Ward health education robot

Abstract

The invention discloses a ward health education robot, and the robot is internally provided with a health education platform, and the health education platform comprises a personalized education module which is responsible for generating a personalized guidance scheme according to the personal information of a patient; the intelligent question and answer module is responsible for intelligently recommending related health knowledge and matters needing attention; the medicine use and treatment nursing management module is responsible for recording the medicine use condition of the patient and generating a schedule about the patient; the intelligent navigation and hospital guide module is responsible for realizing autonomous navigation of the patient from a ward to a designated place; the physical sign monitoring and analyzing module is responsible for collecting vital sign data of a patient and generating a health report and early warning information; according to the system, the emotion recognition and interaction optimization module recognizes the emotional state of the patient and adjusts the communication mode and content according to the emotional state, the anxiety and fear emotion of the patient can be relieved, and the confidence of the patient in fighting the disease is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of ward health education, and in particular to a ward health education robot. Background Art

[0002] As an effective alternative therapy for end-stage renal disease, peritoneal dialysis has a long historical background, a solid technical foundation and broad application prospects. With the continuous progress of medical technology and the continuous improvement of medical policies, peritoneal dialysis will play a more important role in the future. As a home treatment method, peritoneal dialysis requires patients and their families to have certain knowledge and skills. Therefore, it is particularly important to conduct systematic health education for patients. The ward health education robot emerged as the times require, aiming to improve the health education level of patients through intelligent means.

[0003] After retrieval, the invention patent with the Chinese patent number CN110293535A discloses a ward health education robot, belonging to the technical field of robots. It includes a body, a moving base is arranged below the body, the moving base is fixedly connected with the body, and a protective component is sleeved outside the moving base. By setting the protective component in the present invention, when the body collides with the wall during movement, the circular protective plate will first contact the wall, giving the protective plate a pressure. When the buffer spring is compressed, a good buffering effect is provided. The wall will give the outer protective plate a lateral force. When the inner lining plate and the sliding block slide on the fixed plate at the same time, the movement direction of the body is changed accordingly, so that the body moves more smoothly, ensuring that the robot can move without dead angles throughout the process. Compared with the existing robot that cannot move when radially colliding with the wall, this design has a simple structure and higher practicability.

[0004] However, during the above use process, by setting a programmable PLC, a display screen and an infrared sensor, the robot provides personalized health knowledge education and introduction of important examination and test results to patients according to the diagnosis. However, it is single and mechanized, lacking affinity and attraction, which is not conducive to the user experience of patients. Therefore, a ward health education robot is proposed. Summary of the Invention

[0005] The purpose of the present invention is to solve the disadvantages existing in the prior art that the ward health education robot is single and mechanized, lacking affinity and attraction, and not conducive to the user experience of patients, and to propose a ward health education robot.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solutions:

[0007] A ward health education robot. Inside the robot, there is a health education platform. The robot is a mobile robot equipped with medical device interfaces such as a high-definition camera, a microphone, a speaker, a touch screen, a body temperature sensor, a sphygmomanometer, a pulse oximeter, etc. The robot is provided with a user interface. The health education platform includes:

[0008] Personalized education module: responsible for generating personalized guidance plans according to the patient's personal information (such as condition, age, cultural background, eating habits, etc.);

[0009] Intelligent Q&A module: responsible for intelligently recommending relevant health knowledge and precautions according to the patient's historical query records and current health status;

[0010] Drug use and treatment and nursing management module: responsible for recording the patient's medication situation, including medication time, dosage, drug name, etc. The treatment and nursing management can generate a schedule according to the doctor's order, and has functions such as reminder, prompt, and recording of treatment and nursing;

[0011] Intelligent navigation and guidance module: responsible for realizing the patient's autonomous navigation from the ward to designated locations such as the examination room and pharmacy, and providing consultation information such as hospital department distribution and doctor's schedule;

[0012] Vital sign monitoring and analysis module: responsible for collecting the patient's vital sign data such as body temperature, blood pressure, and pulse, automatically analyzing and processing the collected data, and generating a health report and warning information;

[0013] Emotion recognition and interaction optimization module: responsible for recognizing the patient's emotional state and adjusting the communication method and content accordingly;

[0014] Emergency rescue module: Once it is found that the patient's vital signs are abnormal or there is an emergency call, it immediately triggers an alarm mechanism to notify the medical staff and starts the corresponding rescue process, and provides preliminary emergency guidance measures for the patient during the waiting for rescue;

[0015] The personalized education module interacts with the intelligent Q&A module and the drug use and treatment and nursing management module to obtain the patient's personalized information and historical records. The drug use and treatment and nursing management module interacts with the vital sign monitoring and analysis module to monitor the impact of the drug on the patient's vital signs and adjusts the medication plan accordingly. The intelligent navigation and guidance module interacts with the hospital information system to obtain the latest department distribution, doctor's schedule and other consultation information. The emotion recognition and interaction optimization module interacts with the personalized education module and the intelligent Q&A module to adjust the communication method and content according to the patient's emotional state. The emergency rescue module interacts with the vital sign monitoring and analysis module and the hospital emergency rescue system.

