A family doctor robot based on bioelectric and magnetic resonance principles

Abstract

The application discloses a family doctor robot based on bioelectricity and magnetic resonance principles, and relates to the technical field of intelligent medical equipment, and solves the technical problems of single detection dimension, low diagnosis accuracy and lack of personalized health intervention of existing family medical equipment. The robot integrates a bioelectricity detection module, a magnetic resonance detection module, an intelligent analysis processing module, an interactive execution module and a cloud service module, collects bioelectricity signals of multiple parts of the human body through a flexible sensor, realizes non-invasive structure and metabolism detection of local tissues of the human body in combination with low-field magnetic resonance, completes multi-dimensional health data fusion analysis, preliminary diagnosis of diseases and risk assessment relying on an AI diagnosis model, simultaneously outputs a personalized intervention scheme and executes basic health services, and can also realize continuous tracking of health data and docking of remote medical treatment. The application realizes non-invasive, multi-dimensional and intelligent health monitoring and preliminary diagnosis and treatment in a family scene, has high detection accuracy, is convenient to operate and has strong adaptability, and greatly improves the professionalism and convenience of family health management.

Description

Technical Field

[0001] This invention relates to the field of intelligent medical device technology, specifically to a family doctor robot based on the principles of bioelectricity and magnetic resonance, which is suitable for non-invasive health monitoring, preliminary diagnosis of common diseases, and basic health intervention in home settings. Background Technology

[0002] With the increasing health awareness of residents, family health management has become an important development direction in the medical and health field. Family medical devices are also evolving from basic testing equipment such as blood pressure monitors and blood glucose meters to intelligent and multifunctional devices. However, existing family medical devices still have many shortcomings: First, they have limited testing dimensions, often only able to detect a single physiological indicator, failing to achieve multi-dimensional and systematic health screening and thus failing to meet the comprehensive health monitoring needs of families. Second, their diagnostic capabilities are weak, only able to collect and display indicator data, lacking professional analysis and preliminary diagnostic capabilities; users must submit the data to professional doctors themselves, resulting in poor convenience. Third, existing non-invasive imaging devices are mostly large, high-field magnetic resonance imaging (MRI) devices, which are bulky and expensive, making them unsuitable for home use. Fourth, they lack personalized health intervention and continuous health data management, making it difficult to form a closed-loop family health management system encompassing "detection-diagnosis-intervention-tracking."

[0003] Bioelectrical signals are important indicators of human physiological activity. Signals such as electrocardiogram (ECG), electroencephalogram (EEG), and electromyography (EMG) can directly reflect the functional state of tissues such as the cardiovascular system, nervous system, and muscles. Magnetic resonance imaging (MRI) technology can achieve non-invasive imaging of human soft tissues, reflecting changes in tissue structure and metabolism. Combining these two technologies can achieve precise screening of both human function and structure, significantly improving the accuracy of health testing. Currently, there is no home medical robot that organically integrates bioelectrical detection and low-field MRI detection, and possesses intelligent diagnosis, intervention, and data management capabilities. Therefore, developing a home doctor robot that integrates the principles of bioelectrical and MRI detection, is adapted to home environments, and has multi-dimensional intelligent diagnostic and health intervention capabilities is key to solving existing technological problems. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a family doctor robot based on the principles of bioelectricity and magnetic resonance. This robot enables non-invasive, multi-dimensional, and intelligent health monitoring, preliminary diagnosis of common diseases, personalized intervention, and closed-loop management of health data in a home setting. It solves the technical problems of existing devices, such as limited detection dimensions, low diagnostic accuracy, poor adaptability, and lack of closed-loop health management.

[0005] A family doctor robot based on the principles of bioelectricity and magnetic resonance includes a robot body, which integrates a bioelectricity detection module, a magnetic resonance detection module, and an intelligent analysis and processing module. An interactive execution module is located on the outside of the robot body. The intelligent analysis and processing module is connected to a cloud service module through a communication module. A mobility module is located at the bottom of the robot body, and a charging module is located on the back.

[0006] 1. Bioelectric detection module

[0007] The flexible sensor array, made of conductive silicone with a thickness of 0.3-0.5mm and an elongation of ≥150%, can be closely attached to the chest, head, limbs, and other parts of the human body to collect four core bioelectrical signals: electrocardiogram (ECG), electroencephalogram (EEG), electromyography (EMG), and electrodermal conduction (EDC). The module integrates a 1000-5000x signal amplification unit, a second-order active low-pass filter unit with a cutoff frequency of 0.05-100Hz, and an A / D conversion unit with a sampling accuracy of ≥16bit and a sampling frequency of 1-2kHz. After noise reduction, amplification, and analog-to-digital conversion of the raw bioelectrical signals, the digital signals are transmitted to the intelligent analysis and processing module, effectively reducing environmental interference and improving signal acquisition accuracy.

[0008] 2. Magnetic Resonance Detection Module

[0009] This is a portable, low-field permanent magnet magnetic resonance testing unit with a magnetic field strength of 0.5-1.5T. It balances testing accuracy with the needs of non-invasive home use, poses no significant radiation risk, and has a device size of ≤0.8m². 3 Weighing ≤50kg, it is suitable for home space layouts. The module includes a 12-64MHz radio frequency transmitting unit, a phased array signal receiving coil and an imaging processing unit. The imaging resolution is ≥256×256 pixels, and the imaging time for a single part is ≤5min. It can perform non-invasive magnetic resonance imaging detection and tissue metabolic signal acquisition on local tissues such as the human head, chest and limbs, and realize preliminary screening of the structure and function of soft tissues and organs.

