Hospital robot propaganda and education system and method based on medical knowledge base

Abstract

The invention relates to the technical field of medical informatization, in particular to a hospital robot propaganda and education system and method based on a medical knowledge base, and the system comprises an information collection module, a voice interaction module, an intelligent matching module and a terminal display module. The system realizes accurate acquisition of patient information, efficient processing of voice instructions, personalized matching of propaganda and education contents and interactive display in a multimedia form through multi-source data fusion and an intelligent algorithm. The problems of data isolation and inconvenient interaction of a traditional propaganda and education mode can be solved, the patient education quality and the medical care resource allocation efficiency are improved, and intelligent development of medical services is promoted.

Description

Technical Field

[0001] This invention belongs to the field of medical information technology and intelligent robot technology, specifically a hospital robot education system and method based on a medical knowledge base. Background Technology

[0002] Currently, hospital health education primarily includes oral education, written education, image education, slideshow education, or group education. This requires medical staff to conduct face-to-face presentations or explanations. However, brief communication is insufficient for patients and caregivers to fully understand the relevant information, easily affecting the effectiveness and satisfaction of patient education. Furthermore, the content of current educational devices is limited and cannot be personalized to each patient's individual circumstances, preventing patients from accurately learning the educational content based on their own needs. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a hospital education system and method based on the intelligent fusion of a medical knowledge base and patient data, comprising the following modules: Information Acquisition Module: This module acquires multi-dimensional patient information through various sensors and interface devices, including but not limited to vital signs data, medical records, medication information, and personal lifestyle habits. The information acquisition module adopts a distributed architecture design, supporting seamless integration with Hospital Information Systems (HIS), Laboratory Information Systems (LIS), and Picture Archiving and Communication Systems (PACS). The module integrates a high-performance data conversion chip to uniformly convert data from different sources into a standardized format and store it in a central database, providing fundamental support for subsequent processing.

[0004] Voice Interaction Module: This module receives voice commands from patients in the ward via a microphone array and preprocesses the sound signal using a noise suppression algorithm to reduce the impact of environmental noise on recognition accuracy. The voice signal is then converted from analog to digital and transmitted to the natural language processing unit, where a deep learning model parses the semantic content and generates corresponding text commands. The module also includes voice feedback functionality, playing system responses to the patient through a speaker, thus establishing a two-way interactive mechanism.

[0005] Intelligent Matching Module: This module performs correlation analysis between patient information and content in the medical knowledge base to generate personalized education plans. The module incorporates a multi-layered neural network structure. The first layer extracts key patient characteristics, such as disease type, stage of illness, and personal preferences. The second layer matches these characteristics with relevant content in the knowledge base, selecting the most suitable education materials. The third layer optimizes the content combination order through logical reasoning, ensuring that the output conforms to medical standards and patient needs.

[0006] Terminal Display Module: This module consists of a high-definition display screen and a touch panel, used to present educational content and facilitate patient interaction. The display screen supports multi-resolution output and can switch display modes according to different scenarios, such as text descriptions, video demonstrations, or a combination of text and images. The touch panel uses capacitive sensing technology for precise operation; patients can select desired content or adjust playback progress by clicking, swiping, etc. The module also integrates a light sensor to automatically adjust screen brightness to adapt to ambient lighting conditions.

[0007] The education system also includes a dynamic knowledge base management system for nurse-patient collaboration: when a patient's disease education needs exceed the scope of the knowledge base, the robot packages the needs into a structured task and pushes it to the nurse station. The structured task includes the question text, patient ID, and urgency level tag. The knowledge base is compiled by professional medical staff and can only be entered into the knowledge base after being reviewed by superiors.

[0008] As a further aspect of the present invention, a hospital education method based on the intelligent fusion of medical knowledge base and patient data includes the following steps: S1: The information collection module acquires basic patient information and medical data, categorizing patients into several classes based on their primary disease type. Each class corresponds to a specific knowledge base partition. For example, cardiovascular disease patients are assigned to the "Chronic Disease Management" partition, and postoperative rehabilitation patients are assigned to the "Postoperative Care" partition.

[0009] S2: Based on the ward layout, independent sensing zones are set up around each bed, with each zone equipped with a microphone array and a terminal device. The sensing zones are connected via a wireless communication protocol to form a local area network covering the entire ward.

[0010] S3: When the patient issues a voice command, the microphone array captures the sound signal and transmits it to the voice interaction module. The module separates the background noise from the target signal using a spectrum analysis algorithm, and calculates the time difference between the signals received by each microphone to infer the location of the sound source.

