Robot with health file management and medication management functions

Abstract

The invention discloses a robot with health file management and medication management functions, which belongs to the technical field of health management, and comprises a hardware part and a software part, the hardware part comprises a processor, a memory, a communication module, a display module, a voice interaction module and a sensor module. Medical data in different formats are analyzed through a natural language processing technology, and meanwhile, when a personalized medication plan is made according to doctor prescription information, a time sequence prediction model is introduced, so that the efficiency and accuracy of health file management are improved, real-time sharing and dynamic updating of health information are realized, and the user experience is improved. The treatment risk caused by nonstandard medication of the patient is reduced, the treatment effect is improved, intelligent interaction experience is provided, and health management of the patient is facilitated.

Description

Technical Field

[0001] The present invention belongs to the technical field of health management, and particularly relates to a robot with functions of health record management and medication management. Background Art

[0002] With the improvement of people's health awareness and the development of medical informatization, the demand for personal health management is increasing day by day. Traditional health record management and medication management mainly rely on manual operations, with low efficiency and prone to errors. For example, it is difficult to integrate the health information of patients in different medical institutions, resulting in doctors being unable to fully understand the patient's condition; during the medication process, patients may have poor treatment effects due to forgetting the medication time, dosage, etc. Although some existing health management software can solve some problems to a certain extent, they lack intelligent interaction and real-time reminder functions. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to overcome the above-mentioned disadvantages of the prior art and provide a robot with functions of health record management and medication management.

[0004] The technical solution adopted to solve the above technical problem is: a robot with functions of health record management and medication management, including a hardware part and a software part. The hardware part includes a processor, a memory, a communication module, a display module, a voice interaction module, and a sensor module. The processor, as the core component, is responsible for data processing and instruction execution; the memory is used to store health record data, medication information, etc.; the communication module realizes data transmission with external devices; the display module is used to display health information and medication reminders; the voice interaction module facilitates natural language interaction between patients and the robot; the sensor module can collect the vital sign data of patients;

[0005] The software part includes a health record management system and a medication management system.

[0006] Furthermore, the health record management system uses natural language processing technology (NLP) to parse medical data in different formats, classifies and stores the data using a text classification algorithm, and uses a support vector machine (SVM) algorithm. The specific formula is as follows:

[0007]

[0008] Where x is the input data, x i is the training sample, y i is the sample label, α i is the Lagrange multiplier, K(x i , x) is the kernel function, and b is the bias term. Through this algorithm, the text information of the patient's symptom description and inspection report can be accurately classified, facilitating efficient storage and rapid retrieval.

[0009] Furthermore, in terms of providing decision support for doctors through data analysis, the Apriori algorithm for association rule mining is adopted, and the specific formula is as follows:

[0010]

[0011] Among them, support(X∪Y) represents the support degree that item sets X and Y appear simultaneously, that is, the proportion of transactions containing X and Y in the total transaction set. σ(X∪Y) is the number of transactions containing item sets X and Y, and N is the number of the total transaction set;

[0012]

[0013] Among them, represents the confidence degree of inferring Y from X, that is, the proportion of transactions that also contain Y among the transactions containing X, which is used to measure the reliability of the rule. support(X) represents the support degree that item set X appears;

[0014]

[0015] represents the lift, which is used to evaluate the influence degree of the appearance of X on the appearance of Y. If the lift is greater than 1, it indicates a positive correlation between X and Y; equal to 1 indicates that the two are independent of each other; less than 1 indicates a negative correlation.

[0016] Furthermore, when formulating a personalized medication plan according to the doctor's prescription information, the medication management system introduces a time series prediction model and adopts ARIMA(p, d, q), and the specific formula is as follows:

[0017] Φ p (B)(1 - B) d Y t =Θ q (B)∈ t

[0018] Among them, Y t represents the time series data, which can be the historical medication data of the patient and the sequence values of the physical recovery indicators (such as body temperature changes and blood pressure fluctuations) over time here. B represents the backward shift operator, and BY t =Y t-1 , that is, shifting the time series data backward by one time step. Φ p (B) represents the autoregressive polynomial, Among them is the autoregressive coefficient, p is the autoregressive order, which represents the number of past time step data included in the model and is used to describe the linear relationship between the current value and the past values. (1 - B) dDenotes the difference operator, where d is the order of difference, used to transform a non-stationary time series into a stationary time series. By subtracting consecutive terms of the time series, trends and seasonality in the data are eliminated. Θ q (B) denotes the moving average polynomial, Θ q (B) = 1 - θ 1 B - θ 2 B 2 - … - θ p B p , where θ i is the moving average coefficient, and q is the order of moving average, representing the number of past prediction errors included in the model, used to correct prediction errors. ∈ t denotes the white noise sequence, which is a random error term with a mean of 0 and a constant variance, representing the random fluctuations that cannot be explained by the model;

[0019] By combining the patient's historical medication data and factors related to their physical recovery using this model, the optimal medication time and dosage adjustments are predicted, and a medication plan that better meets the patient's needs is developed.