[0010] 3. Intelligent Analysis and Processing Module

[0011] With a high-performance embedded chip as the core controller, it incorporates an AI diagnostic model based on ResNet+LSTM fusion deep learning algorithm, a health data fusion algorithm, and a local health benchmark database. The model achieves a preliminary diagnostic accuracy of ≥88% for common cardiovascular, nervous, and musculoskeletal diseases. It can standardize and de-redundant preprocessing of detection data, extract abnormal feature values, and compare them with benchmark data to achieve preliminary disease diagnosis and low, medium, and high risk level assessment. It also integrates a plan generation module, which can output personalized health monitoring, diet, exercise, and basic intervention plans based on the diagnostic results.

[0012] 4. Interactive Execution Module

[0013] It includes a 10.1-inch 1920×1200 resolution IPS high-definition touch screen, a far-field voice interaction unit, and a detection assistance unit. The voice interaction unit has a pickup distance of ≤5m, a recognition accuracy of ≥95%, and supports Mandarin, multiple dialects, and simple English, making it suitable for users of different ages. The robotic arm has a rated load of ≥2kg and a repeatability of ±0.1mm, enabling basic health interventions such as automatic sensor attachment, adjustable massage with 5-30N intensity, and acupoint stimulation with ±2mm accuracy. The detection assistance unit includes an infrared human body positioning sensor and an audio-visual posture guidance unit to guide users to complete standard detection postures, reducing detection errors caused by improper posture.

[0014] 5. Communication module

[0015] The communication module includes an IEEE 802.11a / b / g / n / ac wireless WiFi module, a Bluetooth 5.0 module, and an NR Sub-6G 5G communication module, with a data transmission rate of ≥100Mbps. It supports wireless networking and real-time data transmission between the robot and cloud service modules, user mobile terminals, and wearable detection devices, enabling real-time synchronization and remote viewing of health data while ensuring the stability and efficiency of data transmission.

[0016] 6. Cloud Service Module

[0017] Deployed on a highly available cloud server cluster, the system includes a health database, a model update unit, a telemedicine integration unit, and a blockchain data security unit. The health database stores ≥500,000 bioelectrical baseline data, magnetic resonance imaging baseline data, and common disease case data for individuals of different ages, genders, and physical conditions. The model update unit iteratively optimizes the AI ​​diagnostic model based on massive clinical data and user test data, improving diagnostic accuracy. The magnetic resonance imaging baseline data and massive common disease case data provide data support for AI diagnosis. The model update unit continuously optimizes the AI ​​diagnostic model algorithm based on massive clinical data and user test data, improving diagnostic accuracy. The telemedicine integration unit can upload user test data and preliminary diagnostic results to a telemedicine platform, enabling connection with professional doctors, supporting remote consultations, and compensating for the lack of professional expertise in home-based preliminary diagnoses. The data security unit uses blockchain encryption technology and the SHA-256 encryption algorithm, supporting data anonymization and tiered access, encrypting and storing and transmitting user health data to effectively prevent data leakage and protect user data privacy.

[0018] 7. Mobility module and charging module

[0019] The mobile module includes dual drive wheels, a laser obstacle avoidance sensor with a detection distance of 0.05-3m, a SLAM navigation unit, and an attitude sensor. It moves at a speed of 0.1-0.5m / s and has a positioning accuracy of ±0.05m. It can achieve autonomous movement, obstacle avoidance, and precise positioning in home scenarios, and actively approach users to complete detection services. The charging module is a 15-30W wireless charging unit with a charging efficiency of ≥88%. When the robot's battery level drops below a preset threshold of 20%, it automatically triggers a return-to-charging command, returns to the charging position, and completes wireless charging to ensure the device's continuous working capability.

[0020] This invention also discloses a method for operating a family doctor robot based on the principles of bioelectricity and magnetic resonance, comprising the following steps:

[0021] S1: Human-computer interaction is initiated. The user inputs detection requirements to the interaction execution module via voice / touch. The intelligent analysis and processing module receives and parses the detection requirements and generates corresponding detection instructions.

[0022] S2: Detection preparation. The mobile module drives the robot body to move autonomously to the user's side. The detection assistance unit guides the user to complete the standard detection posture through voice and light. The flexible sensor array of the bioelectric detection module automatically fits the designated detection part of the human body. The magnetic resonance detection module is adjusted to the corresponding detection working position to complete the detection preparation.

[0023] S3: Multi-dimensional data acquisition. The bioelectric detection module begins to collect bioelectric signals such as human electrocardiogram and electroencephalogram, and transmits them to the intelligent analysis and processing module after signal processing. At the same time, the magnetic resonance detection module performs magnetic resonance detection on local human tissues, generates magnetic resonance images and tissue metabolic data, and transmits them to the intelligent analysis and processing module.

[0024] S4: Data fusion analysis and diagnosis. The intelligent analysis and processing module preprocesses and extracts features from the collected bioelectrical and magnetic resonance data. It then performs fusion analysis through an AI diagnostic model and compares the data with benchmark data in the cloud-based health database to complete anomaly identification, preliminary diagnosis of common diseases, and risk level assessment. If a major data anomaly is detected, an audible and visual alarm is immediately issued through the interactive execution module and automatically pushed to the user's bound mobile terminal and remote medical platform to achieve emergency warning.

[0025] S5: Results output and solution push. The interactive execution module displays test data, magnetic resonance images, preliminary diagnostic information and risk assessment results through a touch screen, and broadcasts key information through voice. The solution generation module outputs personalized health monitoring, diet, exercise and basic intervention plans for users based on the diagnostic results.

[0026] S6: Data synchronization and remote support. The intelligent analysis and processing module uploads the test data, diagnosis results and risk assessment results to the cloud service module for encrypted storage through the communication module and incorporates them into the user's personal health record. If the user needs further professional diagnosis, they can initiate a remote consultation application through the interactive execution module. The remote medical docking unit immediately connects to the remote medical platform and pushes the user's health data to a professional doctor to realize remote consultation.