[0011] S4: Based on the sound source localization results, determine the bed number of the patient who made the sound, and retrieve the relevant information of the patient from the central database. The module compares the patient's current status with historical records to determine whether the educational content needs to be updated.

[0012] S5: Based on the patient's specific situation, appropriate content is extracted from the medical knowledge base and presented to the patient through the terminal display module. Content formats include dynamic diagrams, 3D animations, or real-time audio explanations, facilitating patient understanding and memorization.

[0013] As a further aspect of the present invention, in step S1, the patient classification criteria not only consider the type of disease but also take into account factors such as age, gender, and education level. For example, for elderly patients, simple and easy-to-understand textual explanations are prioritized; for younger patients, multimedia formats are preferred to enhance appeal.

[0014] As a further aspect of this invention, in step S3, the sound source localization algorithm employs beamforming technology, specifically as follows: First, the original audio signal is processed into frames, with each frame length set to 20 milliseconds and an overlap rate of 50%; second, frequency features are extracted using short-time Fourier transform, and a spatial power spectrum is constructed; finally, the sound source direction angle is determined by searching for the peak value of the power spectrum. Furthermore, to improve localization accuracy, the module introduces a Kalman filter algorithm to perform a weighted average of multiple measurement results.

[0015] As a further aspect of the present invention, in step S4, the patient information query process includes a multi-level verification mechanism. The system first initially locates the target patient through the bed number index, then verifies the patient's identification code (such as a wristband QR code), and finally performs subsequent operations after confirming that the information is correct.

[0016] As a further aspect of the present invention, in step S5, the selection strategy for educational content is based on a patient behavior pattern prediction model. The model's input variables include the patient's historical interaction records, current emotional state, and environmental parameters, while the output variable is the priority ranking of the recommended content. For example, for patients who have repeatedly asked a certain question recently, the system will proactively push relevant answers.

[0017] Beneficial Effects: This invention integrates multi-source data and intelligent algorithms to construct a complete hospital education solution. The information collection module solves the problem of data isolation in traditional education methods, the voice interaction module improves the convenience of human-computer communication, the intelligent matching module enables highly personalized education content, and the terminal display module enhances the user experience. The overall design fully reflects technological advancement and practicality, helping to improve the quality of patient education, optimize the allocation of medical resources, and promote the intelligent development of medical services. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall system architecture of the present invention.

[0019] Figure 2 This is a structural block diagram of the information acquisition module.

[0020] Figure 3 This is a flowchart of the voice interaction module.

[0021] Figure 4 This is the algorithm logic diagram for the intelligent matching module.

[0022] Figure 5 This is a schematic diagram of the hardware components of the terminal display module.

[0023] Figure 6 A schematic diagram showing the layout of the sensing area in the ward. Detailed Implementation

[0024] This invention provides a hospital education system and method based on the intelligent fusion of medical knowledge base and patient data, and its specific implementation is described in detail with reference to the accompanying drawings. Figure 1 As shown, the overall system architecture includes an information acquisition module, a voice interaction module, an intelligent matching module, and a terminal display module. Each module transmits and processes information through data flow. The connections, positions, and cooperation between these modules will be described in detail below.

[0025] The information acquisition module, as the core data input unit of the system, has the following structural block diagram: Figure 2 As shown, this module connects to various medical devices via sensor interfaces to acquire multi-dimensional patient information, including vital signs data, medical records, medication information, and personal lifestyle habits. The sensor interfaces support standardized protocols, enabling seamless integration with Hospital Information Systems (HIS), Laboratory Information Systems (LIS), and Picture Archiving and Communication Systems (PACS), thus achieving unified acquisition of multi-source data. A data conversion chip, integrated within the data acquisition module, is responsible for converting data from different sources into a unified format and storing it in a central database. In practical applications, the data acquisition modules are deployed in different areas of the ward using a distributed architecture. The sensor interfaces are located close to the patient's bedside for real-time monitoring of vital signs, while the data conversion chips are centrally located in the ward's technical room to reduce interference during signal transmission. This layout not only optimizes data acquisition efficiency but also ensures data accuracy and integrity.

