Personalized injection system for rheumatoid joint multi-dimensional symptom monitoring

Abstract

The invention relates to a personalized injection system for rheumatoid joint multi-dimensional symptom monitoring, and belongs to the technical field of medical instruments, and the personalized injection system comprises a multi-dimensional pain sensor which can collect joint pressure, temperature, vibration and other data; the wearable continuous monitoring equipment can monitor physiological parameters such as joint movement and heart rate; and the two are wirelessly connected with the data analysis cloud platform. And the data analysis cloud platform generates personalized treatment suggestions through an artificial intelligence algorithm according to the collected data in combination with the individual information of the patient, and accurately regulates and controls the personalized drug delivery device. Meanwhile, the doctor-patient communication interaction terminal facilitates doctor-patient communication. The system can accurately monitor the illness state in multiple dimensions, realizes personalized accurate treatment, supports remote medical treatment, effectively improves the treatment effect of rheumatoid arthritis and the life quality of a patient, and assists the remarkable improvement of the medical service level.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically to a personalized injection system for multidimensional symptom monitoring of rheumatoid arthritis. Background Technology

[0002] Rheumatoid arthritis is a common chronic autoimmune disease that severely impacts patients' quality of life and physical health. Currently, the treatment of rheumatoid arthritis faces many challenges, and traditional treatment methods often fail to meet the individualized needs of patients.

[0003] In terms of symptom monitoring, rheumatoid arthritis symptoms are diverse and complex, including joint pain, swelling, morning stiffness, and limited mobility. These symptoms not only fluctuate over time but are also influenced by various factors, such as environmental factors like changes in temperature and humidity, as well as the patient's own lifestyle habits and diet. Traditional monitoring methods rely heavily on patient self-reporting, which is subject to subjective bias and cannot comprehensively and accurately reflect the condition.

[0004] In terms of drug treatment, there are many types of medications for rheumatoid arthritis, and different patients respond differently to them. Various side effects may also occur during treatment. Existing treatment plans often lack precision and personalization, failing to be optimized and adjusted according to the patient's specific condition and physical status.

[0005] With advancements in technology, intelligent monitoring devices and advanced data analysis techniques have provided new solutions for multi-dimensional symptom monitoring and personalized injection systems for rheumatoid arthritis. However, existing systems are complex, and the use of some components may be limited by practical operation and cost. Therefore, optimizing the system by removing some components while ensuring monitoring and treatment effectiveness can simplify the system and improve its usability and operability. Summary of the Invention

[0006] In view of this, the purpose of this invention is to propose a personalized injection system for multi-dimensional symptom monitoring of rheumatoid arthritis. By integrating components such as a multi-dimensional pain sensor, a personalized drug delivery device, a wearable continuous monitoring device, a data analysis cloud platform, and a doctor-patient communication and interaction terminal, this system addresses the problems existing in current rheumatoid arthritis treatment and monitoring technologies. Specifically, the multi-dimensional pain sensor and wearable continuous monitoring device overcome the subjective bias of traditional monitoring methods, providing comprehensive, objective, and accurate monitoring of joint pain and physiological parameters, offering data support for disease assessment. The personalized drug delivery device ensures the accuracy and safety of drug delivery, enabling personalized drug treatment, reducing side effects, and improving treatment efficacy. The data analysis cloud platform uses data analysis and artificial intelligence algorithms to generate personalized treatment suggestions, enhancing the targeting and effectiveness of treatment. The doctor-patient communication and interaction terminal promotes information exchange between doctors and patients, improving patient compliance and treatment outcomes, providing remote medical support, and improving the customization, monitoring, evaluation, and management services of clinical treatment plans. Through the collaborative work of these components, a more comprehensive, convenient, and efficient treatment system is provided for rheumatoid arthritis patients, improving treatment levels and patient experience, and driving innovation and progress in this field.

[0007] This invention is achieved through the following technical solution:

[0008] A personalized injection system for multidimensional symptom monitoring of rheumatoid arthritis includes a wearable continuous monitoring device fixedly installed on the outside of the patient's joint. The wearable continuous monitoring device is connected to a multidimensional pain sensor fixedly attached to the outside of the patient's joint via a wire. The wearable continuous monitoring device is also connected to a personalized drug delivery device and a doctor-patient communication and interaction terminal via a data analysis cloud platform through a wireless signal.

[0009] Furthermore, the multi-dimensional pain sensor includes a sensor housing, a pressure sensing chip for sensing changes in joint surface pressure is disposed at the center of the bottom of the sensor, temperature-sensitive resistors for measuring local joint temperature are evenly distributed around the pressure sensing chip, a vibration acceleration sensor is fixedly disposed on the outside of the sensor housing, a main control circuit is fixedly disposed in the center of the inside of the sensor housing, a wireless transmission module is fixedly disposed on one side of the main control circuit, and a battery is fixedly disposed at the bottom of the sensor housing, the battery being connected to the main control circuit via wires.

[0010] Furthermore, the personalized drug delivery device includes a main support frame, a micro drug storage device fixedly mounted on the top of the main support frame, a high-precision injection pump fixedly mounted on the bottom of the micro drug storage device, an intelligent control chip fixedly mounted on the side of the high-precision injection pump, the intelligent control chip using a dedicated microcontroller mounted on a circuit board next to the high-precision injection pump to control the drug injection volume, and a power module fixedly mounted on the bottom of the main support frame, the power module being connected to each electrical component via wires.

[0011] Furthermore, the high-precision injection pump includes a stepper motor, a lead screw, a piston, and an injection syringe. The stepper motor is connected to the lead screw via a coupling. The piston is located inside the injection syringe. The lead screw drives the piston to move within the injection syringe to achieve precise drug injection. A drug concentration sensor is fixedly installed near the outlet of the injection syringe. A flow sensor is installed on the pipe between the injection syringe and the injection needle.