[0027] S7: Basic intervention execution. If the user confirms the execution of the generated health intervention plan, the multi-degree-of-freedom robotic arm of the motion execution unit will perform basic health intervention operations such as local massage and acupoint stimulation according to the plan. At the same time, the robot will set detection reminders according to the plan and remind the user to perform health monitoring at regular intervals.

[0028] S8: Health Data Tracking. The cloud service module integrates the current test data, diagnosis results, intervention implementation status, and the user's historical health data to achieve continuous tracking and management of health data, providing data support for subsequent diagnosis and intervention.

[0029] 1. This invention integrates the principles of bioelectric detection and 0.5-1.5T low-field magnetic resonance detection technology to achieve dual non-invasive detection of human organ functional characteristics (bioelectric signals) and structural characteristics (magnetic resonance signals). It breaks through the limitation of the single detection dimension of existing home medical devices and can complete the preliminary screening and diagnosis of common diseases in multiple systems such as cardiovascular, nervous system, and musculoskeletal system with a diagnostic accuracy of ≥88%, greatly improving the professionalism of home health monitoring.

[0030] 2. The magnetic resonance detection module of this invention adopts a portable low-field magnetic resonance detection unit design, with a magnetic field strength of 0.5-1.5T. It is small in size, low in cost, and has no significant radiation risk, with a volume ≤0.8m³. 3 Weighing ≤50kg, the cost is reduced by more than 90% compared to traditional high field strength magnetic resonance imaging (MRI) equipment, solving the technical problem that traditional MRI equipment cannot be used in homes, and making non-invasive imaging detection possible in home settings.

[0031] 3. This invention incorporates an AI diagnostic model and a health data fusion algorithm, ResNet+LSTM, into a deep learning AI diagnostic model. Combined with massive amounts of clinical data in the cloud, it can achieve automated and professional analysis and preliminary diagnosis of test data. At the same time, it can generate personalized health management and basic intervention plans, realizing a closed-loop service of "detection-diagnosis-intervention", which greatly improves the convenience and professionalism of family health management.

[0032] 4. This invention is equipped with an autonomous movement module, a voice / touch dual interaction unit, and an automatic sensor attachment function, enabling autonomous movement, active detection, sensor-assisted attachment, and basic health intervention functions. It is suitable for use by all family members, especially the elderly, children, and other people with limited mobility or weak operational abilities. At the same time, the device is small in size and flexible in layout, adapting to various family spaces and improving the device's versatility.

[0033] 5. This invention constructs a two-tier service system of "local detection + cloud management". The cloud enables encrypted storage, continuous tracking and trend analysis of health data. At the same time, it connects to a telemedicine platform to enable remote doctor consultations, making up for the lack of professionalism in home initial diagnosis, forming a two-tier medical service system of "home initial diagnosis and treatment + remote professional consultation", thereby improving the security of family health management.

[0034] 6. The robot of this invention has functions such as automatic recharging when the battery is low, obstacle avoidance navigation, and posture detection guidance, realizing fully intelligent operation without much human intervention. It meets the usage needs of home scenarios and promotes the popularization and application of intelligent medical devices in homes.

[0035] 7. The core modules of this invention all adopt domestically produced technologies and components, the industrial chain is independent and controllable, the overall cost of the equipment is low, and subsequent maintenance is convenient. It is suitable for large-scale promotion and application, promotes the popularization of intelligent medical devices to the home, and helps the construction of grassroots health management system. Attached Figure Description

[0036] The present invention will now be described in further detail with reference to the accompanying drawings, which are all schematic structural diagrams:

[0037] Figure 1 Overall structural block diagram of the present invention

[0038] This diagram is a schematic block diagram of the robot's system architecture, which is laid out in layers of "hardware module - software module - cloud module". It clearly shows the connection relationship between the bioelectric detection module, magnetic resonance detection module, intelligent analysis and processing module, interactive execution module, communication module, mobility module, charging module and cloud service module.

[0039] Each module is clearly labeled, and the connection lines indicate the direction of data transmission. The intelligent analysis and processing module is the core hub, which is bidirectionally connected to all other hardware modules. The communication module is the only data transmission channel between the intelligent analysis and processing module and the cloud service module.

[0040] The block diagram adopts a standardized rectangular design, the module names are labeled in bold, the connecting lines are solid lines, and the data transmission direction is indicated by arrows.

[0041] Figure 2. Block diagram of bioelectric detection module

[0042] The diagram illustrates the internal components and signal processing flow of the bioelectric detection module, from left to right: "flexible sensor array → signal amplification unit → filtering unit → A / D conversion unit → signal output terminal";

[0043] The core technical parameters of each component (amplification factor, filter cutoff frequency, sampling accuracy / frequency) are labeled. The signal output terminal is connected to the intelligent analysis and processing module, and the signal processing flow is labeled with a unidirectional arrow.

[0044] The components in the attached diagram are rectangles of equal size, and the parameters are labeled below the components.

[0045] Figure 3. Block diagram of magnetic resonance detection module

[0046] This diagram illustrates the internal components and imaging process of the magnetic resonance imaging module, which is divided into four parts: "transmitter, detector, receiver, and processor". The transmitter is the radio frequency transmitting unit, the detector is the low field strength permanent magnet, the receiver is the signal receiving coil, and the processor is the imaging processing unit.

[0047] The core technical parameters of each component (magnetic field strength, radio frequency pulse frequency, imaging resolution) are labeled. The output of the imaging processing unit is divided into two paths: "image data" and "metabolic data", both of which are connected to the intelligent analysis and processing module.

[0048] The attached diagram uses a modular layout, with each part separated by dashed boxes, and the core parameters are marked in bold.

[0049] Figure 4. Block diagram of the intelligent analysis and processing module

[0050] The diagram illustrates the internal functional modules of the intelligent analysis and processing module. The core is the AI ​​diagnosis module, surrounded by data preprocessing module, feature extraction module, solution generation module, and local health benchmark database.