[0016] The above technical solution further includes:

[0017] Furthermore, the personalized education module includes a data collection unit, a data processing unit, a solution generation unit, a push unit, a learning progress tracking unit, and a personalized recommendation unit. The data collection unit is responsible for collecting various personalized information of patients, including medical records, questionnaire survey results, etc. The data processing unit uses data analysis algorithms to comprehensively process the information of patients and generate customized solutions. The solution generation unit generates specific diet, exercise, and life guidance solutions according to the data processing results. The push unit pushes the generated solutions to patients through the robot (1). The learning resource library stores and manages various learning materials, including videos, pictures and texts, Q&A, etc. The learning progress tracking unit records the learning progress and feedback of users for subsequent recommendation and optimization of learning materials. The personalized recommendation unit uses random forest to comprehensively analyze the personal information of patients, generates personalized guidance solutions, and dynamically adjusts the education content and difficulty according to the real-time health status and learning progress of patients to ensure that patients can gradually master the required knowledge. After the solution generation unit completes the solution formulation, it transfers the solution content to the push unit. The learning progress tracking unit transfers the learning history and feedback of users to the personalized recommendation unit. The personalized recommendation unit recommends learning materials that meet the interests and needs of users based on these data.

[0018] Furthermore, the intelligent Q&A module includes a speech recognition unit, a semantic understanding unit, a natural language generation unit, a speech synthesis unit, a historical record management unit, a health status assessment unit, and a recommendation algorithm unit. The speech recognition unit converts the voice input of patients into text form. The semantic understanding unit performs semantic analysis on the converted text to understand the real intentions and query contents of patients. The natural language generation unit generates answer texts in natural language form according to the results of semantic understanding. The speech synthesis unit converts the answer texts into voice outputs. The historical record management unit stores and manages the historical query records of patients, including query contents, times, frequencies, etc. The health status assessment unit evaluates the current health status of patients according to information such as the physical signs data and disease diagnoses of patients. The recommendation algorithm unit uses collaborative filtering and combines historical records and health status to generate a recommendation list.

[0019] Furthermore, the recommendation algorithm unit uses collaborative filtering and combines historical records and health status to generate a recommendation list. The specific steps are as follows:

[0020] Construct a patient-health guide matrix: According to the historical consultation records of patients and the feedback of health guides, the health guides include information such as health knowledge, drug usage instructions, precautions, etc., to construct a patient-health guide interaction matrix. Suppose there are m patients and n health guides, and the patient-health guide matrix R can be expressed as:

[0021]

[0022] Among them, r ij represents the satisfaction score or the number of consultations of patient i for health guideline j. If a certain patient has not consulted a certain health guideline, the corresponding element can be set to 0 or left blank;

[0023] Calculate similarity: Calculate the similarity between patients or the similarity between health guidelines:

[0024]

[0025] Among them, I uv is the set of health guidelines that both patient u and patient v have interacted with, r ui is the score or a certain metric value of patient u for health guideline i, r vi is the score or metric value of patient v for health guideline i;

[0026] Generate a recommendation list: Based on the consultation history and health guideline preferences of similar patients, generate a recommendation list for the current patient.

[0027] Furthermore, the drug use and treatment care management module includes a doctor's order transcription unit, a medication record unit, a precautions query unit, a schedule generation unit, and a medication reminder unit. The doctor's order transcription unit is responsible for accurately transcribing the doctor's orders (including drug use, treatment, examinations, and nursing content) into the system. The medication record unit is responsible for detailed recording of the patient's medication situation. The precautions query unit is responsible for providing the patient with a detailed drug instruction manual and a function for querying medication precautions to guide the patient to use drugs rationally. The schedule generation unit automatically or manually generates a treatment care schedule according to the doctor's orders, clarifying the time, content, and responsible person of each treatment care task. The medication reminder unit is responsible for setting a timing reminder function and reminding the patient to take medicine at the specified time according to the patient's treatment care schedule. The data in the medication record unit serves as the basis for the medication reminder unit. The precautions query unit provides the detailed information of the drug to the medication reminder unit.

[0028] Furthermore, the intelligent navigation and guidance module includes an autonomous navigation unit and a virtual guidance unit. The autonomous navigation unit uses the hospital map data and the navigation system of the robot to provide the patient with autonomous navigation services from the current location (such as the ward) to the target location (such as the peritoneal dialysis center, dialysis equipment replacement point, or specific nursing area, etc.), reducing the patient's trouble in finding the path. The virtual guidance unit provides virtual guidance services through the robot. The virtual guidance unit is provided with a hospital information database for storing the department distribution, doctor scheduling, and medical service items of the hospital, enabling the patient to retrieve relevant information from the hospital information database.

[0029] Furthermore, the physical sign monitoring and analysis module includes a data acquisition unit, an automatic analysis unit, a monitoring and correction unit, and a data visualization unit. The data acquisition unit is responsible for collecting the patient's vital sign data and peritoneal dialysis-related parameters, such as heart rate, blood pressure, body temperature, respiratory rate, effluent volume, and effluent composition (such as electrolyte concentration, urea nitrogen, etc.). The automatic analysis unit is responsible for automatically analyzing and processing the real-time collected vital sign data and peritoneal dialysis parameters, comprehensively evaluating the patient's health status and peritoneal dialysis effect, and generating a detailed health report, peritoneal dialysis effect evaluation, and warning information. The monitoring and correction unit is responsible for the real-time monitoring of specific behaviors of patients who need to strictly control their diet. The data visualization unit is responsible for converting the vital sign data and peritoneal dialysis parameters into a visual graphical interface.