[0020] Through the above technical solution, a time series prediction model is introduced, thereby improving the efficiency and accuracy of health record management, realizing real-time sharing and dynamic updating of health information, reducing the treatment risks caused by non-standard medication for patients, improving the treatment effect, providing an intelligent interaction experience, and facilitating patients' health management.

[0021] Furthermore, when the medication management system records the patient's medication situation and feeds it back to the doctor and the patient, it uses data fusion technology to fuse and analyze the medication information confirmed by the voice interaction module and the patient behavior data detected by the sensor, and applies the Dempster - Shafer evidence theory, including the following specific formula:

[0022]

[0023] Among them, m(A) represents the fused information function, indicating the degree of trust in event A, m 1 (X i ) and m 2 (Y i ) are the information functions of different data sources respectively, X i and Y i are the sets of events supported by different data sources, represents the sum of the products of the belief functions that simultaneously support event A in the two data sources, represents the sum of the products of the belief functions where the events supported by the two data sources are contradictory, It is used to process conflict information, ensure that the fused belief function is within a reasonable range, improve the accuracy and reliability of the medication record through this theory, and provide more comprehensive data support for doctors to evaluate the treatment effect.

[0024] Furthermore, when the patient uses it for the first time, personal basic information and health record authorization information are input through the voice interaction module and the display module. The robot obtains the patient's health record data through the communication module and stores it in the local memory. After the doctor issues a prescription according to the patient's condition, the prescription information is synchronized to the robot's medication management system. The robot reminds the patient to take medicine on time according to the medication plan, and confirms and records when the patient takes the medicine. The patient can query their own health information and medication situation at any time through the voice interaction module.

[0025] The beneficial effects of the present invention are as follows: The present invention analyzes medical data in different formats through natural language processing technology, and at the same time, when formulating a personalized medication plan according to the doctor's prescription information, a time series prediction model is introduced, thereby improving the efficiency and accuracy of health record management, realizing real-time sharing and dynamic updating of health information, reducing the treatment risk caused by the patient's irregular medication, improving the treatment effect, providing an intelligent interaction experience, and facilitating the patient to manage their health. Brief Description of the Drawings

[0026] Figure 1 It is the system structure diagram of the present invention. Detailed Embodiment

[0027] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below with reference to 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 used to limit the present invention.

[0028] As Figure 1 shown, a robot with health record management and medication management functions in this embodiment includes a hardware part and a software part. The hardware part includes a processor, a memory, a communication module, a display module, a voice interaction module, and a sensor module. The processor, as the core component, is responsible for data processing and instruction execution; the memory is used to store health record data, medication information, etc.; the communication module realizes data transmission with external devices; the display module is used to display health information and medication reminders; the voice interaction module facilitates natural language interaction between the patient and the robot; the sensor module can collect the patient's vital sign data;

[0029] The software part includes a health record management system and a medication management system.

[0030] Furthermore, the health record management system uses natural language processing technology (NLP) to parse medical data in different formats, classifies and stores the data using text classification algorithms, and uses the support vector machine (SVM) algorithm. The specific formula is as follows:

[0031]

[0032] Among them, x is the input data, x i is the training sample, y i is the sample label, α i is the Lagrange multiplier, K(x i , x) is the kernel function, b is the bias term. Through this algorithm, the text information of the patient's symptom description and examination report can be accurately classified, facilitating efficient storage and rapid retrieval.