[0051] The data preprocessing module is the data input end, connected to the bioelectric and magnetic resonance detection module; the solution generation module is the result output end, connected to the interactive execution module; and the AI ​​diagnosis module is bidirectionally connected to the cloud service module.

[0052] The core algorithms and diagnostic accuracy of the AI ​​diagnostic model are labeled, and the attached figures adopt a central radial layout.

[0053] Figure 5. Work Method Flowchart

[0054] This diagram illustrates the entire workflow of the robot, using a step-by-step linear layout, arranged sequentially from "S1 Human-Computer Interaction Startup" to "S8 Health Data Tracking," with each step labeled with its core operational content.

[0055] Step S4 sets up the branch process, labeled "Major Anomaly → Audible and Visual Alarm + Terminal Push"; Step S6 sets up the optional process, labeled "Remote Consultation Application → Connect to Remote Medical Platform".

[0056] The steps in the attached diagram are shown in rectangular boxes, the flow is indicated by unidirectional arrows, branch flows are indicated by dashed arrows, and the core operation content is marked in Song typeface, size 5.

[0057] Figure 6. Schematic diagram of the overall appearance structure of the robot

[0058] This image is a three-dimensional schematic diagram of the robot's appearance, presented from a frontal view, showing the robot's overall shape and the layout of its external components.

[0059] The head features a 10.1-inch touchscreen display and a voice pickup hole; the front of the device has a magnetic resonance detection module window and a bioelectric sensor storage compartment; the side of the device has a robotic arm; the bottom has a motion drive wheel and an obstacle avoidance sensor; and the back has a wireless charging receiver.

[0060] The location and core dimensions of each external component are marked (body height 20cm, width 20cm, thickness 10cm). The attached diagram uses a line drawing design and the component labels are clear.

[0061] Figure 7 Schematic diagram of the attachment location of the bioelectric sensor array

[0062] This diagram shows the front and side views of the human body, and marks the standard attachment positions of the bioelectric sensor array: the electrocardiogram sensor is attached to the precordial region of the chest (6-electrode layout), the electroencephalogram sensor is attached to the forehead and temporal side of the head (8-electrode layout), the electromyogram sensor is attached to the biceps brachii of the upper limb and the quadriceps femoris of the lower limb, and the electrodermal sensor is attached to the inner side of the wrist.

[0063] The sensor is represented by a circle, with the attachment position marked by dashed lines. The number and layout of the electrodes are clearly marked, and the human figure in the attached diagram uses a standardized line design.

[0064] Figure 8. Schematic diagram of the working principle of the magnetic resonance detection module

[0065] This diagram illustrates the core principles of low-field magnetic resonance imaging (MRI), including the static magnetic field generated by the low-field permanent magnet, the radio frequency pulses emitted by the radio frequency transmitting unit, the MRI echo signal from human tissue, and the signal acquisition process of the signal receiving coil. The diagram shows the direction of the static magnetic field, the frequency of the radio frequency pulse, and the direction of the echo signal transmission, illustrating the process of generating MRI images. The diagram is designed with schematic illustrations, and the principles are clearly and concisely explained. Detailed Implementation

[0066] The technical solution of the present invention will be described in detail below with reference to specific embodiments. All embodiments are based on the technical solution of the present invention, and the technical parameters meet the practicality requirements of the patent application and have no technical features that exceed the level of the prior art.

[0067] Example 1: Hardware Configuration and Module Parameters of a Family Doctor Robot

[0068] A family doctor robot based on bioelectricity and magnetic resonance principles. The robot body is 20cm high, 20cm wide, 10cm thick, and weighs 2kg. It is suitable for home living room, bedroom, and other scene layouts. The specific configuration and technical parameters of each module are as follows:

[0069] 1. Bioelectric detection module: It adopts a 0.4mm thick conductive silicone flexible sensor array with a stretchability of 180%, and has a layout of 6 electrodes for ECG, 8 electrodes for EEG, 4 electrodes for EMG, and 2 electrodes for ED. The signal amplification unit has a magnification factor of 3000 times, the second-order active low-pass filter unit has a cutoff frequency of 0.05-80Hz, and the A / D conversion unit has a sampling accuracy of 18bit and a sampling frequency of 1.5kHz, which can realize the synchronous acquisition of bioelectric signals from multiple sites.

[0070] 2. Magnetic Resonance Imaging Module: 0.8T low field strength permanent magnet, RF transmitting unit frequency 21MHz, phased array signal receiving coil, imaging processing unit output image resolution 512×512 pixels, single-site imaging time 3min, can perform non-invasive imaging detection of human brain, heart, knee joint and other parts.

[0071] 3. Intelligent Analysis and Processing Module: It adopts a domestically produced high-performance embedded chip with a main frequency of 2.0GHz, 8GB of memory, and 128GB of storage; it has a built-in ResNet+LSTM fusion deep learning AI diagnostic model, and a local health benchmark database storing 500,000 basic health data. The diagnostic accuracy rate for common diseases such as hypertension, insufficient cerebral blood supply, osteoarthritis, and frozen shoulder is 88%.

[0072] 4. Interactive Execution Module: 10.1-inch IPS high-definition touch screen with a resolution of 1920×1200; voice interaction unit with a pickup distance of 4m, recognition accuracy of 96%, supporting Mandarin, Cantonese, and Sichuan dialect; robotic arm with a rated load of 2.5kg, repeatability accuracy of ±0.08mm, adjustable massage intensity range of 5-28N, and acupoint stimulation accuracy of ±1.5mm; infrared human body positioning sensor with a detection accuracy of ±0.03m.