[0012] Furthermore, the wearable continuous monitoring device includes: a wearable body, an accelerometer fixedly mounted near the joint movement area of ​​the wearable body, a gyroscope fixedly mounted on the side of the accelerometer fixedly mounted, a heart rate sensor fixedly mounted on the inner side of the wearable body in contact with the skin, a sleep monitoring sensor fixedly mounted on the side of the heart rate sensor, a main control unit fixedly mounted in the middle of the wearable body, a Bluetooth module fixedly mounted on the side of the main control unit, and a rechargeable battery fixedly mounted at the bottom of the wearable body, the rechargeable battery being connected to each power-consuming component via wires.

[0013] Furthermore, the data analysis cloud platform consists of a server cluster, data processing software, artificial intelligence algorithm modules, a database management system, and a network interface for communicating with external devices and connecting the server cluster to the Internet.

[0014] Furthermore, the doctor-patient communication and interaction terminal includes a terminal control panel. A display screen for displaying the doctor's replies and treatment suggestions is fixedly installed on the top of the terminal control panel. An input module is fixedly installed below the display screen. A communication module, a processor, and a power supply are fixedly installed inside the terminal control panel.

[0015] Furthermore, the method of using the above system includes the following steps:

[0016] S1 attaches a multi-dimensional pain sensor to the patient's painful joint. It collects joint pain-related data through its internal pressure sensing chip, temperature-sensitive resistor, and vibration acceleration sensor, and uses a wireless transmission module to send the data to a wearable continuous monitoring device or directly to a data analysis cloud platform.

[0017] S2, the patient wears a wearable continuous monitoring device that collects various physiological parameters through an accelerometer, gyroscope, heart rate sensor and sleep monitoring sensor, and transmits the data to a data analysis cloud platform via a Bluetooth module;

[0018] S3, the data analysis cloud platform receives data from multi-dimensional pain sensors and wearable continuous monitoring devices, and uses data processing software and artificial intelligence algorithm modules to analyze and process the data to generate personalized treatment suggestions;

[0019] S4. Doctors view information sent by the data analysis cloud platform through the doctor-patient communication interaction terminal, and input the adjusted treatment plan through the input module. The plan is then processed by the processor and transmitted to the data analysis cloud platform through the communication module. The data analysis cloud platform sends the adjustment information to the intelligent control chip of the personalized drug delivery device to control the drug delivery. At the same time, the drug concentration sensor and flow sensor of the personalized drug delivery device feed back the drug status data to the data analysis cloud platform.

[0020] The beneficial effects of this invention are as follows:

[0021] This invention utilizes a rheumatoid arthritis management system that integrates multi-dimensional symptom monitoring, personalized drug delivery, and real-time doctor-patient interaction. Through the collaboration of multi-dimensional pain sensors and wearable devices, it comprehensively captures joint pain and physiological changes, providing an objective reflection of the condition and avoiding subjective bias. A data analysis cloud platform integrates multi-source data to generate personalized treatment plans, which, combined with a precise drug delivery device, facilitates on-demand medication, reduces side effects, and enhances treatment targeting. A doctor-patient communication terminal promotes real-time information exchange, enabling timely adjustments to the treatment plan and improving patient compliance. The synergy of these components provides patients with convenient and efficient full-cycle management, ensuring treatment safety and improving treatment outcomes and patient experience. Attached Figure Description

[0022] Figure 1 A diagram illustrating the overall architecture of a multidimensional symptom monitoring and personalized injection system for rheumatoid arthritis.

[0023] Figure 2 This is a schematic diagram of a multi-dimensional pain sensor structure.

[0024] Figure 3 Rear view of the multi-dimensional pain sensor;

[0025] Figure 4 This is a schematic diagram of the structure of a wearable continuous monitoring device.

[0026] Figure 5 Top view of a wearable continuous monitoring device;

[0027] Figure 6This is a schematic diagram of a personalized drug delivery device.

[0028] Figure 7 A schematic diagram of the structure of a doctor-patient communication and interaction terminal;

[0029] Figure 8 A cross-sectional view of the internal workings of the doctor-patient communication and interaction terminal.

[0030] Explanation of reference numerals in the attached figures:

[0031] 100. Multi-dimensional pain sensor; 101. Sensor housing; 102. Pressure sensing chip; 103. Temperature-sensitive resistor; 104. Vibration accelerometer; 105. Main control circuit; 106. Wireless transmission module; 107. Battery; 200. Wearable continuous monitoring device; 201. Wearable casing; 202. Accelerometer; 203. Gyroscope; 204. Heart rate sensor; 205. Sleep monitoring sensor; 206. Main control unit; 207. Bluetooth module; 300. Personalization Drug delivery device; 301, main support; 302, miniature drug storage device; 303, stepper motor; 304, lead screw; 305, syringe; 306, piston; 307, intelligent control chip; 308, drug concentration sensor; 309, flow sensor; 310, power supply module; 400, data analysis cloud platform; 500, doctor-patient communication and interaction terminal; 501, terminal operating console; 502, display screen; 503, input module; 504, communication module; 505, processor; 506, power supply. Detailed Implementation

[0032] like Figure 1-6 As shown, one embodiment of the present invention provides a personalized injection system for multidimensional symptom monitoring of rheumatoid arthritis, comprising a multidimensional pain sensor, a wearable continuous monitoring device, a data analysis cloud platform, a personalized drug delivery device, and a doctor-patient communication and interaction terminal.