[0030] Furthermore, the emotion recognition and interaction optimization module includes a face recognition unit, an emotion analysis unit, a communication strategy generation unit, and a personalized communication execution unit. The face recognition unit is responsible for capturing and recognizing the patient's facial image and extracting key emotion features. The emotion analysis unit is responsible for deeply analyzing the facial features, quickly judging the patient's emotional state, and evaluating its impact on peritoneal dialysis treatment. The communication strategy generation unit is responsible for generating personalized communication strategies according to the emotion analysis results, combined with the patient's personal information and peritoneal dialysis treatment needs, including language style, topic selection, and communication methods. The personalized communication execution unit is responsible for converting the communication strategy into actual actions, such as playing warm reminders through a smart voice assistant, displaying encouraging information on the screen, or adjusting the interaction interface of the treatment device, to communicate in the most suitable way for the patient's current emotional state. The face recognition unit captures the patient's facial image in real time and transmits it to the emotion analysis unit. The emotion analysis unit provides the emotion type, and the emotion type includes the emotion intensity, which is the basis for the communication strategy generation unit to formulate personalized strategies.

[0031] Furthermore, the emergency rescue module includes a data collection unit, an anomaly detection unit, an alarm trigger unit, a notification unit, a rescue process management unit, a peritoneal dialysis emergency guidance database, a guidance generation unit, and a guidance communication unit. The data collection unit is responsible for collecting real-time data from peritoneal dialysis machines, vital sign monitoring devices, and other sensors. The anomaly detection unit identifies abnormal patterns related to peritoneal dialysis. The alarm trigger unit intelligently judges and triggers an alarm based on the output of the anomaly detection unit or an emergency call signal. The notification unit ensures that emergency information is conveyed to relevant medical staff through text messages, phone calls, the hospital's internal communication system, etc. The rescue process management unit is specifically designed for the process management of peritoneal dialysis emergencies, including rapid dispatching of medical resources, activation of emergency plans, recording of event details, etc. The peritoneal dialysis emergency guidance database stores emergency treatment measures and steps in peritoneal dialysis emergencies. The guidance generation unit retrieves and generates personalized emergency guidance plans from the peritoneal dialysis emergency guidance database and the rescue process management unit according to the specific situation of the patient and the type of emergency event. The guidance communication unit uses various methods such as voice prompts, screen displays, or wearable device prompts to ensure that emergency guidance measures are accurately conveyed to the patient. The guidance communication unit monitors the patient's operation process in real time and immediately corrects and reminds if any non-standard operation or risk is found.

[0032] Furthermore, the emotion analysis unit judges the patient's emotional state and evaluates its impact on peritoneal dialysis treatment. The specific steps are as follows:

[0033] Facial feature extraction: Use a convolutional neural network to extract emotion-related features from the facial recognition unit, such as the curvature of the eyebrows, the upward or downward movement of the corners of the mouth, etc.;

[0034] Physiological feature integration: Combine facial features with physiological indicators (such as heart rate variability, skin conductance, etc.) to provide a more comprehensive evaluation of the emotional state. These physiological indicators can reflect the activities of the autonomic nervous system and are closely related to the emotional state;

[0035] Model application: Input the extracted features into a trained neural network, and the model will judge the patient's emotional state (such as happy, sad, anxious, calm, etc.) based on the extracted features;

[0036] Emotion assessment: The model assesses the intensity and duration of the emotional state, as well as the possible impact of this emotional state on peritoneal dialysis treatment. For example, an anxious emotion may cause the patient to not comply with the treatment plan or overly focus on physical symptoms;

[0037] Result output: The emotion analysis unit outputs the results of emotion classification and assessment to the communication strategy generation unit;

[0038] Feedback loop: Based on the results of sentiment analysis, the communication strategy generation unit adjusts the communication method with the patient, provides personalized psychological support, or triggers an emergency rescue process. At the same time, it continuously collects the patient's feedback and treatment effect data to continuously optimize the performance and accuracy of the sentiment analysis model.

[0039] The present invention has the following beneficial effects:

[0040] 1. In the present invention, the emotion recognition and interaction optimization module uses facial recognition and sentiment analysis technologies to identify the patient's emotional state and adjusts the communication method and content accordingly, providing personalized comfort, encouragement, or advice based on the patient's emotional changes to enhance the communication effect. This humanized communication method helps to relieve the patient's anxiety and fear emotions and enhance their confidence in overcoming the disease.

[0041] 2. In the present invention, by comprehensively considering multi-dimensional factors such as the patient's condition, age, cultural background, eating habits, etc., the personalized education module, intelligent Q&A module, drug use and treatment nursing management module, and physical sign monitoring and analysis module generate a highly personalized medical guidance plan. This customized service not only improves the pertinence and effectiveness of treatment but also significantly enhances the patient's satisfaction and compliance.

[0042] 3. In the present invention, the integrated emergency rescue module ensures that the patient can be treated in a timely manner when an abnormal situation occurs. This rapid response mechanism helps to minimize the losses and sufferings of the patient due to delayed treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a system block diagram of a ward health education robot proposed by the present invention;

[0044] Figure 2 is a structural diagram of the robot in the present invention;

[0045] Figure 3 is a flowchart of the sentiment analysis unit in the present invention for judging the patient's emotional state and evaluating its impact on peritoneal dialysis treatment.