[0073] 5. Communication Module: Integrates IEEE802.11ac wireless WiFi, Bluetooth 5.0, and NR Sub-6G 5G communication modules, with a data transmission rate of 150Mbps, supporting networking with mobile terminals such as mobile phones and tablets, as well as wearable detection devices.

[0074] 6. Cloud Service Module: Deployed on Alibaba Cloud server cluster, the health database stores 600,000 health baseline data and disease case data, the AI ​​model is iterated once a month, and it connects to the telemedicine platforms of 20 top-tier hospitals in China, using SHA-256 encryption algorithm and data anonymization technology.

[0075] 7. Mobility and charging module: Dual drive wheels, movement speed 0.1-0.45m / s, laser obstacle avoidance sensor detection distance 0.05-2.8m, SLAM navigation and positioning accuracy ±0.04m; 30W wireless charging unit, charging efficiency 88%, automatic recharging when the battery is low at 20%, and a full charge battery life of 8 hours.

[0076] The robot in this embodiment works in concert with its various modules to achieve non-invasive, multi-dimensional health monitoring and preliminary diagnosis in a home setting. All components are domestically produced, the industrial chain is self-reliant and controllable, and the overall production cost of the equipment is ≤50,000 yuan, making it suitable for large-scale home promotion and application.

[0077] Example 2: Application Process of Robots in Cardiovascular and Cerebrovascular Health Detection in Middle-aged and Elderly People

[0078] Using the family doctor robot in Example 1, a cardiovascular health check was performed on a 65-year-old user who had a 5-year history of hypertension. The specific workflow is as follows:

[0079] 1. S1 Human-Computer Interaction Startup: When the user is 3m in front of the robot, they can issue the command "cardiovascular and cerebrovascular health test" via voice. The voice interaction unit recognizes and transmits the command to the intelligent analysis and processing module. After parsing, the module generates a cardiovascular and cerebrovascular specific test command and simultaneously wakes up each test module.

[0080] 2. S2 Detection Preparation: The mobile module drives the robot to move autonomously from the charging position to the user's side, and the obstacle avoidance sensor avoids furniture obstacles along the way, with a positioning accuracy of ±0.03m; the detection assistance unit guides the user to sit on the detection chair through voice and light, maintaining the standard detection posture with the upper body upright; the robotic arm takes out the bioelectric sensor array from the sensor storage compartment and accurately attaches it to the user's chest precordial area (ECG), forehead (EEG), and inner wrist (TEG), with an attachment accuracy of ±1mm; the magnetic resonance detection module is adjusted to 15cm directly in front of the user's chest to complete the detection working position calibration.

[0081] 3. S3 Multi-dimensional Data Acquisition: The bioelectric detection module is activated, acquiring the user's ECG, EEG, and ductal nerve signals at 18-bit precision and a sampling frequency of 1.5kHz. After 3000x amplification and 0.05-80Hz filtering, the signals are converted into digital signals and transmitted to the intelligent analysis and processing module. Simultaneously, the magnetic resonance detection module activates 0.8T low-field magnetic resonance detection, transmitting 21MHz radio frequency pulses. The phased array coil acquires magnetic resonance echo signals from the heart and brain tissues, generating 512×512 pixel images of cardiac soft tissue, brain structure, and tissue metabolic data within 3 minutes, which are then synchronously transmitted to the intelligent analysis and processing module.

[0082] 4. S4 Data Fusion Analysis and Diagnosis: The intelligent analysis and processing module standardizes and de-redundantizes the bioelectrical data, extracting mild ventricular premature beats (5 times / minute) from the electrocardiogram (ECG) signal. The electroencephalogram (EEG) signal shows no obvious abnormalities, and the electrodermal signal indicates normal autonomic nerve function. Feature extraction from the magnetic resonance imaging (MRI) images reveals mild thickening of the left ventricular wall (12mm thick), no obvious stenosis of cerebral blood vessels, and normal tissue metabolism. The AI ​​diagnostic model integrates the bioelectrical and MRI data and compares it with the baseline data of 65-year-old hypertensive individuals in the cloud-based health database, diagnosing the condition as "early stage of hypertensive cardiac changes, cardiac function class I, low risk," with no major abnormalities and no alarms triggered.

[0083] 5. S5 Result Output and Solution Push: The robot's 10.1-inch touchscreen displays the user's electrocardiogram waveform, brain / cardiac MRI images, and various bioelectrical indicators in high definition, while simultaneously broadcasting the diagnosis results and risk level via voice. The solution generation module outputs a personalized intervention plan based on the diagnosis results and the user's history of hypertension: ① Monitor blood pressure / ECG at 8 am and 8 pm daily; ② Low-salt diet (daily salt intake ≤5g), reduce high-fat and high-sugar foods; ③ Walk for 30 minutes daily, avoid strenuous exercise; ④ Massage legs for 10 minutes (15N intensity) before bed each night, stimulating acupoints such as Zusanli and Sanyinjiao.

[0084] 6. S6 Data Synchronization and Remote Support: The intelligent analysis and processing module, through the 5G communication module, encrypts and uploads the test data, MRI images, and diagnostic results to the cloud service module, which then incorporates them into the user's personal health record. The cloud integrates the data with the user's health data for the past 6 months; the user does not need remote consultation and has not initiated an application.

[0085] 7. S7 Basic Intervention Execution: The user confirms the execution of the intervention plan through the touch screen. The 6-DOF robotic arm stimulates the user's Zusanli and Sanyinjiao acupoints on the legs with a precision of ±1.5mm, while simultaneously performing a 10-minute kneading massage with a force of 15N. The robot also sets timed detection reminders locally and in the cloud, automatically reminding the user to perform the detection at 8 am and 8 pm every day.

[0086] 8. S8 Health Data Tracking: The cloud service module compares the current test data with the user's historical data to generate a cardiovascular health trend curve, showing that the user's left ventricular wall thickness has not changed significantly compared to 3 months ago, the number of premature ventricular contractions has decreased slightly, and the health trend is stable; the cloud synchronizes the trend analysis results to the user's mobile APP to provide data support for subsequent diagnosis and intervention.