[0033] In this embodiment, the multi-dimensional pain sensor 100 is designed as a compact structure that can closely fit the patient's joint. Its core supporting component is a sensor mounting shell 101 made of medical-grade silicone, which ensures biocompatibility and can flexibly adapt to the shape of the joint. A high-precision pressure sensing chip 102 is securely mounted at the bottom center of the sensor mounting shell 101. Due to the piezoelectric properties of the material, even extremely slight pressure changes on the joint surface can be quickly detected by the pressure sensing chip 102, and converted into an electrical signal with a high sensitivity of 0.1 N / mV. The multi-dimensional pain sensor 100 includes a sensor mounting shell 101, a pressure sensing chip 102 for sensing pressure changes on the joint surface located at the bottom center of the sensor, temperature-sensitive resistors 103 for measuring local joint temperature evenly distributed around the pressure sensing chip 102, a vibration acceleration sensor 104 fixedly mounted on the outside of the sensor mounting shell 101, and a main control circuit 105 fixedly mounted in the center of the inside of the sensor mounting shell 101. Multiple temperature-sensitive resistors 103 are arranged around the pressure sensing chip 102 and are independently connected to the main control circuit 105 via high-precision metal wires. This design achieves:

[0034] 1. Temperature compensation function: The main control circuit 105 synchronously acquires pressure and temperature data. When the temperature change causes the zero point of the pressure sensing chip 102 to drift (temperature coefficient: ±0.05%FS / ℃), it performs compensation calculations in real time.

[0035] 2. Spatial Coordinated Layout: The temperature-sensitive resistors 103 are arranged around the joint to simultaneously monitor the temperature field gradient on the joint surface (radial temperature difference ≤ 0.3℃ / cm);

[0036] 3. Synchronous signal acquisition: Pressure and temperature signals are sampled through the same clock source (sampling interval 10ms).

[0037] Therefore, these resistors can accurately sense the local temperature of the joint, with a measurement accuracy of ±0.1℃. Once an abnormal temperature is detected, a corresponding electrical signal is generated promptly. The vibration acceleration sensor 104 is securely mounted on the outside of the sensor mounting housing 101, and its measurement range is ±10g, which is sufficient to effectively monitor various vibrations caused by joint movement or pain.

[0038] Inside the sensor housing 101, at its center, is a small multilayer printed circuit board, on which the main control circuit 105 performs crucial data processing. It rapidly acquires electrical signals from the pressure sensing chip 102, temperature-sensitive resistor 103, and vibration acceleration sensor 104 at a frequency of 100 times per second, using a Kalman filter algorithm to remove noise interference and extract effective data features. A wireless transmission module 106 is tightly connected to the main control circuit 105, employing Bluetooth 5.0 or ZigBee Pro 2017 technology to ensure stable and high-speed data transmission. At the bottom of the sensor housing 101, a rechargeable battery 107 is connected to the main control circuit 105 via low-resistance wires, providing continuous power to the entire sensor. Its 500mAh capacity ensures the sensor can operate continuously for 72 hours on a single full charge. All components work together. The pressure sensing chip 102, temperature sensitive resistor 103, and vibration acceleration sensor 104 are responsible for collecting multi-dimensional pain-related data. After processing by the main control circuit 105, the wireless transmission module 106 transmits the data, while the battery 107 ensures the operation of the device.

[0039] The wearable continuous monitoring device 200 uses a wearable sleeve 201 made of skin-friendly, breathable, and highly elastic medical Lycra fabric as its carrier, allowing patients to wear it comfortably without affecting normal joint movement. A high-precision accelerometer 202 is fixedly installed on the wearable sleeve 201 near areas of frequent joint movement, with a measurement accuracy of ±0.05 m / s². 2 It can capture real-time acceleration changes during joint movement; adjacent to the accelerometer 202, the gyroscope 203 is rationally positioned, and the two work together to accurately measure joint movement posture, with an angle measurement accuracy of ±0.5°. On the inner side of the wearable body 201, where it contacts the skin, a heart rate sensor 204 using advanced photoelectric reflective measurement technology is cleverly embedded, accurately measuring heart rate at a frequency of once per second, with an error controlled within ±1 beat / minute. To one side of the heart rate sensor 204, a sleep monitoring sensor 205, composed of a pressure sensor and a temperature sensor, is responsible for monitoring pressure distribution and changes in body surface temperature during the patient's sleep.

[0040] Although the multi-dimensional pain sensor has already collected pressure and temperature data, the reason why the wearable continuous monitoring device still needs to collect data repeatedly is as follows:

[0041] 1. Since the vast majority of patients with rheumatoid arthritis have sleep disorders, and sleep intervention can reduce the intensity of morning stiffness, it is necessary to monitor sleep status and then adjust medication according to the corresponding sleep status to ensure sleep and thus reduce the intensity of morning stiffness.

[0042] 2. The sleep monitoring sensor 205 consists of a pressure sensor and a temperature sensor. The pressure sensor monitors the pressure distribution in non-joint areas of the trunk (such as pressure on the lower back in the supine position) and analyzes sleep posture and turning frequency. The temperature sensor monitors the core temperature of the trunk (such as the axillary / sternal area) and assesses sleep stages through temperature changes.