[0046] In the figure: 1. Robot. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0048] Please refer to Figures 1-3As shown in the figure, the present invention is a ward health education robot. Inside the robot 1, there is a health education platform. The robot 1 is a mobile robot, equipped with medical device interfaces such as a high-definition camera, microphone, speaker, touch screen, body temperature sensor, sphygmomanometer, pulse oximeter, etc. The health education platform includes:

[0049] Personalized education module: responsible for generating personalized guidance plans according to the patient's personal information (such as condition, age, cultural background, eating habits, etc.);

[0050] Intelligent Q&A module: responsible for intelligently recommending relevant health knowledge and precautions according to the patient's historical query records and current health status;

[0051] Drug use and treatment nursing management module: responsible for recording the patient's medication situation, including medication time, dosage, drug name, etc. At the same time, it can transcribe the treatment, examination and related nursing content of the doctor's order. The treatment nursing management can generate a schedule according to the doctor's order, with functions such as reminder, prompt and recording of treatment and nursing;

[0052] Intelligent navigation and guidance module: responsible for realizing the autonomous navigation of patients from the ward to designated locations such as the examination room and pharmacy, and providing consultation information such as hospital department distribution and doctor scheduling;

[0053] Vital sign monitoring and analysis module: responsible for collecting the patient's vital sign data such as body temperature, blood pressure, and pulse, automatically analyzing and processing the collected data, and generating health reports and warning information;

[0054] Emotion recognition and interaction optimization module: responsible for identifying the patient's emotional state and adjusting the communication method and content accordingly;

[0055] Emergency rescue module: Once it detects that the patient's vital signs are abnormal or there is an emergency call, it immediately triggers an alarm mechanism to notify the medical staff and initiate the corresponding rescue process, and provides preliminary emergency guidance measures for the patient during the waiting for rescue;

[0056] The personalized education module interacts with the intelligent Q&A module and the drug use and treatment nursing management module to obtain the patient's personalized information and historical records. The drug use and treatment nursing management module interacts with the vital sign monitoring and analysis module to monitor the impact of drugs on the patient's vital signs and adjust the medication plan accordingly. The intelligent navigation and guidance module interacts with the hospital information system to obtain the latest department distribution, doctor scheduling and other consultation information. The emotion recognition and interaction optimization module interacts with the personalized education module and the intelligent Q&A module to adjust the communication method and content according to the patient's emotional state. The emergency rescue module interacts with the vital sign monitoring and analysis module and the hospital emergency rescue system.

[0057] In one embodiment, for the above-mentioned personalized education module, the personalized education module includes a data collection unit, a data processing unit, a solution generation unit, a push unit, a learning progress tracking unit, and a personalized recommendation unit. The data collection unit is responsible for collecting various personalized information of the patient, including medical records, questionnaire survey results, etc. The data processing unit uses data analysis algorithms to comprehensively process the patient's information and generate a customized solution. The solution generation unit generates specific diet, exercise, and life guidance solutions according to the data processing results. The push unit pushes the generated solutions to the patient through the robot (1). The learning resource library stores and manages various learning materials, including videos, pictures and texts, Q&A, etc. The learning progress tracking unit records the user's learning progress and feedback for subsequent recommendation and optimization of learning materials. The personalized recommendation unit comprehensively analyzes the patient's personal information using random forest, generates a personalized guidance solution, and dynamically adjusts the education content and difficulty according to the patient's real-time health status and learning progress to ensure that the patient can gradually master the required knowledge. After the solution generation unit completes the solution formulation, it transfers the solution content to the push unit. The learning progress tracking unit transfers the user's learning history and feedback to the personalized recommendation unit. The personalized recommendation unit recommends learning materials that meet the user's interests and needs based on this data.

[0058] The personalized recommendation unit comprehensively analyzes the patient's personal information using random forest, including the following steps:

[0059] Data collection and preprocessing: Collect the patient's personal information, including age, gender, medical history, living habits, medication records, etc. Clean and preprocess the collected data to remove redundant and incorrect information and ensure the accuracy and integrity of the data.

[0060] Feature selection and extraction: Extract features related to the patient's health needs and education focus from the preprocessed data. Use feature selection algorithms to further screen key features to improve the prediction performance of the model.

[0061] Build a random forest model: Use the processed data to train the random forest model. Random forest is a classifier that contains multiple decision trees, and its output result is determined by the mode of the categories output by individual trees. During the training process, optimize the performance of the model by adjusting the parameters of the random forest (such as the number of decision trees, maximum depth, selection of splitting features, etc.). Decision tree splitting criterion: The Gini index is used to select the optimal splitting feature. Gini index formula: Gini(D)=1-∑i = 1k(pi)2 where D represents the data set, k represents the number of categories, and pi represents the probability of the i-th category.

[0062] Generate personalized guidance plans: For new patients, input their personal information into the trained random forest model to obtain the prediction results of their health needs and education priorities. The prediction results output by the random forest are determined by the mode of the categories output by individual trees. Suppose there are T decision trees, and the prediction result of each tree for a certain sample is ci (i = 1, 2,..., T), then the prediction result output by the random forest is the ci that appears most frequently. According to the prediction results, generate personalized guidance plans for patients, including education content, learning methods, suggestions, etc. Dynamically adjust the education content and difficulty: During the process of patients receiving health education, monitor their health status and learning progress in real time. According to the patients' real-time feedback and learning effects, use the random forest model to re-evaluate their health needs and education priorities. According to the evaluation results, dynamically adjust the difficulty and depth of the education content to ensure that patients can gradually master the required knowledge.

[0063] Model evaluation and optimization: Regularly evaluate the random forest model, including calculating indicators such as accuracy and recall. Optimize the model according to the evaluation results, such as adjusting parameters, adding features, etc.

[0064] Use the random forest algorithm to provide personalized health education services for patients.

[0065] Data collection: The platform collected the personal information of a patient, including age (45 years old), gender (male), medical history (hypertension, diabetes), lifestyle habits (smoking, drinking), etc.