[0087] In this embodiment, the robot automates the entire process of detection, diagnosis, and intervention, making it easy to operate, providing accurate test results, and offering personalized intervention plans. This achieves closed-loop management of cardiovascular and cerebrovascular health for middle-aged and elderly users, meeting the actual needs of family health management.

[0088] Example 3: Application process of robots in detecting muscle damage after exercise in adolescents

[0089] Using the family doctor robot in Example 1, a specific detection of lower limb muscle injury after exercise was performed on a 16-year-old user. The user experienced right knee pain due to basketball. The specific workflow is as follows:

[0090] 1. S1 Human-Computer Interaction Startup: The user clicks "Musculoskeletal Injury Detection - Right Knee Joint" on the robot's touch screen. The intelligent analysis and processing module parses the command and generates a specific detection command.

[0091] 2. S2 Detection Preparation: The robot moves to the user's side and guides the user to sit with their right knee naturally extended. The robotic arm attaches the electromyography sensor to the quadriceps and patellar ligament around the user's right knee joint, and the magnetic resonance imaging module is adjusted to 10cm in front of the right knee joint.

[0092] 3. S3 Data Acquisition: Acquire electromyographic signals around the right knee joint (sampling accuracy 18bit, frequency 1.5kHz), while the magnetic resonance imaging module performs 3 minutes of imaging on the right knee joint, generating a 512×512 pixel soft tissue image of the knee joint.

[0093] 4. S4 Analysis and Diagnosis: After preprocessing, abnormal discharge characteristics of electromyography signals were extracted. Magnetic resonance imaging showed mild edema of the patellar ligament in the right knee joint without tear. The AI ​​diagnostic model diagnosed it as "minor patellar ligament injury after exercise, low risk".

[0094] 5. S5 Solution Push: Display the test results on the screen and output the intervention plan: ① Stop strenuous sports such as basketball for 1 week; ② Massage the right knee joint twice a day (10N intensity, 10 minutes each time); ③ Apply local heat to help reduce swelling.

[0095] 6. S6-S8: The detection data is uploaded to the cloud for archiving. After the user confirms the intervention plan, the robotic arm completes the knee joint massage. The cloud sets a reminder for a follow-up examination one week later, and tracks the recovery of muscle damage.

[0096] In this embodiment, the robot can quickly perform non-invasive detection and diagnosis of muscle damage after exercise in adolescents, and promptly output personalized intervention plans to prevent the damage from worsening. It is suitable for rapid screening of sports injuries in home settings.

[0097] The above-described embodiments are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the design concept of the present invention should be included within the scope of protection of the present invention.

[0098] Instruction manual illustrations:

[0099] Figure 1 Overall structural block diagram of the present invention

[0100] Structural Diagram Description: The layout is layered and modular, with the intelligent analysis and processing module as the core hub. All hardware modules are bidirectionally connected to the core, and the communication module serves as the sole data channel between the core and the cloud. Arrows indicate the direction of data transmission. Core modules: Bioelectrical Detection Module, Magnetic Resonance Detection Module, Intelligent Analysis and Processing Module, Interactive Execution Module, Communication Module, Mobility Module, Charging Module, and Cloud Service Module.

[0101] Figure 2. Block diagram of bioelectric detection module

[0102] Structural Diagram Description: A linear flow layout, from left to right: signal acquisition → processing → output. Core technical parameters of each unit are labeled, and unidirectional arrows indicate the signal processing flow. Core components: Flexible sensor array, signal amplification unit (300x), filtering unit (0.05-80Hz), A / D conversion unit (18bit / 1.5kHz), and signal output terminal.

[0103] Figure 3. Block diagram of magnetic resonance detection module

[0104] Structural Diagram Description: The system features a four-terminal modular layout (transmit / detect / receive / process), with dashed boxes separating each terminal. Core technical parameters are labeled, and data types are labeled for both output terminals. Core components: RF transmitting unit (21MHz), low-field permanent magnet (0.8T), phased array signal receiving coil, imaging processing unit (512×512 pixels / 3min), image data output terminal, and metabolic data output terminal.

[0105] Figure 4. Block diagram of the intelligent analysis and processing module

[0106] Structural Diagram Description: The layout is radial, with the AI ​​diagnostic module at its core, surrounded by other modules. Core algorithms and performance parameters are labeled, and bidirectional arrows indicate data interaction. Core annotations include: Data Preprocessing Module, Feature Extraction Module, AI Diagnostic Module (ResNet+LSTM / 88% accuracy), Solution Generation Module, Local Health Benchmark Database, Data Input Terminal, Result Output Terminal, and Cloud Interaction Terminal.

[0107] Figure 5. Work Method Flowchart

[0108] Structure Diagram Explanation: The layout is linear, with 8 core steps arranged sequentially. Branch processes are marked with dashed arrows (major anomaly alarms), optional processes are marked with parentheses (remote consultations), and unidirectional arrows indicate workflows. Core steps: S1 Human-Computer Interaction Startup, S2 Detection Preparation, S3 Multi-Dimensional Data Acquisition, S4 Data Fusion Analysis and Diagnosis (Major Anomalies → Audible and Visual Alarms + Terminal Push), S5 Result Output and Solution Push, S6 Data Synchronization and Remote Support (Remote Consultation Request), S7 Basic Intervention Execution, S8 Health Data Tracking.