[0026] The workflow of the voice interaction module is as follows: Figure 3As shown, its core components include a microphone array and a speaker. The microphone array is arranged in the sensing area around each bed to capture the patient's voice commands. To reduce the impact of environmental noise on recognition accuracy, the voice interaction module uses a noise suppression algorithm to preprocess the sound signal. Specifically, the raw audio signal is first converted into a digital signal by an analog-to-digital converter, then frequency features are extracted using short-time Fourier transform, and a spatial power spectrum is constructed to separate background noise from the target signal. The sound source localization algorithm uses beamforming technology, which divides the audio signal into frames with a length of 20 milliseconds and an overlap rate of 50%, calculates the time difference of the signals received by each microphone, and thus infers the direction angle of the sound source. To further improve localization accuracy, a Kalman filter algorithm is used to perform a weighted average of multiple measurement results to finally determine the specific location of the sound source. The speaker plays the system response content to the patient based on the text commands generated by the natural language processing unit, forming a two-way interactive mechanism. In the actual ward environment, the microphone array and speaker are installed on both sides of the bed to ensure that the acquisition and feedback of the voice signal can cover the patient's range of motion.

[0027] The intelligent matching module is the core processing unit of the system, and its algorithm logic is as follows: Figure 4 As shown, this module uses a multi-layer neural network structure to perform correlation analysis between patient information and medical knowledge base content. The first layer of the neural network is responsible for extracting key patient features from the central database, such as disease type, disease stage, and personal preferences. The second layer of the neural network matches relevant content in the medical knowledge base based on the extracted features, selecting the most suitable educational materials. The third layer of the neural network optimizes the combination order of content through logical reasoning, ensuring that the output results conform to medical standards and patient needs. In actual operation, the intelligent matching module first receives patient data from the information collection module, then processes it layer by layer through the neural network, and finally generates a personalized educational plan. For example, for cardiovascular disease patients, the system will prioritize pushing content from the "Chronic Disease Management" section; while for postoperative rehabilitation patients, it will recommend relevant materials from the "Postoperative Care" section. In addition, the intelligent matching module also combines a patient behavior pattern prediction model to dynamically adjust the priority order of content based on historical interaction records, current emotional state, and environmental parameters, thereby meeting the personalized needs of patients.

[0028] The hardware components of the terminal display module are as follows: Figure 5As shown, the system includes a high-definition display screen, a touch panel, and a light sensor. The high-definition display screen is located directly in front of the patient's bed and is used to present educational content and interact with the patient. The display screen supports multi-resolution output and can switch display modes according to different scenarios, such as text descriptions, video demonstrations, or a combination of text and images. The touch panel uses capacitive sensing technology for precise operation; patients can select desired content or adjust playback progress by clicking and swiping. The light sensor, integrated above the display screen, detects ambient light conditions and automatically adjusts screen brightness to adapt to different usage environments. In practical applications, the terminal display module and voice interaction module work together. When the patient issues a voice command, the system presents the corresponding educational content on the high-definition display screen. For example, for elderly patients, the system prioritizes simple and easy-to-understand text descriptions; while for younger patients, it tends to use multimedia formats to enhance engagement.

[0029] The layout of the sensor area in the ward is as follows Figure 6 As shown, each bed is surrounded by an independent sensing area containing a microphone array and terminal devices. These sensing areas are connected via a wireless communication protocol, forming a local area network covering the entire ward. The wireless communication protocol uses Bluetooth Low Energy technology to ensure the stability and security of data transmission. In actual operation, when a patient issues a voice command, the microphone array captures the sound signal and transmits it to the voice interaction module. The module uses a spectrum analysis algorithm to separate background noise from the target signal and calculates the time difference between the signals received by each microphone to infer the location of the sound source. Based on the sound source localization result, the system determines the bed number of the patient and retrieves the patient's relevant information from the central database. To ensure the accuracy of information retrieval, the system employs a multi-level verification mechanism. First, the target patient is initially located using the bed number index; then, the patient's identification code (such as a wristband QR code) is verified; finally, subsequent operations are performed only after the information is confirmed to be correct.

[0030] In this invention, the health education system also includes dynamic knowledge base management for nurse-patient collaboration: when a patient's health education needs exceed the scope of the knowledge base, the robot packages the needs into a structured task and pushes it to the nurse station. The structured task includes the question text, patient ID, and urgency level tag. The knowledge base is compiled by professional medical staff and can only be entered into the knowledge base after being reviewed by superiors.