[0043] 3. Comparison of characteristics between the multi-dimensional pain sensor 100 and the sleep monitoring sensor 205

[0044] Table 1 Feature Comparison Table

[0045] Monitoring Projects Multidimensional pain sensor 100 Sleep monitoring sensor 205 Pressure sampling site Local pressure on joint surface Trunk pressure distribution Temperature acquisition site Joint local temperature Core temperature of the torso Technical Purpose Assess joint inflammation status Analyze sleep quality

[0046] In the middle of the wearable sleeve 201, a 32-bit microprocessor forms the main control unit 206. This unit collects data from the accelerometer 202 and gyroscope 203 at a frequency of 50 times per second, while simultaneously acquiring data from the heart rate sensor 204 and sleep monitoring sensor 205. The data is preprocessed using a Fast Fourier Transform (FFT) algorithm to extract key data features before being stored in 1GB of internal memory. A Bluetooth module 207 is connected to the side of the main control unit 206. When within Bluetooth signal range, it transmits the stored data to the data analysis cloud platform 400 every 5 minutes at a transmission rate of 2Mbps. At the bottom of the wearable sleeve 201, a 1500mAh rechargeable battery is connected to all power-consuming components via wires, providing up to 48 hours of continuous power for the device. All components work together: the accelerometer 202, gyroscope 203, heart rate sensor 204, and sleep monitoring sensor 205 collect physiological parameters; the main control unit 206 processes and stores the data; the Bluetooth module 207 transmits the data; and the rechargeable battery provides power.

[0047] The main support 301 of the personalized drug delivery device 300 is made of high-strength, corrosion-resistant medical-grade aluminum alloy, ensuring the stability and reliability of the entire device. A transparent medical-grade plastic miniature drug reservoir 302 is fixedly mounted on top of the main support 301. Its 50mL capacity meets the needs of multiple drug injections, and its transparent design allows medical staff to easily check the remaining drug level. At the bottom of the miniature drug reservoir 302, the components of the high-precision infusion pump are tightly connected. A stepper motor 303 is rigidly connected to a lead screw 304 via a coupling. The stepper motor 303 has a step angle accuracy of ±0.05°. During operation, the lead screw 304 converts rotational motion into linear motion. The lead screw 304 has a pitch accuracy of ±0.01mm, thus precisely driving the piston 306 to move within the medical-grade glass syringe 305, achieving accurate drug injection.

[0048] On the circuit board next to the high-precision injection pump, a 32-bit microcontroller, serving as the intelligent control chip 307, is installed. This chip receives instructions from the data analysis cloud platform 400, achieving a drug injection volume control accuracy of ±0.01 mL. Near the outlet of the syringe 305, a drug concentration sensor 308, based on spectral analysis technology, is fixedly installed. Its concentration measurement accuracy is ±0.5%, enabling real-time monitoring of drug concentration. A flow sensor 309, operating on the principle of electromagnetic induction, is installed in the tubing between the syringe 305 and the needle, with a flow measurement accuracy of ±0.1 mL / min. Both sensors monitor the drug status in real time. At the bottom of the main support 301, a power module 310, consisting of a 2000mAh rechargeable lithium-ion battery, connects to various electrical components via wires, providing power for 36 hours of continuous operation. The intelligent control chip 307 controls the stepper motor 303 according to instructions, driving the lead screw 304 and piston 306 to inject the drug. The drug concentration sensor 308 and flow sensor 309 provide feedback data, ensuring safe and accurate injection.

[0049] The data analytics cloud platform 400 is deployed in a professional data center. Multiple high-performance servers within the server cluster are connected via a high-speed fiber optic network. Data storage servers are responsible for storing massive amounts of data, computing servers handle complex calculations, and application servers provide user interaction services. The data processing software installed on the computing servers includes the following modules:

[0050] Data cleaning module: Employs density-based noise spatial clustering (DBSCAN) algorithm to remove outlier data points. Specific parameters and operations are as follows:

[0051] Parameter settings for different data types:

[0052] Joint surface pressure (N): Neighborhood radius Eps = 0.5N, minimum number of points MinPts = 5, outliers are defined as pressure < 0N or > 10N;

[0053] Joint local temperature (°C): neighborhood radius Eps = 0.3°C, minimum number of points MinPts = 3, outliers are defined as temperatures <35°C or >40°C;

[0054] Vibration acceleration (g): neighborhood radius Eps = 0.5g, minimum number of points MinPts = 5, outliers are defined as vibration peak value < 0.1g or > 5g;

[0055] Turning frequency (times / hour): neighborhood radius Eps = 2 times / hour, minimum number of points MinPts = 3, outliers are judged as frequency < 0 times / hour or > 30 times / hour.

[0056] Anomaly handling rules:

[0057] Pressure and temperature data: If three consecutive points are abnormal, replace them with the average of the three normal points before and after (e.g., if the abnormality occurs at the 10th minute, use the average of the 7th-9th minute and the 11th-13th minute).

[0058] Vibration and turning data: Individual abnormal points are deleted directly, and the gaps are filled with the previous normal point (e.g., if the abnormal point is in the 5th minute, the data from the 4th minute is used instead).

[0059] Data standardization module: Z-score normalization is used (a conventional existing technology). The calculation steps are as follows: calculate the arithmetic mean of all data points (denoted as mean); calculate the standard deviation of each data point from the mean (denoted as standard deviation); normalized value = (original value - mean) / standard deviation.

[0060] Feature extraction module: Principal component analysis (PCA) algorithm (a conventional existing technique) is used to reduce data dimensionality. The specific operation is as follows:

[0061] The original data consisted of 10 dimensions, all from system devices, namely: (1) mean joint pressure (N, multi-dimensional pain sensor); (2) joint pressure fluctuation amplitude (N, multi-dimensional pain sensor); (3) mean local joint temperature (°C, multi-dimensional pain sensor); (4) local joint temperature gradient (°C / cm, multi-dimensional pain sensor); (5) peak-to-peak vibration acceleration (g, multi-dimensional pain sensor); and (6) maximum joint motion acceleration (m / s²). 2 (7) Average joint motion angular velocity (° / s, wearable device); (8) Turning frequency (times / hour, wearable device sleep sensor); (9) Core body temperature fluctuation (°C, wearable device sleep sensor); (10) Heart rate variability during sleep (ms, wearable device heart rate sensor).