[0066] Feature extraction: Extract features related to health education from the collected data, such as age, gender, type of medical history, lifestyle habits, etc.

[0067] Model training: Use historical data to train a random forest model that can predict patients' health needs and education priorities.

[0068] Generate guidance plans: Input the patient's personal information into the model to obtain the prediction results of their health needs and education priorities. According to the prediction results, generate a personalized guidance plan for the patient, including suggestions such as a low-salt and low-fat diet, appropriate exercise, smoking cessation and alcohol restriction.

[0069] Dynamic adjustment: During the process of the patient receiving health education, monitor their health status and learning progress. It is found that the patient's blood pressure control is not ideal and their understanding of the low-salt and low-fat diet is not deep enough. Therefore, the platform uses the random forest model to re-evaluate the patient's health needs and education priorities, and adjusts the difficulty and depth of the education content, increasing the detailed explanation and examples of hypertension diet control.

[0070] Model Evaluation and Optimization: Regularly evaluate and optimize the random forest model to ensure that it can continuously provide accurate and personalized health education services for patients.

[0071] In one embodiment, for the above intelligent Q&A module, the intelligent Q&A module includes a speech recognition unit, a semantic understanding unit, a natural language generation unit, a speech synthesis unit, a history record management unit, a health status evaluation unit, and a recommendation algorithm unit. The speech recognition unit converts the patient's speech input into text form. The semantic understanding unit performs semantic analysis on the converted text to understand the patient's true intention and query content. The natural language generation unit generates an answer text in natural language form based on the results of semantic understanding. The speech synthesis unit converts the answer text into voice output. The history record management unit stores and manages the patient's historical query records, including query content, time, frequency, etc. The health status evaluation unit evaluates the patient's current health status based on the patient's physical sign data, disease diagnosis, and other information. The recommendation algorithm unit uses collaborative filtering and combines historical records and health status to generate a recommendation list.

[0072] In one embodiment, for the above recommendation algorithm unit, the recommendation algorithm unit uses collaborative filtering and combines historical records and health status to generate a recommendation list. The specific steps are as follows:

[0073] Construct a patient-health guide matrix: According to the patient's historical consultation records and the feedback of health guides, where health guides include information such as health knowledge, medication instructions, precautions, etc., construct a patient-health guide interaction matrix. Assuming there are m patients and n health guides, the patient-health guide matrix R can be represented as:

[0074]

[0075] where r ij represents the satisfaction score or consultation times of patient i for health guide j. If a patient has not consulted a certain health guide, the corresponding element can be set to 0 or left blank;

[0076] Calculate similarity: Calculate the similarity between patients or the similarity between health guides:

[0077]

[0078] where I uv is the set of health guides that both patient u and patient v have interacted with, r ui is the score or some metric value of patient u for health guide i, and r vi is the score or metric value of patient v for health guide i;

[0079] Generate a recommendation list: Based on the consultation history of similar patients and their preferences for health guidelines, generate a recommendation list for the current patient.

[0080] In one embodiment, for the above-mentioned drug use and treatment care management module, the drug use and treatment care management module includes a doctor's order transcription unit, a medication record unit, a precautions query unit, a schedule generation unit, and a medication reminder unit. The doctor's order transcription unit is responsible for accurately transcribing the doctor's orders (including drug use, treatment, examinations, and care content) into the system. The medication record unit is responsible for detailed recording of the patient's medication situation, including key information such as medication time, dosage, and drug name, for subsequent analysis and tracking. The precautions query unit is responsible for providing the patient with a detailed drug instruction manual and the function of querying medication precautions to guide the patient to use drugs rationally and avoid potential risks. The schedule generation unit automatically or manually generates a treatment care schedule according to the doctor's orders, clarifying the time, content, and responsible person of each treatment care task. The medication reminder unit is responsible for setting a timed reminder function. According to the patient's treatment care schedule, it reminds the patient to take medicine at the designated time to ensure that the treatment is completed on time and in the correct dosage. The data in the medication record unit (such as medication time, drug name, dosage, etc.) serves as the basis for the medication reminder unit. The precautions query unit provides detailed information such as the drug instruction manual, ingredients, usage and dosage, and precautions to the medication reminder unit.

[0081] In one embodiment, for the above-mentioned intelligent navigation and guiding module, the intelligent navigation and guiding module includes an autonomous navigation unit and a virtual guiding unit. The autonomous navigation unit uses the hospital map data and the navigation system built into Robot 1 to provide the patient with autonomous navigation services from the current location (such as the ward) to the target location (such as the peritoneal dialysis center, dialysis equipment replacement point, or specific care area, etc.), reducing the patient's trouble in finding the path. The virtual guiding unit provides virtual guiding services through Robot 1. The virtual guiding unit is provided with a hospital information database for storing the department distribution, doctor's schedule, and medical service items in the hospital, enabling the patient to retrieve relevant information from the hospital information database.

[0082] In one embodiment, for the above-mentioned vital sign monitoring and analysis module, the vital sign monitoring and analysis module includes a data acquisition unit, an automatic analysis unit, a monitoring and correction unit, and a data visualization unit. The data acquisition unit is responsible for collecting the patient's vital sign data and peritoneal dialysis-related parameters, such as heart rate, blood pressure, body temperature, respiratory rate, effluent volume, effluent composition (such as electrolyte concentration, urea nitrogen, etc.). The automatic analysis unit is responsible for automatically analyzing and processing the real-time collected vital sign data and peritoneal dialysis parameters, comprehensively evaluating the patient's health status and peritoneal dialysis effect, and generating a detailed health report, peritoneal dialysis effect evaluation, and warning information. The monitoring and correction unit is responsible for the real-time monitoring of the specific behaviors of patients who need to strictly control their diet. The data visualization unit is responsible for converting the vital sign data and peritoneal dialysis parameters into a visual graphical interface.