[0033] Furthermore, in terms of providing decision support for doctors in data analysis, the Apriori algorithm for association rule mining is adopted. The specific formula is as follows:

[0034]

[0035] Among them, support(X∪Y) represents the support degree that item sets X and Y appear simultaneously, that is, the proportion of transactions containing X and Y in the total transaction set. σ(X∪Y) is the number of transactions containing item sets X and Y, and N is the number of the total transaction set;

[0036]

[0037] Among them, represents the confidence degree of inferring Y from X, that is, the proportion of transactions containing Y among the transactions containing X, which is used to measure the reliability of the rule. support(X) represents the support degree that item set X appears;

[0038]

[0039] represents the lift, which is used to evaluate the influence degree of the appearance of X on the appearance of Y. If the lift is greater than 1, it means there is a positive correlation between X and Y; equal to 1 means the two are independent; less than 1 means a negative correlation.

[0040] Furthermore, when formulating a personalized medication plan according to the doctor's prescription information, the medication management system introduces a time series prediction model and adopts ARIMA(p, d, q). The specific formula is as follows:

[0041] Φ p (B)(1 - B) d Y t =Θ q (B)∈ t

[0042] Among them, Y t represents time series data, which can be the patient's historical medication data here, and the sequence values of the patient's physical recovery indicators (such as body temperature changes, blood pressure fluctuations) over time. B represents the backward shift operator, and BY t = Y t-1 , that is, shifting the time series data backward by one time step. Φ p (B) represents the autoregressive polynomial, where are the autoregressive coefficients, p is the autoregressive order, which represents the number of past time step data included in the model and is used to describe the linear relationship between the current value and the past values. (1 - B) d represents the difference operator, d is the difference order, which is used to transform a non - stationary time series into a stationary time series by performing the operation of subtracting each period of the time series to eliminate the trend and seasonality in the data. Θ q (B) represents the moving average polynomial, Θ q (B)= 1 - θ 1 B - θ 2 B 2 -… - θ p B p , where θ i are the moving average coefficients, q is the moving average order, which represents the number of past prediction errors included in the model and is used to correct the prediction errors. ∈ t represents the white noise sequence, which is a random error term with a mean of 0 and a constant variance, representing the random fluctuations that cannot be explained by the model;

[0043] By combining this model with the patient's historical medication data and physical recovery factors, the optimal medication time and dose adjustment are predicted, and a medication plan that better meets the patient's needs is formulated.

[0044] Through the above - mentioned technical solution, a time series prediction model is introduced, thereby improving the efficiency and accuracy of health record management, realizing the real - time sharing and dynamic update of health information, reducing the treatment risks caused by non - standard medication of patients, improving the treatment effect, providing an intelligent interaction experience, and facilitating patients' health management.

[0045] Furthermore, when the medication management system records the patient's medication situation and feeds it back to the doctor and the patient, it adopts data fusion technology to fuse and analyze the medication information confirmed by the voice interaction module and the patient behavior data detected by the sensor, and uses the Dempster - Shafer evidence theory, including the following specific formula:

[0046]

[0047] Among them, m(A) represents the fused information function, indicating the degree of trust in event A, and m 1 (X i ) and m 2 (Y i ) are the information functions of different data sources respectively. X i and Y i are the sets of events supported by different data sources. represents the sum of the products of the belief functions that simultaneously support event A in the two data sources. represents the sum of the products of the belief functions whose supported events are contradictory in the two data sources. represents the processing of conflicting information to ensure that the fused belief function is within a reasonable range, improving the accuracy and reliability of the medication record through this theory, and providing more comprehensive data support for doctors to evaluate the treatment effect.

[0048] Further, when the patient uses it for the first time, personal basic information and health record authorization information are input through the voice interaction module and the display module. The robot obtains the patient's health record data through the communication module and stores it in the local memory. After the doctor issues a prescription according to the patient's condition, the prescription information is synchronized to the robot's medication management system. The robot reminds the patient to take medicine on time according to the medication plan, and confirms and records when the patient takes medicine. The patient can query their own health information and medication situation at any time through the voice interaction module.

[0049] In the hardware part, according to the design requirements, components such as the processor, memory, communication module, display module, voice interaction module, and sensor module are assembled to ensure correct electrical connections between the components. A high-performance ARM architecture processor is selected, and a large-capacity solid-state drive is used as the memory to ensure the high efficiency of data storage and reading; the communication module uses 5G wireless communication technology to achieve high-speed data transmission with external devices.

[0050] The above is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention.

Claims

1. A robot with health record management and medication management functions, characterized in that: It includes hardware and software parts. The hardware part includes a processor, a memory, a communication module, a display module, a voice interaction module, and a sensor module. The processor is the core component responsible for data processing and instruction execution. The memory is used to store health record data, medication information, etc.; the communication module realizes data transmission with external devices; the display module is used to display health information and medication reminders; The voice interaction module facilitates natural language interaction between patients and robots; the sensor module can collect patients’ vital signs data; The software part includes a health record management system and a medication management system.