[0062] Dimensionality reduction steps: Using SPSS software, after importing the 10-dimensional data, click "Analyze → Dimensionality Reduction → Factor Analysis," select the 10-dimensional data as "Variables," check "Principal Components," and set "Eigenvalue > 1." The system automatically filters out principal components with a cumulative variance contribution rate ≥ 85%. In practice, the first three principal components meet the requirements: Principal Component 1 (45%) reflects joint mechanical load, Principal Component 2 (25%) reflects joint inflammation, and Principal Component 3 (16%) reflects sleep quality, ultimately reducing the 10-dimensional data to 3 dimensions.

[0063] The artificial intelligence algorithm module integrated into the data processing software includes the following components:

[0064] Deep learning models:

[0065] Deep learning models are built based on convolutional neural networks (CNNs) and recurrent neural networks (RNNs). The specific steps are as follows:

[0066] Training set preparation: 7 days × 24 hours of data from 200 patients (1400 records in total). Each data point includes input and label: input is 3D principal component data after PCA dimensionality reduction (1 value per hour, 24 values ​​per day); label is the pain score (0-10 points) entered by the patient daily through the doctor-patient interaction terminal. The data is divided into a training set (1000 records), a validation set (300 records), and a test set (100 records) in a 7:2:1 ratio.

[0067] Model building (using Baidu AI Studio platform):

[0068] Create a new "Deep Learning Project" and import the training set (CSV format, each row contains 24 input values ​​+ 1 label);

[0069] Add the following layers in sequence: convolutional layer (3×3 convolutional kernels, 32 kernels, ReLU activation), bidirectional LSTM layer (64 hidden units), and fully connected layer (1 output unit, linear activation);

[0070] Training settings: Optimizer selected: Adam (learning rate: 0.001), batch size: 32, number of training epochs: 50, automatic stop when validation set error is ≤1 point.

[0071] Model output: Predicts patient pain scores; an error ≤ 1.2 is considered acceptable.

[0072] Data correlation analysis module:

[0073] The correlation between joint pain and activity was calculated using the Pearson correlation coefficient (a conventional technique): pressure and range of motion values ​​were collected at n time points; the covariance between pressure and range of motion values ​​was calculated; the correlation coefficient = covariance / (pressure standard deviation × range of motion standard deviation); the coefficient range is [-1, 1], and the larger the absolute value, the stronger the correlation between pain and activity.

[0074] Inflammation trend prediction: I = a·CRP + b·ΔT 24h

[0075] Where: I: Inflammation index (dimensionless, used to quantify the severity of inflammation, range 0-10);

[0076] CRP: C-reactive protein level (unit: mg / L, a clinical blood test indicator that reflects the degree of inflammatory activity);

[0077] ΔT 24h : Local temperature change of joint over 24 hours (unit: °C, calculated by subtracting the average joint temperature 24 hours ago from the current average joint temperature);

[0078] a, b: Weighting coefficients (a = 0.6, b = 0.4, obtained by fitting clinical data from 1000 patients with rheumatoid arthritis, with a fitting error ≤ 5%);

[0079] Data Management System: Oracle Database (conventional existing technology) manages patient data;

[0080] Communication module: Transmits data at a rate of 100MB / s via a 10 Gigabit Ethernet interface (conventional existing technology).

[0081] The terminal console 501 of the doctor-patient communication interaction terminal 500 is made of high-strength engineering plastic, making it sturdy and durable. A 10.1-inch high-definition touchscreen display 502 with a resolution of 1920×1200 is fixedly installed on top of the console 501, clearly displaying doctor's replies and treatment suggestions. Below the display 502, an input module 503 integrates touchscreen, keyboard, and voice input functions, facilitating patient information input. Inside the console 501, a communication module 504 employing 4G / 5G and Wi-Fi dual-mode enables high-speed communication with the data analysis cloud platform 400; a high-performance quad-core processor 505 handles patient input and received data; and a rechargeable power supply 506 provides stable power to the terminal. Patients input information through the input module 503, which is then processed by the processor 505 and transmitted to the data analysis cloud platform 400 by the communication module 504. Simultaneously, information received from the platform is displayed on the display 502, enabling efficient doctor-patient communication.

[0082] Implementation steps of this embodiment

[0083] Before use, medical personnel need to conduct a comprehensive inspection of the multi-dimensional pain sensor 100. Using professional sensor performance testing equipment, key indicators such as the sensitivity and accuracy of the pressure sensing chip 102, temperature-sensitive resistor 103, and vibration acceleration sensor 104 are calibrated and tested to ensure they meet factory standards. Simultaneously, using a high-precision power detection instrument, the battery 107 is confirmed to have sufficient power; it needs to be recharged when the power level drops below 80% to ensure the sensor can operate stably for at least 24 hours. Furthermore, the sensor mounting housing 101 is carefully inspected for cracks, damage, or other defects to ensure its sealing and protection are in good condition and to prevent external factors from interfering with the normal operation of the sensor.

[0084] Next, medical staff first used medical alcohol swabs to wipe and disinfect the skin of the joint area where the sensor would be installed. After the skin dried naturally, they used a suitable 3M Tegaderm transparent dressing to attach the multi-dimensional pain sensor 100 to the patient's joint area. The installation position was determined according to the following criteria: Locate the 'force center area':

[0085] First, locate the knee joint: the midpoint of the line connecting the lower pole and the upper pole of the patella;

[0086] Next, locate the wrist joint: the midpoint of the line connecting the radial styloid process and the ulnar styloid process;

[0087] Verification of whether it is an 'effective vibration monitoring side': The patient actively flexes and extends the joint 3 times. When the peak-to-peak value of the vibration acceleration sensor 104 output is >2.0g, it is determined to be an effective position.