[0109] Figure 6. Schematic diagram of the overall appearance structure of the robot (front view)

[0110] Structural Diagram Description: The design features proportional, three-dimensional lines, with a body size adapted to home environments. External components are arranged according to actual function, with core dimensions and component names clearly marked. There are no decorative patterns. Core components: Body (20cm L x 20cm H x 10cm D), 10.1-inch IPS touchscreen display, voice pickup port, magnetic resonance imaging window, bioelectric sensor storage compartment, laser obstacle avoidance sensor, and wireless charging receiver. The front of the body features a data display screen, with circular detection electrodes at the bottom, connected to the body via data transmission lines. A data transmission cable is located on the side of the body, enabling data transmission and interaction with external devices. The overall structure is compact and suitable for home health monitoring and data transmission.

[0111] Figure 7 Schematic diagram of the attachment location of the bioelectric sensor array

[0112] Structural Diagram Description: Standardized line drawing of the human body from the front, with sensors represented by circles, dotted lines marking the attachment areas, and the number and layout of electrodes labeled to fit the human physiological structure. Core annotations: ECG sensor (precordial region of the chest / 6 electrodes), EEG sensor (forehead / 8 electrodes), EMG sensor (upper limb / lower limb / 4 electrodes), TESA sensor (inner side of the wrist / 2 electrodes), and the outline of the human body from the front.

[0113] Figure 8. Schematic diagram of the working principle of the portable magnetic resonance imaging module.

[0114] Diagram Explanation: This diagram illustrates the principle layout, labeling the directions of the magnetic field, radio frequency pulse, and echo signal. It simplifies the core process of portable magnetic resonance imaging (MRI). The device is 20cm wide, 15cm high, and 15cm thick. No complex formulas are used; it is intuitive and easy to understand. Key elements: Low-field permanent magnet, static magnetic field (0.8T), radio frequency transmitting coil (21MHz), radio frequency pulse, human detection area (chest / brain / knee joint), MRI echo signal, signal receiving coil, imaging processing unit, and MRI image.

Claims

1. A family doctor robot based on the principle of bioelectricity and magnetic resonance, characterized in that, The robot includes a main body, which integrates a bioelectric detection module, a magnetic resonance detection module, and an intelligent analysis and processing module. An interactive execution module is located on the outside of the main body. The intelligent analysis and processing module is connected to a cloud service module through a communication module. A mobility module is located at the bottom of the main body, and a charging module is located at the back. (1) The bioelectric detection module is used to collect bioelectric signals of electrocardiogram, electroencephalogram, electromyogram and electrodermal conduction on the human body surface, and is equipped with a wearable flexible bioelectric sensor array that is compatible with the detection of the chest, head and limbs of the human body. (2) The magnetic resonance detection module is a portable low field strength magnetic resonance detection unit with a magnetic field strength of 0.5-1.5T. It is used to perform non-invasive magnetic resonance imaging detection and tissue metabolic signal acquisition on local human tissues, so as to achieve preliminary screening of the structure and function of soft tissues and organs. (3) The intelligent analysis and processing module is the core control unit of the robot. It has a built-in AI diagnostic model and health data fusion algorithm to receive and preprocess the detection data of the bioelectric detection module and the magnetic resonance detection module, and complete data feature extraction, abnormal identification, preliminary diagnosis of common diseases and risk level assessment. (4) The interactive execution module is used to realize human-computer interaction, receive detection instructions, display detection results and diagnostic information, and perform basic health intervention operations. (5) The cloud service module is used for encrypted storage of health data, iterative updates of AI diagnostic models, and connection to remote medical platforms to achieve data synchronization and support for remote consultation; (6) The mobile module is based on the SLAM algorithm to realize autonomous movement, obstacle avoidance and positioning in the home scene; the charging module is a wireless charging unit that supports automatic recharging when the battery is low.

2. The home doctor robot based on bioelectric and magnetic resonance principles according to claim 1, characterized in that, The bioelectric detection module includes a flexible sensor array, a signal amplification unit, a filtering unit, and an A / D conversion unit. The flexible sensor array is made of conductive silicone material with a thickness of 0.3-0.5 mm and an elongation of ≥150%. The signal amplification unit has a magnification of 1000-5000 times, the filtering unit is a second-order active low-pass filter circuit with a cutoff frequency of 0.05-100 Hz, and the A / D conversion unit has a sampling accuracy of ≥16 bits and a sampling frequency of 1-2 kHz. The signal amplification unit and the filtering unit perform noise reduction and amplification processing on the acquired raw bioelectric signals. After the A / D conversion unit converts the analog signal into a digital signal, it is transmitted to the intelligent analysis and processing module.

3. The home doctor robot based on bioelectric and magnetic resonance principles according to claim 1, characterized in that, The magnetic resonance imaging (MRI) module includes a low-field magnet, a radio frequency (RF) transmitting unit, a signal receiving unit, and an imaging processing unit. The RF magnet has a magnetic field strength of 0.5-1.5T, suitable for non-invasive home testing. The RF transmitting unit emits RF pulses towards the human body to be tested, and the signal receiving unit collects the MRI echo signals from the human tissue. After processing by the imaging processing unit, MRI images and tissue metabolic data are generated and transmitted to the intelligent analysis and processing module. The RF transmitting unit emits RF pulses at a frequency of 12-64MHz, the signal receiving unit is a phased array coil, and the imaging processing unit outputs an image resolution ≥256×256 pixels with an imaging time ≤5min / site.

4. The home doctor robot based on bioelectric and magnetic resonance principles according to claim 1, characterized in that, The intelligent analysis and processing module includes a data preprocessing module, a feature extraction module, an AI diagnosis module, and a solution generation module. The AI ​​diagnosis module is based on a ResNet+LSTM fusion deep learning algorithm. The data preprocessing module standardizes and removes redundancy from bioelectrical digital signals and magnetic resonance data. The feature extraction module extracts abnormal feature values ​​from the data and compares them with a health benchmark database. The AI ​​diagnosis module, based on a deep learning algorithm, fuses bioelectrical signal features and magnetic resonance features to complete the preliminary diagnosis and risk level assessment of common diseases such as cardiovascular and cerebrovascular diseases, minor neurological disorders, and musculoskeletal diseases. The preliminary diagnosis accuracy is ≥88%, and the risk level is divided into low, medium, and high levels. Based on the diagnostic results, the solution generation module outputs personalized health monitoring, diet, exercise, and basic intervention plans.