[0083] Suppose Ms. Li, a patient, needs to strictly control her diet and avoid consuming high-sugar and high-fat foods. The ward health education robot will monitor Ms. Li's diet through a camera or sensor. If the robot detects that Ms. Li is eating high-sugar foods (such as cakes, candies, etc.), it will immediately give a voice reminder: "The food you are eating has a high sugar content. It is recommended to replace it with low-sugar or sugar-free food." If Ms. Li violates the diet regulations repeatedly, the robot will strengthen the reminder and suggest communicating with medical staff to adjust the diet plan.

[0084] In one embodiment, for the above-mentioned emotion recognition and interaction optimization module, the emotion recognition and interaction optimization module includes a face recognition unit, an emotion analysis unit, a communication strategy generation unit, and a personalized communication execution unit. The face recognition unit is responsible for capturing and recognizing the patient's facial image and extracting key emotion features. The emotion analysis unit is responsible for deeply analyzing the facial features, quickly judging the patient's emotional state, and evaluating its impact on peritoneal dialysis treatment. The communication strategy generation unit is responsible for generating personalized communication strategies according to the emotion analysis results, combining the patient's personal profile and peritoneal dialysis treatment needs, including language style, topic selection, and communication methods. The personalized communication execution unit is responsible for converting the communication strategy into actual actions, such as playing warm reminders through a smart voice assistant, displaying encouraging information on the screen, or adjusting the interaction interface of the treatment device, to communicate in the most suitable way for the patient's current emotional state. The face recognition unit captures the patient's facial image in real-time and transmits it to the emotion analysis unit. The emotion analysis unit provides the emotion type, and the emotion type includes the emotion intensity, which is the basis for the communication strategy generation unit to formulate personalized strategies.

[0085] In one embodiment, for the above-mentioned emergency rescue module, the emergency rescue module includes a data collection unit, an anomaly detection unit, an alarm trigger unit, a notification unit, a rescue process management unit, a peritoneal dialysis emergency guidance database, a guidance generation unit, and a guidance communication unit. The data collection unit is responsible for collecting real-time data from peritoneal dialysis machines, vital sign monitoring devices, and other sensors. The anomaly detection unit identifies anomaly patterns related to peritoneal dialysis. The alarm trigger unit makes an intelligent judgment and triggers an alarm based on the output of the anomaly detection unit or an emergency call signal. The notification unit ensures that emergency information is conveyed to relevant medical staff through text messages, phone calls, the hospital's internal communication system, etc. The rescue process management unit is specifically designed for the process management of peritoneal dialysis emergencies, including quickly dispatching medical resources, activating emergency response plans, recording event details, etc. The peritoneal dialysis emergency guidance database stores emergency treatment measures and steps in peritoneal dialysis emergencies. The guidance generation unit retrieves and generates personalized emergency guidance plans from the peritoneal dialysis emergency guidance database and the rescue process management unit according to the patient's specific situation and the type of emergency event. The guidance communication unit uses various methods such as voice prompts, screen displays, or wearable device prompts to ensure that emergency guidance measures are accurately conveyed to the patient. The guidance communication unit monitors the patient's operation process in real time and immediately corrects and reminds if it finds non-standard operations or risks.

[0086] For patients who need to undergo peritoneal dialysis, the robot will provide detailed operation guidance through video tutorials or virtual reality technology. When the patient is operating, the robot will monitor the patient's operation steps and techniques in real time. If it finds non-standard operations or risks (such as incorrect connection of pipelines, incomplete disinfection, etc.), it will immediately correct them through voice or visual prompts. If the situation is serious, the robot will immediately contact medical staff and initiate the emergency rescue process.

[0087] Suppose patient Mr. Wang suddenly shows abnormal vital signs such as a rapid heart rate and difficulty breathing. The ward health education robot will immediately detect these abnormal changes through sensors. The robot will automatically trigger the early warning mechanism and remind Mr. Wang through voice: "Your vital signs are abnormal. Please stay calm immediately and don't panic. We will immediately contact medical staff to come and handle it." At the same time, the robot will provide Mr. Wang with preliminary emergency guidance according to the preset emergency guidance measures, such as "Please try to lie flat as much as possible and don't move around casually, waiting for the medical staff to arrive."

[0088] In one embodiment, for the above-mentioned emotion analysis unit, the emotion analysis unit judges the patient's emotional state and evaluates its impact on peritoneal dialysis treatment. The specific steps are as follows:

[0089] Facial feature extraction: Using a convolutional neural network, extract emotion-related features from the facial recognition unit, such as the curvature of the eyebrows, the upward or downward turn of the corners of the mouth, etc.;

[0090] Physiological feature integration: Combine facial features with physiological indicators (such as heart rate variability, skin conductance, etc.) to provide a more comprehensive assessment of the emotional state. These physiological indicators can reflect the activity of the autonomic nervous system and are closely related to the emotional state;

[0091] Model application: Input the extracted features into a trained neural network. The model will judge the patient's emotional state (such as happy, sad, anxious, calm, etc.) based on the extracted features;