[0088] Adaptation when the patient is a special patient: for patients with joint deformities, ultrasound imaging is used to determine the point of maximum pressure in the joint cavity; for obese patients, pressure-sensitive membranes are used to locate areas of significant pressure change.

[0089] Relationship between installation location and wearable device location: Spacing ≤ 5.0cm (based on Bluetooth 5.0 effective transmission distance requirements);

[0090] Relative orientation: The long axis of the pain sensor is parallel to the joint's flexion and extension direction. This step ensures accurate attachment. This step is crucial for obtaining high-quality pain-related data; precise installation ensures the sensor comprehensively and accurately captures various joint status information.

[0091] After the multi-dimensional pain sensor 100 is powered on, the pressure sensing chip 102, leveraging its piezoelectric material properties, converts changes in joint surface pressure into electrical signals with a high sensitivity of 0.1 N / mV. The temperature-sensitive resistor 103, based on the change in its resistance with temperature, accurately measures the local joint temperature with an accuracy of ±0.1℃ and converts the temperature information into an electrical signal. The vibration acceleration sensor 104 monitors joint vibration in real time, with a measurement range of ±10g, effectively detecting vibrations caused by joint movement and pain, and generating corresponding electrical signals. The main control circuit 105 rapidly acquires these electrical signals at a frequency of 100 times per second, processes the data using an advanced Kalman filter algorithm to remove noise interference, and extract effective data features.

[0092] The processed data is transmitted via the wireless transmission module 106, employing a two-level transmission architecture, prioritizing relay through the wearable device 200: when the signal strength is >-70dBm, it is transmitted to the wearable device 200 via Bluetooth 5.0 at a rate of 1Mbps; the wearable device 200 performs data preprocessing; the preprocessed data is then uploaded to the cloud platform 400 via 4G / 5G; this path reduces transmission power consumption and extends sensor battery life.

[0093] When a fault occurs, the device will connect directly to the cloud platform. The trigger conditions are three consecutive Bluetooth handshake failures or the wearable device's battery level is less than 10%. ZigBee Pro 2017 direct connection will be enabled to transmit raw data at a rate of 250kbps. The device fault alarm will be sent to the medical-patient interaction terminal 500 simultaneously.

[0094] The advantages of the above architecture are: Bluetooth 5.0 transmission power consumption is much lower than ZigBee direct connection to the cloud, which can significantly extend the battery life of sensors; the built-in filtering algorithm of wearable devices removes motion artifacts in real time, greatly reducing the amount of invalid data uploaded; the packet loss rate of point-to-point Bluetooth transmission is extremely low, far lower than that of ZigBee direct connection.

[0095] In this way, multi-dimensional pain data can be transmitted to subsequent processing stages in a timely and accurate manner, providing doctors with rich data support for a comprehensive understanding of the patient's pain status.

[0096] Before a patient wears the wearable continuous monitoring device 200, medical staff must conduct a thorough inspection. First, check the wearable sleeve 201 for damage or deformation, ensuring it is skin-friendly, breathable, and elastic, and will not cause discomfort to the patient's skin or affect joint movement. Next, use professional sensor calibration equipment to comprehensively calibrate the accelerometer 202, gyroscope 203, heart rate sensor 204, and sleep monitoring sensor 205. Through calibration, ensure the accelerometer 202 achieves a measurement accuracy of ±0.05 m / s². 2 It can accurately measure the acceleration changes of joint movement; the gyroscope 203 has an angle measurement accuracy of ±0.5°, and works in conjunction with the accelerometer 202 to accurately measure joint movement posture; the heart rate sensor 204 has a measurement error controlled within ±1 beat / minute, accurately measuring heart rate. Simultaneously, using the device's built-in power detection function, it confirms that the rechargeable battery has a charge of over 70%, ensuring that the device can operate continuously and stably for at least 12 hours.

[0097] Next, under the guidance of medical staff, the patient correctly wears the wearable sleeve 201 of the wearable continuous monitoring device 200 on the corresponding joint, such as the wrist, ankle, or knee. When wearing it, the placement of the accelerometer 202 must meet the following requirements: it should be positioned in the most frequently used area, namely the knee joint in the region 3-5 cm proximal to the upper edge of the patella, and the wrist joint in the region along the radial styloid process where the extensor carpi radialis tendon runs; it should be positioned in an area where acceleration can be accurately sensed, i.e., by having the patient perform standardized joint movements (such as walking / holding a cup), and then the main control unit 206 detects the area with the maximum amplitude of the accelerometer 202 signal to determine the area where acceleration can be accurately sensed. For special patients, such as those with joint deformities, priority should be given to ensuring that the accelerometer covers the area of ​​maximum range of motion; for patients with skin injuries, the wound should be avoided and the integrity of the sensor-skin contact should be ensured.

[0098] The gyroscope 203 and accelerometer 202 are placed close together to achieve accurate measurement of joint movement posture. The heart rate sensor 204 must fit snugly against the skin, avoiding obstruction by clothing or other items, to ensure accurate heart rate measurement. The sleep monitoring sensor 205 should be placed in a position where the wearable sleeve 201 has a large contact area with the body, effectively monitoring pressure distribution and changes in body surface temperature during sleep. Correct wearing of the devices is crucial to ensuring that each sensor can accurately collect various physiological parameters, providing a comprehensive and reliable data foundation for subsequent comprehensive analysis of the patient's condition.