2. A robot with health record management and medication management functions according to claim 1, characterized in that: The health record management system uses natural language processing technology (NLP) to parse medical data in different formats, uses text classification algorithm to classify and store data, and uses support vector machine (SVM) algorithm. The specific formula is as follows: Among them, x is the input data, x i is the training sample, y i is the sample label, α i is the Lagrange multiplier, K(x i , x) is the kernel function, b is the bias term, and the algorithm can accurately classify the text information of the patient's symptom description and examination report, which is convenient for efficient storage and fast retrieval.

3. A robot with health record management and medication management functions according to claim 2, characterized in that: In terms of data analysis to provide decision support for doctors, the association rule mining algorithm Apriori is used. The specific formula is as follows: Where support(X∪Y) represents the support of the itemsets X and Y appearing at the same time, that is, the proportion of transactions containing X and Y in the total transaction set, σ(X∪Y) is the number of transactions containing itemsets X and Y, and N is the number of total transaction sets; in, It indicates the confidence of inferring I from X, that is, the proportion of transactions that also contain I among the transactions that contain X, which is used to measure the reliability of the rule. support(X) indicates the support of the occurrence of item set X. It represents the lift, which is used to evaluate the influence of the appearance of X on the appearance of I. If the lift is greater than 1, it means that there is a positive correlation between X and I; equal to 1 means that the two are independent of each other; less than 1 means negative correlation.

4. A robot with health record management and medication management functions according to claim 3, characterized in that: When the medication management system formulates a personalized medication plan based on the doctor's prescription information, a time series prediction model is introduced, using ARIMA (p, d, q). The specific formula is as follows: F p (B)(1-B) d Y t =Θ q (B)∈ t Among them, Y t represents time series data, which can be the patient's historical medication data or the sequence value of physical recovery indicators (such as temperature changes and blood pressure fluctuations) changing over time. B represents the backward shift operator, and BY t =Y t-1 , that is, moving the time series data back one time step, Φ p (B) represents the autoregressive polynomial, in is the autoregressive coefficient, p is the autoregressive order, which indicates the number of past time step data included in the model and is used to describe the linear relationship between current values ​​and past values, (1-B) d represents the difference operator, d is the difference order, which is used to transform the non-stationary time series into a stationary time series, and eliminate the trend and seasonality in the data by subtracting the time series period by period. q (B) represents the moving average polynomial, Θ q (B) = 1 - θ1B - θ2B 2 -…-θ p B p , where θ i is the moving average coefficient, q is the moving average order, which indicates the number of past forecast errors contained in the model and is used to correct the forecast error, ∈ t Represents a white noise sequence, which is a random error term with a mean of 0 and a constant variance, representing random fluctuations that cannot be explained by the model; This model combines the patient's historical medication data and physical recovery factors to predict the optimal medication time and dosage adjustment, and develop a medication plan that better suits the patient's needs.

5. A robot with health record management and medication management functions according to claim 4, characterized in that: When recording the patient's medication status and feeding back to the doctor and the patient, the medication management system uses data fusion technology to fuse and analyze the medication information confirmed by the voice interaction module and the patient behavior data detected by the sensor, and uses the Dempster-Shafer evidence theory, including the following specific formulas: Among them, m(A) represents the fused information function, which indicates the degree of trust in event A, and m1(X i ) and m2(Y i ) are information functions of different data sources, X i and Y i It is a collection of events supported by different data sources. It represents the sum of the products of the credibility functions of event A in two data sources. It represents the sum of the products of the contradictory belief functions of the events supported by the two data sources. It is used to process conflicting information and ensure that the fused reliability function is within a reasonable range. This theory can improve the accuracy and reliability of medication records and provide more comprehensive data support for doctors to evaluate treatment effects.

6. A robot with health record management and medication management functions according to claim 5, characterized in that: When the patient uses it for the first time, he / she enters personal basic information and health record authorization information through the voice interaction module and display module. The robot obtains the patient's health record data through the communication module and stores it in the local storage. After the doctor prescribes according to the patient's condition, the prescription information is synchronized to the robot's medication management system. The robot reminds the patient to take the medicine on time according to the medication plan, and confirms and records when the patient takes the medicine. The patient can check his / her health information and medication status at any time through the voice interaction module.

Images