[0092] Emotional assessment: The model evaluates the intensity and duration of the emotional state, as well as the possible impact of this emotional state on peritoneal dialysis treatment. For example, an anxious emotion may cause the patient to not comply with the treatment plan or overly focus on physical symptoms;

[0093] Result output: The emotion analysis unit outputs the results of emotion classification and assessment to the communication strategy generation unit;

[0094] Feedback loop: According to the results of emotion analysis, the communication strategy generation unit adjusts the communication method with the patient, provides personalized psychological support or triggers an emergency rescue process. At the same time, continuously collect the patient's feedback and treatment effect data to continuously optimize the performance and accuracy of the emotion analysis model;

[0095] Suppose a peritoneal dialysis patient is undergoing dialysis treatment, and the facial camera captures changes in his facial expressions. After preprocessing and feature extraction, the emotion analysis unit finds that his eyebrows are furrowed, the corners of his mouth are drooping, and his heart rate is accelerating. Input these features into the model, and the model determines that the patient is in an anxious state and assesses the degree of his anxiety as moderate. The system then triggers an alarm mechanism to notify the medical staff to pay attention to the patient's situation and automatically play a soothing voice guidance to help the patient relax and reduce anxiety.

[0096] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A ward health education robot, wherein a health education platform is arranged inside the robot (1), characterized in that, The health education platform includes: Personalized education module: responsible for generating personalized guidance plans based on the personal information of patients; Intelligent Q&A module: responsible for intelligently recommending relevant health knowledge and precautions according to the patient's historical query records and current health status; Drug use and treatment care management module: responsible for recording the patient's medication situation and generating a schedule for the patient; Intelligent navigation and guiding module: responsible for realizing the autonomous navigation of patients from the ward to the designated location and providing consultation information; Vital sign monitoring and analysis module: responsible for collecting the patient's vital sign data, automatically analyzing and processing the collected data, and generating health reports and warning information; Emotion recognition and interaction optimization module: responsible for recognizing the patient's emotional state and adjusting the communication method and content accordingly; Emergency rescue module: Once it is found that the patient's vital signs are abnormal or an emergency call is made, it immediately triggers an alarm mechanism to notify the medical staff and initiate the corresponding rescue process, and provides preliminary emergency guidance measures for the patient during the waiting for rescue; The personalized education module interacts with the intelligent Q&A module, the drug use and treatment care management module to obtain the patient's personalized information and historical records. The drug use and treatment care management module interacts with the vital sign monitoring and analysis module to monitor the impact of drugs on the patient's vital signs and adjust the medication plan accordingly. The intelligent navigation and guiding module interacts with the hospital information system to obtain the latest consultation information. The emotion recognition and interaction optimization module interacts with the personalized education module and the intelligent Q&A module to adjust the communication method and content according to the patient's emotional state. The emergency rescue module interacts with the vital sign monitoring and analysis module and the hospital emergency rescue system.

2. The ward health education robot according to claim 1, characterized in that, The personalized education module includes a data collection unit, a data processing unit, a plan generation unit, a push unit, a learning progress tracking unit, and a personalized recommendation unit. The data collection unit is responsible for collecting various personalized information of patients. The data processing unit uses data analysis algorithms to comprehensively process the patient's information and generate a customized plan. The plan generation unit generates specific diet, exercise, and life guidance plans according to the data processing results. The push unit pushes the generated plan to the patient through the robot (1). The learning resource library stores and manages various learning materials. The learning progress tracking unit records the user's learning progress and feedback for the subsequent recommendation and optimization of learning materials. The personalized recommendation unit uses random forest to comprehensively analyze the patient's personal information, generates a personalized guidance plan, and dynamically adjusts the education content and difficulty according to the patient's real-time health status and learning progress to ensure that the patient can gradually master the required knowledge. After the plan generation unit completes the plan formulation, it transfers the plan content to the push unit. The learning progress tracking unit transfers the user's learning history and feedback to the personalized recommendation unit. The personalized recommendation unit recommends learning materials that meet the user's interests and needs based on this data.

3. The ward health education robot according to claim 1, wherein, The intelligent Q&A module includes a speech recognition unit, a semantic understanding unit, a natural language generation unit, a speech synthesis unit, a history record management unit, a health status assessment unit, and a recommendation algorithm unit. The speech recognition unit converts the patient's voice input into text form. The semantic understanding unit performs semantic analysis on the converted text to understand the patient's true intention and query content. The natural language generation unit generates an answer text in natural language form based on the result of semantic understanding. The speech synthesis unit converts the answer text into voice output. The history record management unit stores and manages the patient's historical query records. The health status assessment unit evaluates the patient's current health status based on the patient's information. The recommendation algorithm unit uses collaborative filtering, combines historical records and health status, and generates a recommendation list.

4. The ward health education robot according to claim 3, wherein, The recommendation algorithm unit uses collaborative filtering, combines historical records and health status, and generates a recommendation list. The specific steps are as follows: Construct a patient-health guide matrix: According to the patient's historical consultation records and the feedback of health guides, construct a patient-health guide interaction matrix. Suppose there are m patients and n health guides. The patient-health guide matrix R can be expressed as: where r ij represents the satisfaction score or the number of consultations of patient i regarding health guideline j. If a patient has not consulted a certain health guideline, the corresponding element can be set to 0 or left blank; Calculate similarity: Calculate the similarity between patients or the similarity between health guides: Among them, I uv is the set of health guidelines that both patient u and patient v have interacted with, and r ui is the score or some measure of patient u for health guideline i, and r vi is the score or measure of patient v for health guideline i; Generate a recommendation list: Based on the consultation history and health guide preferences of similar patients, generate a recommendation list for the current patient.