[0031] In its implementation, the system operates on the following principles and procedures: First, the information acquisition module obtains basic patient information and medical data through sensor interfaces, categorizing patients into several classes based on their primary disease types. Each class corresponds to a specific knowledge base partition. For example, cardiovascular disease patients are assigned to the "Chronic Disease Management" partition, and postoperative rehabilitation patients are assigned to the "Postoperative Care" partition. Second, the voice interaction module captures the patient's voice commands through a microphone array and uses beamforming technology and Kalman filtering algorithms to determine the sound source location. Based on the sound source localization results, the system retrieves relevant patient information from the central database and generates personalized educational plans through the intelligent matching module. Finally, the terminal display module presents the educational content on a high-definition screen, allowing patients to interact via a touch panel. For example, for patients who have repeatedly asked a particular question recently, the system will proactively push relevant answers, thereby improving the educational effect.

[0032] In practical applications, this system significantly improves the quality of patient education and optimizes the allocation of medical resources. For example, during the education of cardiovascular disease patients, the system pushes relevant content on diet control, exercise suggestions, and medication use based on the patient's historical records and current status. For postoperative rehabilitation patients, the system provides detailed nursing guidance and rehabilitation plans. By integrating multi-source data and intelligent algorithms, this system achieves a high degree of personalization in educational content while improving the convenience of human-computer communication and user experience.

[0033] To enable those skilled in the art to fully understand and implement this invention, the specific implementation principle of this invention will be further explained below in conjunction with a specific application scenario.

[0034] During patient education for cardiovascular diseases, the system first interfaces with the Hospital Information System (HIS) through the sensor interface in the information acquisition module to acquire patients' vital signs data, medical records, and medication information in real time. This data is then processed by a data conversion chip and stored in a unified format in a central database. For example, for a patient with hypertension, the system records their blood pressure fluctuation trends, recent medication types and dosages, and notes their lifestyle habits such as dietary preferences and exercise frequency. Because the sensor interface is located close to the patient's bedside, the real-time nature and accuracy of vital signs data are ensured, while the centralized placement of the data conversion chips in the ward's technical area further reduces interference during signal transmission.

[0035] Subsequently, the voice interaction module captures the patient's voice commands via a microphone array. When the patient asks, "How should I control my blood pressure?", the raw audio signal is first converted into a digital signal by an analog-to-digital converter, and frequency features are extracted using a short-time Fourier transform. Based on this, beamforming technology calculates the time difference between the signals received by each microphone to preliminarily infer the direction angle of the sound source. To further improve positioning accuracy, a Kalman filter algorithm performs a weighted average of multiple measurements, ultimately determining the patient's bed number as bed number 3. Simultaneously, the system uses a multi-level verification mechanism, combined with the patient's wristband QR code, to confirm the identity information before retrieving the patient's relevant data from the central database.

[0036] After receiving patient data from the information collection module, the intelligent matching module analyzes and processes it based on a multi-layer neural network structure. The first layer of the neural network extracts key patient characteristics from the central database, such as the type of disease (hypertension), the stage of the disease (stable), and the patient's preference for a low-salt diet. The second layer of the neural network matches these characteristics with relevant content in the medical knowledge base, selecting the most suitable educational materials, such as advice on a low-salt diet and exercise guidance. The third layer of the neural network optimizes the order of content combination through logical reasoning, ensuring that the output conforms to medical standards and patient needs. In addition, the intelligent matching module also incorporates a patient behavior pattern prediction model to dynamically adjust the content priority ranking. For example, if the system detects that a patient has recently asked multiple questions about medication use, it will prioritize pushing relevant answers.

[0037] The terminal display module presents educational content on a high-definition screen and integrates with a touch panel for human-computer interaction. For the aforementioned hypertensive patients, the system prioritizes pushing simple and easy-to-understand text instructions, such as "Daily salt intake should be controlled within 5 grams" and "It is recommended to engage in 30 minutes of aerobic exercise daily." Simultaneously, a light sensor detects ambient light conditions and automatically adjusts the screen brightness to adapt to different usage scenarios within the ward. Patients can select desired content or adjust playback progress by clicking on the touch panel. For example, patients can switch to a video demonstration by swiping to watch a detailed explanation of a low-salt diet.

[0038] In actual operation, the layout of the sensing areas within the ward ensures the system's efficient operation. Each bed is equipped with an independent sensing area containing a microphone array and terminal devices. These sensing areas are connected via a wireless communication protocol, forming a local area network covering the entire ward. When a patient issues a voice command, the microphone array captures the sound signal and transmits it to the voice interaction module. The module uses a spectrum analysis algorithm to separate background noise from the target signal and, combined with the sound source localization results, pinpoints the patient's location. For example, when a patient in bed 3 asks a question, the system quickly determines their location via the wireless communication protocol and retrieves relevant information to generate a personalized education plan.