5. The home doctor robot based on bioelectric and magnetic resonance principles according to claim 1, characterized in that, The interactive execution module includes a touch screen, a voice interaction unit, a motion execution unit, and a detection assistance unit. The touch screen and voice interaction unit enable bidirectional human-computer interaction, supporting voice / touch command input, visualization of detection results, and health knowledge push. The motion execution unit is a multi-degree-of-freedom robotic arm that can assist in basic health interventions such as sensor attachment, local massage, and acupoint stimulation. The touch screen is a 10.1-inch IPS high-definition screen with a resolution of 1920×1200. The voice interaction unit supports far-field voice pickup with a pickup distance ≤5m, a recognition accuracy ≥95%, and supports multiple languages ​​and dialects. The robotic arm has a rated load ≥2kg and a repeatability accuracy of ±0.1mm. The detection assistance unit includes an infrared human body positioning sensor and a detection posture guidance unit to guide users to complete standard detection postures and improve the accuracy of detection data.

6. The home doctor robot based on bioelectric and magnetic resonance principles according to claim 1, characterized in that, The communication module integrates a wireless WiFi module (IEEE 802.11a / b / g / n / ac), a Bluetooth module (Bluetooth 5.0), and a 5G communication module (NR Sub-6G), with a data transmission rate of ≥100Mbps. It supports simultaneous networking of multiple devices and wireless data transmission between robots and cloud service modules, wearable detection devices, and mobile terminals, enabling real-time synchronization and remote viewing of health data.

7. The family doctor robot based on bioelectricity and magnetic resonance principles according to claim 1, characterized in that, The cloud service module includes a health database, a model update unit, a telemedicine connection unit, and a blockchain data security unit. The health database stores ≥500,000 sets of health baseline data and disease case data for different age groups, genders, and physical conditions. The data security unit uses the SHA-256 encryption algorithm and supports data anonymization and tiered access. The health database stores bioelectrical baseline data, magnetic resonance imaging baseline data, and common disease case data for different population groups. The model update unit continuously optimizes the diagnostic accuracy of the AI ​​diagnostic model based on massive amounts of clinical data. The telemedicine connection unit can upload test data and preliminary diagnostic results to the telemedicine platform, enabling connection with professional doctors and supporting remote consultations. The data security unit uses blockchain encryption technology to encrypt, store, and transmit user health data, ensuring data privacy.

8. The family doctor robot based on bioelectricity and magnetic resonance principles according to claim 1, characterized in that, The robot body is also equipped with a movement module and a charging module. The movement module includes dual drive wheels, a laser obstacle avoidance sensor, a SLAM navigation unit, and an attitude sensor. The drive wheels have a movement speed of 0.1-0.5 m / s, the obstacle avoidance sensor has a detection distance of 0.05-3 m, and a positioning accuracy of ±0.05 m. The robot can move and locate autonomously in the home environment and can actively approach the user to complete the detection service. The charging module is a wireless charging unit that supports automatic recharging when the battery is low. The charging power is 15-30W and the charging efficiency is ≥85%. The robot will automatically trigger recharging when the battery level is 20%.

9. A working method of a family doctor robot based on the bioelectric and magnetic resonance principle, characterized in that, Includes the following steps: S1: Human-computer interaction is initiated. The user inputs detection requirements to the interaction execution module via voice / touch. The intelligent analysis and processing module receives and parses the detection requirements and generates detection instructions. S2: Detection preparation, the mobile module moves the robot body closer to the user, the detection auxiliary unit guides the user to complete the standard detection posture, the flexible sensor array of the bioelectric detection module fits the designated detection part of the human body, and the magnetic resonance detection module is adjusted to the detection working position. S3: Multi-dimensional data acquisition. The bioelectric detection module collects human bioelectric signals, processes them, and transmits them to the intelligent analysis and processing module. The magnetic resonance detection module performs magnetic resonance detection on local human tissues, generates signals and metabolic data, and transmits them to the intelligent analysis and processing module. S4: Data fusion analysis and diagnosis. The intelligent analysis and processing module preprocesses and extracts features from the data. It then uses an AI diagnostic model to fuse and analyze the data and compares it with benchmark data in the cloud to complete anomaly identification, preliminary diagnosis, and risk assessment. If a major anomaly is identified, an audible and visual alarm is immediately issued and pushed to the user's mobile terminal and remote medical platform. S5: Results output and solution push. The interactive execution module displays test results, preliminary diagnostic information and risk assessment results through a touch screen / voice. The solution generation module outputs personalized health management and basic intervention solutions. S6: Data synchronization and remote support. The intelligent analysis and processing module uploads the detection data and diagnostic results to the cloud service module for storage through the communication module. If needed by the user, the remote medical docking unit connects to the remote medical platform to realize remote doctor consultation. S7: Basic intervention execution. If the user confirms the execution of the intervention plan, the action execution unit will complete basic health intervention operations such as massage, acupoint stimulation, and detection reminders according to the plan. S8: Health data tracking. The cloud integrates current data with historical data to achieve continuous tracking and trend analysis, providing data support for subsequent diagnosis and intervention.

10. The method of claim 9, wherein, In step S4, if the AI ​​diagnostic model identifies a major anomaly in the data, it will immediately issue an audible and visual alarm through the interactive execution module and automatically push it to the user's bound mobile terminal and remote medical platform to achieve emergency warning; after step S7, the cloud service module will incorporate the test data, diagnostic results and intervention execution status into the user's health record to achieve continuous tracking and management of health data.

Images