitoring and feedback mechanism helps to improve patients' health management level and reduce the risk of emergencies.

[0112] S5, a storage module, used to store the used data;

[0113] Furthermore, storing the used data means storing the medicine box data, network connection data, hospital information system NIS data, drug information data, distribution records, medication time calculation data, reminder adjustment data, medication education, patient feedback data, health testing equipment data, health assessment data, health testing data and treatment plan adjustment data in the database, and setting security access measures. The database will back up the stored data in the cloud, and regularly perform integrity checks on the stored data and backup data. After the test is completed, the integrity test record is generated and stored synchronously in the database.

[0114] The data used is stored in the database. This centralized management method is conducive to the unified management and maintenance of data and ensures the integrity and consistency of data. By setting security access measures, it is ensured that only authorized personnel can access and operate the data in the database. Security measures may include user identity authentication, access control, data encryption, etc., which can effectively prevent data leakage and unauthorized access and improve data security. The database will back up the stored data in the cloud to ensure that the data can still be restored through cloud backup in the event of local storage device failure or other disasters. This backup mechanism provides a highly reliable data protection measure to ensure the persistence and recoverability of data. The integrity of the stored data and the backup data is regularly tested to ensure that the data has not been tampered with or damaged during storage and transmission. The integrity test verifies the data by generating a checksum or hash value. Once a data integrity problem is found, repair measures can be taken in time to ensure the reliability of the data. After the test is completed, an integrity test record is generated and stored synchronously in the database. The integrity test record provides detailed logs and audit trails for data management, which is convenient for subsequent queries and audits, and helps to make data management transparent and standardized.

[0115] Example 2

[0116] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0117] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0118] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0119] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

Claims

1. An intelligent oral medicine box system for hospitalized patients, characterized by: include, A device dispensing module for dispensing patient smart pill boxes; The drug dispensing and reminder module is used to automatically dispense drugs according to the patient's drug information and remind the patient to take the drug through sound and visual signals; Medication education module, used to provide patients with medication information using smart pill boxes; Real-time monitoring and adjustment module, which is used by the smart medicine box to collect the patient's health data through health monitoring equipment, build a multivariate regression model to analyze the health data, and build a support vector machine (SVM) model based on the analyzed data to adjust medication; The storage module is used to store the used data.

2. The intelligent oral medicine box system for hospitalized patients as claimed in claim 1, characterized in that: The distribution of the patient's smart medicine box means that the reception nurse will distribute a smart medicine box when registering at the hospital, and the compartment of the medicine box is equipped with a high-resolution camera and a weight sensor; The nurse enters the serial number of the medicine box into the hospital information system (NIS) by scanning the patient's ID card, linking the medicine box to the patient's electronic medical record (EMR); The patient wears a health monitoring device to check his health status. The device is connected to the smart medicine box via Wi-Fi. The system automatically detects the device connection status and displays a successful connection prompt on the screen of the smart medicine box. After successful connection, the smart medicine box displays the patient's prescription, medical personnel, and designated contacts on the screen; The smart medicine box synchronizes data frequently with the hospital information system NIS via Wi-Fi.

3. The intelligent oral medicine box system for hospitalized patients as claimed in claim 2, characterized in that: The automatic distribution of drugs according to the patient's drug information means that the pharmacist first registers each batch of newly arrived drugs into the warehouse, including scanning the barcode of the drug and entering the name, batch number and expiration date of the drug; The robotic arm grabs the specified medicine through the prescription provided by the smart medicine box and dispenses it into the patient's medicine box according to the doctor's instructions; Once the medication is dispensed, the smart pill box automatically records the dispensing and gives it to the patient after the pharmacist checks and confirms it is correct; The data of the allocated smart medicine box is synchronously updated to the hospital information system NIS, including the type of medicine, dosage and medication time information.

4. The intelligent oral medicine box system for hospitalized patients as claimed in claim 3, characterized in that: Prompting the patient to take medicine by audio means automatically setting a reminder time for taking medicine according to the prescription information of the patient's smart medicine box; Use audible and visual signals for reminders based on patient preference; If the patient does not take out the medicine within 5 minutes after the set time, the smart medicine box will activate a vibration reminder. If it is still not taken away within 30 minutes, the smart medicine box will automatically notify medical staff and designated contacts.

5. The intelligent oral medicine box system for hospitalized patients as claimed in claim 4, characterized in that: The use of the smart medicine box to provide patients with medication education materials means that when the patient opens the smart medicine box and takes medicine, the medicine box touch screen automatically displays relevant medication education content, and the medication education content includes drug instructions, usage guidelines, precautions and side effect information, and the medicine box touch screen displays the materials in a multimedia form; After presenting medication education content, the smart pill box provides interactive Q&A, where patients can ask questions and view answers via the touch screen; After the patient takes the medicine, the smart medicine box prompts the patient to enter medication feedback. The preset options are displayed on the touch screen of the medicine box. The patient provides feedback by selecting preset options and manually entering text, and the smart medicine box automatically records the feedback information.

6. The intelligent oral medicine box system for hospitalized patients as claimed in claim 5, characterized in that: The collecting of health monitoring device data worn by the patient through the smart medicine box and constructing a multivariate regression model to analyze the monitoring data means that the smart medicine box collects the patient's health data and performs standardized processing; Collect public health data, standardize it, and generate MR training sets; Construct a multivariate regression model to analyze monitoring data; Use the MR training set to train the multivariate regression model, and use the loss function and Adam optimizer to iteratively optimize the model parameters; The smart medicine box uses the trained multivariate regression model to analyze the standardized collected patient health data to obtain the health index F.

7. The intelligent oral medicine box system for hospitalized patients as claimed in claim 6, characterized in that: The constructing of a support vector machine (SVM) model based on the analyzed data for medication adjustment refers to collecting the patient's historical health data for preprocessing to generate an SVM training set; Construct a support vector machine SVM model to predict the target health parameter value F': Build the SVM model; Gaussian kernel function; Use the training set to train the SVM model; Use the trained support vector machine SVM model to calculate the current target health parameter value F′; Dynamic adjustments are made based on the monitoring data analysis results F and the target health parameter value F'.

8. The intelligent oral medicine box system for hospitalized patients as claimed in claim 7, characterized in that: The storing of the data to be used refers to storing the medicine box data, network connection data, hospital information system NIS data, drug information data, distribution records, medication time calculation data, reminder adjustment data, medication education, patient feedback data, health testing equipment data, health assessment data, health testing data and treatment plan adjustment data in the database, and setting security access measures. The database will back up the stored data in the cloud, and regularly perform integrity checks on the stored data and backup data. After the test is completed, an integrity test record is generated and stored synchronously in the database.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that when the processor executes the computer program, the steps of the intelligent oral medicine box system for hospitalized patients according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent oral medicine box system for hospital inpatients according to any one of claims 1 to 8 are implemented.

Images