[0099] Then, after the wearable continuous monitoring device 200 is activated, the accelerometer 202 collects acceleration data of joint movement at a frequency of 50 times per second, and the gyroscope 203 simultaneously collects angular velocity data of joint movement. The combination of the two can accurately reconstruct the joint's motion trajectory and posture. The heart rate sensor 204 measures heart rate at a frequency of 1 time per second by emitting light of a specific wavelength and detecting changes in reflected light. The pressure sensor and temperature sensor in the sleep monitoring sensor 205 collect pressure distribution and body surface temperature change data during sleep at a frequency of 1 time per minute, respectively. The main control unit 206 uses the Fast Fourier Transform (FFT) algorithm to preprocess the collected data, extract data features, and store the processed data in an internal memory with a capacity of 1GB. When the device is within the Bluetooth signal coverage range, the Bluetooth module 207 sends the stored data to the data analysis cloud platform 400 every 5 minutes at a transmission rate of 2Mbps; if the Bluetooth signal is unstable or cannot connect, the data is temporarily stored in the internal memory and resent after the signal is restored. In this way, the wearable continuous monitoring device 200 can continuously and stably provide comprehensive and continuous physiological parameter data for disease analysis, helping doctors to understand the patient's physical condition more accurately and providing strong support for developing personalized treatment plans.

[0100] The data analysis cloud platform 400 receives monitoring data via a 10 Gigabit Ethernet interface (a conventional existing technology). The specific process is as follows:

[0101] Data transmission: Transmitted at a rate of 100MB / s (using conventional existing technology) using the TCP / IP protocol;

[0102] Data caching: A temporary cache area stored on a data storage server (conventional existing technology);

[0103] Data Validation: Integrity Validation Formula:

[0104] Where: Checksum: Data integrity check value (dimensionless, range 0-255, used to verify the integrity of transmitted data);

[0105] Datai The i-th data packet (unit: bytes, referring to the i-th segment of data split during transmission, including monitoring indicator values ​​and timestamps);

[0106] n: Total number of data packets (dimensionless, referring to the total number of data packets in a single transmission);

[0107] If Checksum = 0, it is determined that the data transmission was lost or erroneous, and the system will automatically trigger the retransmission mechanism (retransmission count ≤ 3 times, interval 10 seconds).

[0108] The data processing software performs the following analysis process:

[0109] Data cleaning: Outliers are removed using the DBSCAN algorithm (a conventional existing technique);

[0110] Data standardization: Z-score normalization (conventional existing techniques);

[0111] Feature extraction: PCA dimensionality reduction (a conventional existing technique);

[0112] In-depth analysis:

[0113] Pain-activity association model:

[0114] Where: P score Pain association score (dimensionless, range 0-10, used to quantify the degree of association between joint pain and activity status; the higher the value, the more significant the effect of activity on pain);

[0115] P max Peak joint surface pressure (unit: N, refers to the maximum joint surface pressure value monitored by multi-dimensional pain sensors within 24 hours);

[0116] : Rate of change of joint range of motion (unit: ° / s, where A represents the range of motion, measured by the gyroscope and accelerometer of the wearable device, and t represents time, which reflects the speed of joint movement);

[0117] α, β: weighting coefficients (α = 0.6, β = 0.4, obtained through training with clinical monitoring data of 300 patients with rheumatoid arthritis; during training, it was verified that these coefficients can keep the model prediction error ≤ 1.5 points);

[0118] Inflammation trend prediction: Prediction is performed using the same formula as the inflammation trend prediction formula mentioned above;

[0119] Pattern recognition: Analyzing temporal changes in physiological parameters using CNN / RNN (conventional existing technology).

[0120] Treatment suggestion generation process:

[0121] Input integration (conventional prior art) The input integration steps are as follows:

[0122] Data source and format:

[0123] Medical record data: from the hospital's HIS system, in structured tables, including CRP (mg / L), erythrocyte sedimentation rate (mm / h), age, gender, etc.

[0124] Genetic data: from sequencing reports from the hospital's laboratory department, in text format, such as "HLA-DRB1*04 positive";

[0125] Historical treatment records: archived from the doctor-patient interaction terminal, in semi-structured text format, such as "2025-06-01: Methotrexate 10mg / week".

[0126] Integration Operations:

[0127] Medical record data: Export an Excel spreadsheet from the HIS system, keeping the four columns "CRP, ESR, Age, and Gender". Missing values ​​are filled with the mean of the same gender / age group.

[0128] Genetic data: In Excel, "1" represents a positive gene (e.g., "HLA-DRB1*04"), and "0" represents a negative gene;

[0129] Treatment records: Extract "drug name, dosage, and start date" from the terminal archive text and copy it to a new column in Excel;

[0130] Integration: Using "Patient ID" as the keyword, the above three types of Excel spreadsheets are merged into a single integrated table, which includes fields such as Patient ID, medical record data, genetic data, and treatment records.

[0131] Decision Model: Drug Dosage Adjustment Formula: D adj =D curr ×(1+k·ΔVAS)

[0132] Where: D adj Adjusted drug dosage (unit: mg, personalized dosage calculated based on the patient's real-time pain changes);

[0133] D curr Current medication dosage (unit: mg, the patient's current standard dosage);

[0134] k: Dose adjustment factor (dimensionless, default value 0.1, determined based on clinical validation, ensuring that a single adjustment does not exceed 20% of the current dose);

[0135] ΔVAS: Change in VAS pain score (unit: points, calculated as "current VAS score - baseline VAS score", where the baseline is the score at the time of the first monitoring, ranging from -10 to +10, with positive values ​​indicating increased pain and negative values ​​indicating relief).