[0039] Through the steps described above, this system achieves a high degree of personalization and convenience in its patient education content. For example, for postoperative recovery patients, the system provides detailed nursing guidance and rehabilitation plans based on their surgical type and recovery stage. For elderly patients, the system prioritizes text-based instructions, while for younger patients, it tends to use multimedia formats to enhance engagement. The overall design not only significantly improves the quality of patient education but also optimizes the allocation of medical resources and promotes the intelligent development of healthcare services.

[0040] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0041] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A hospital robot-based educational system based on a medical knowledge base, characterized in that, Includes the following modules: Information acquisition module: Acquires multi-dimensional information about patients through sensor interfaces, including vital signs data, medical records, medication information and personal lifestyle habits, and converts the data into a standardized format for storage in a central database; Voice interaction module: Receives the patient's voice commands through a microphone array, preprocesses the sound signal using a noise suppression algorithm, parses the semantic content through a natural language processing unit to generate text commands, and plays the system response content to the patient through a speaker; Intelligent matching module: Based on the correlation analysis between patient information and the content in the medical knowledge base, it generates personalized education plans; Terminal display module: Presents educational content and allows interaction with patients through a high-definition display screen and touch panel.

2. The hospital robot education system based on a medical knowledge base according to claim 1, characterized in that, The information acquisition module adopts a distributed architecture design, which supports seamless integration with hospital information systems, laboratory information systems, and image archiving and communication systems. The module integrates a data conversion chip to achieve standardized conversion of data from different sources.

3. A hospital robot education system based on a medical knowledge base according to claim 1, characterized in that, The voice interaction module uses beamforming technology to determine the location of the sound source. The audio signal is processed in frames, each frame is 20 milliseconds long and has an overlap rate of 50%. The Kalman filter algorithm is used to perform a weighted average of multiple measurement results.

4. A hospital robot education system based on a medical knowledge base according to claim 1, characterized in that, The intelligent matching module has a built-in multi-layer neural network structure. The first layer extracts the patient's key features, the second layer matches relevant content in the medical knowledge base based on the features, and the third layer optimizes the order of content combination through logical reasoning.

5. A hospital robot-based educational system based on a medical knowledge base according to claim 1, characterized in that, The education system also includes a dynamic knowledge base management system for nurse-patient collaboration: when a patient's disease education needs exceed the scope of the knowledge base, the robot packages the needs into a structured task and pushes it to the nurse station. The structured task includes the question text, patient ID, and urgency level tag. The knowledge base is compiled by professional medical staff and can only be entered into the knowledge base after being reviewed by superiors.

6. A hospital robot-based educational method based on a medical knowledge base, characterized in that, Includes the following steps: S1: Acquire basic patient information and diagnosis data through the information collection module, and classify patients into several categories according to their main disease types, with each category corresponding to a specific knowledge base partition; S2: Set up a sensing area in the ward, with each area equipped with a microphone array and terminal equipment, and form a local area network through a wireless communication protocol; S3: When the patient issues a voice command, the microphone array captures the sound signal and transmits it to the voice interaction module. The module separates the background noise from the target signal and calculates the location of the sound source through a spectrum analysis algorithm. S4: Determine the bed number of the patient who made the sound based on the sound source localization results, and retrieve the relevant information of the patient from the central database; S5: Extract appropriate content from the medical knowledge base based on the patient's specific situation and present it to the patient through the terminal display module.

7. A hospital robot education method based on a medical knowledge base according to claim 6, characterized in that, In step S1, the patient classification criteria comprehensively consider factors such as disease type, age, gender, and education level.

8. A hospital robot education method based on a medical knowledge base according to claim 6, characterized in that, In step S3, the sound source localization algorithm uses beamforming technology to extract frequency features and construct a spatial power spectrum through short-time Fourier transform, and searches for the peak value of the power spectrum to determine the direction angle of the sound source.

9. A hospital robot education method based on a medical knowledge base according to claim 6, characterized in that, In step S4, the patient information query process includes a multi-level verification mechanism. First, the target patient is initially identified by the bed number index, and then the patient's identity code is verified to confirm that the information is correct.

10. A hospital robot education method based on a medical knowledge base according to claim 6, characterized in that, In step S5, the selection strategy for the educational content is based on a patient behavior pattern prediction model. The model input variables include the patient's historical interaction records, current emotional state, and environmental parameters.

Images