5. The ward health education robot according to claim 1, characterized in that, The drug use and treatment care management module includes a doctor's order transcription unit, a medication record unit, a precautions query unit, a schedule generation unit, and a medication reminder unit. The doctor's order transcription unit is responsible for accurately transcribing the doctor's orders into the system. The medication record unit is responsible for detailed recording of the patient's medication situation. The precautions query unit is responsible for providing the patient with a detailed drug instruction manual and a function to query medication precautions to guide the patient to use drugs rationally. The schedule generation unit automatically or manually generates a treatment care schedule according to the doctor's orders, clarifying the time, content, and responsible person of each treatment care task. The medication reminder unit is responsible for setting a timing reminder function. According to the patient's treatment care schedule, it reminds the patient to take medicine at the specified time. The data in the medication record unit serves as the basis for the medication reminder unit. The precautions query unit provides detailed information about the drug to the medication reminder unit.

6. The ward health education robot according to claim 1, characterized in that, The intelligent navigation and guiding module includes an autonomous navigation unit and a virtual guiding unit. The autonomous navigation unit uses the hospital map data and the navigation system built into the robot (1) to provide the patient with autonomous navigation services from the current location to the target location, reducing the patient's trouble in finding the path. The virtual guiding unit provides virtual guiding services through the robot (1). The virtual guiding unit is provided with a hospital information database for storing the department distribution, doctor's schedule, and medical service items in the hospital, enabling the patient to retrieve relevant information from the hospital information database.

7. The ward health education robot according to claim 1, characterized in that, The physical sign monitoring and analysis module includes a data acquisition unit, an automatic analysis unit, a monitoring and correction unit, and a data visualization unit. The data acquisition unit is responsible for collecting the patient's vital sign data and peritoneal dialysis-related parameters. The automatic analysis unit is responsible for automatically analyzing and processing the real-time collected vital sign data and peritoneal dialysis parameters, comprehensively evaluating the patient's health status and peritoneal dialysis effect, generating a detailed health report, peritoneal dialysis effect evaluation, and warning information. The monitoring and correction unit is responsible for the real-time monitoring of specific behaviors of patients who need to strictly control their diet. The data visualization unit is responsible for converting the vital sign data and peritoneal dialysis parameters into a visual graphical interface.

8. The ward health education robot according to claim 1, characterized in that, The emotion recognition and interaction optimization module includes a face recognition unit, an emotion analysis unit, a communication strategy generation unit, and a personalized communication execution unit. The face recognition unit is responsible for capturing and recognizing the patient's facial images and extracting key emotion features. The emotion analysis unit is responsible for deeply analyzing the facial features, quickly judging the patient's emotional state, and evaluating its impact on peritoneal dialysis treatment. The communication strategy generation unit is responsible for generating personalized communication strategies based on the emotion analysis results, combined with the patient's personal information and peritoneal dialysis treatment needs. The personalized communication execution unit is responsible for converting the communication strategies into actual actions. The face recognition unit captures the patient's facial images in real time and transmits them to the emotion analysis unit. The emotion analysis unit provides the emotion type, and the emotion type includes the emotion intensity, which is the basis for the communication strategy generation unit to formulate personalized strategies.

9. The ward health education robot according to claim 1, wherein, The emergency rescue module includes a data acquisition unit, an anomaly detection unit, an alarm trigger unit, a notification unit, a rescue process management unit, a peritoneal dialysis emergency guidance database, a guidance generation unit, and a guidance communication unit. The data acquisition unit is responsible for collecting real-time data from peritoneal dialysis machines, vital sign monitoring devices, and other sensors. The anomaly detection unit identifies abnormal patterns related to peritoneal dialysis. The alarm trigger unit intelligently judges and triggers an alarm based on the output of the anomaly detection unit or an emergency call signal. The notification unit conveys the emergency information to relevant medical staff. The rescue process management unit is specifically designed for the process management of peritoneal dialysis emergencies. The peritoneal dialysis emergency guidance database stores the emergency treatment measures and steps in peritoneal dialysis emergencies. The guidance generation unit retrieves and generates personalized emergency guidance plans from the peritoneal dialysis emergency guidance database and the rescue process management unit according to the patient's specific situation and the type of emergency event. The guidance communication unit ensures that the emergency guidance measures are accurately conveyed to the patient. The guidance communication unit monitors the patient's operation process in real time and immediately corrects and reminds if it finds that the operation is not standardized or there are risks.

10. The ward health education robot according to claim 8, wherein, The emotion analysis unit judges the patient's emotional state and evaluates its impact on peritoneal dialysis treatment. The specific steps are as follows: Facial feature extraction: Use a convolutional neural network to extract emotion-related features from the face recognition unit. Physiological feature integration: Combine facial features with physiological indicators. Model application: Input the extracted features into the trained neural network, and the model will judge the patient's emotional state based on the extracted features; Emotional assessment: The model assesses the intensity and duration of the emotional state, as well as the possible impact of this emotional state on peritoneal dialysis treatment; Result output: The sentiment analysis unit outputs the results of sentiment classification and assessment to the communication strategy generation unit; Feedback loop: According to the results of sentiment analysis, the communication strategy generation unit adjusts the communication method with the patient, provides personalized psychological support or triggers an emergency rescue process. At the same time, continuously collect the patient's feedback and treatment effect data to continuously optimize the performance and accuracy of the sentiment analysis model.

Images