[0136] This involves obtaining the current dosage (in mg); calculating the change in the current pain score from baseline (using a 0-10 visual analog scale); and then calculating the new dose as: current dose × (1 + 0.1 × change in pain score). Example: If the current dose is 20 mg and the pain score increases by 2 points, then the new dose is 20 × (1 + 0.1 × 2) = 24 mg.

[0137] Physical therapy trigger conditions: the range of motion (ROM) is less than 50% of the normal value, i.e., the maximum range of motion of the joint is measured by the gyroscope 203 of the wearable device 200; the inflammation index (T_{inf}) is greater than 5, i.e., the value is calculated by the formula: inflammation index = 0.6 × C-reactive protein (CRP) value + 0.4 × 24-hour average joint temperature.

[0138] Output scheme determination table:

[0139] Table 2. Treatment Plan Determination Output Table

[0140] Pain score Inflammation index Treatment plan ≥7 >10 Increase TNF inhibitor by 0.1 mg / kg + hyperthermia 3 times / week 4-6 5-10 Maintenance dose + joint flexion and extension exercises 10 times / day

[0141] Data storage: Structured storage to a database (conventional existing technology).

[0142] Next, medication delivery will begin. Before using the personalized medication delivery device 300, medical staff must conduct a comprehensive inspection. The main support 301 is checked for stability, deformation, cracks, or other damage to ensure that structural issues will not affect medication delivery during use. The sealing performance of the micro-drug reservoir 302 is checked, and a compression test is conducted to ensure there is no risk of drug leakage. The remaining drug level is observed through the transparent reservoir casing to ensure sufficient medication to meet the patient's current treatment needs. The stepper motor 303 is tested using high-precision motor testing equipment to ensure its step angle accuracy reaches ±0.05°, enabling precise control of the lead screw 304's rotation angle. The lead screw 304's pitch accuracy is measured using a lead screw measuring instrument, ensuring it is within ±0.01mm to achieve precise control of the piston 306's movement distance. The sealing performance between the piston 306 and the syringe 305 is checked, and a simulated injection test is conducted to ensure no leakage during the medication injection process. In addition, professional chip testing tools are needed to perform functional tests on the intelligent control chip 307 to ensure that it can accurately receive and execute instructions from the data analysis cloud platform 400. Simultaneously, it is confirmed that the power module's charge level is above 60% to support continuous operation of the device for at least 8 hours, ensuring that the drug delivery process is not affected by power interruptions.

[0143] Next, medication filling is completed. Under aseptic conditions, medical staff, using a sterile syringe, slowly inject the required medication into the micro-drug reservoir 302 using a sterile syringe on a sterile operating table, strictly following aseptic procedures. During filling, the concentration and dosage of the medication are precisely controlled according to the drug preparation guidelines and the patient's treatment plan. For example, for a specific drug, it is prepared at a ratio of 10 mg of active ingredient per milliliter. Based on factors such as the patient's weight and the severity of their condition, the required total volume of medication is calculated, and 30 milliliters of medication are precisely injected into the micro-drug reservoir 302. After filling, the sealing performance of the micro-drug reservoir 302 is checked again to ensure the safety and stability of the medication during storage and delivery. The precise operation of this step directly affects the efficacy and safety of drug therapy and must be strictly controlled.

[0144] Meanwhile, the intelligent control chip 307 receives instructions from the data analysis cloud platform 400 in real time. Based on information such as drug injection volume and injection frequency in the instructions, it precisely controls the operation of the stepper motor 303. The stepper motor 303 drives the lead screw 304 to rotate via a coupling. The lead screw 304 converts the rotational motion into linear motion, driving the piston 306 to move within the injection syringe 305 with extremely high precision, achieving precise drug injection with an injection volume control accuracy of ±0.01mL. During drug injection, the drug concentration sensor 308 monitors the drug concentration in real time based on spectral analysis technology, with a concentration measurement accuracy of ±0.5%; the flow sensor monitors the drug flow rate using the principle of electromagnetic induction, with a flow measurement accuracy of ±0.1mL / min. These two sensors feed back the real-time monitored drug concentration and flow rate data to the intelligent control chip 307, which adjusts the drug injection process in real time according to the preset safety range and treatment plan requirements. If the drug concentration or flow rate becomes abnormal, the intelligent control chip 307 will promptly issue an alarm and suspend the drug injection. Simultaneously, it will feed the abnormal information back to the data analysis cloud platform 400 so that medical staff can handle the situation promptly. This closed-loop control mechanism ensures the accuracy, safety, and stability of drug delivery, providing patients with precise and reliable drug treatment.

[0145] Before using the doctor-patient communication interaction terminal 500, medical staff or patients need to inspect it. Check the terminal control panel 501 for damage or deformation, ensuring its structural stability. Check the display screen 502 for dead pixels or display abnormalities, ensuring the 10.1-inch, 1920×1200 resolution high-definition touchscreen can clearly and accurately display information by displaying test patterns. Test the touchscreen, keyboard, and voice input functions of the input module 503 to ensure they are normal, by inputting characters and voice commands, ensuring patients can easily and quickly input information. Check the 4G / 5G and Wi-Fi connectivity of the communication module 504, ensuring high-speed and stable communication with the data analysis cloud platform 400 by connecting to the network and accessing test websites. Use the device's built-in power detection function to confirm the power supply 506 has sufficient power; it needs to be charged when the power is below 20% to ensure continuous and stable operation of the terminal.

[0146] Patients view doctor's replies and treatment suggestions on display screen 502. The clear and intuitive interface design allows patients to easily understand the content and requirements of the treatment plan. Patients can provide feedback through input module 503 based on their symptom changes and feelings. If using touchscreen input, the screen provides a virtual keyboard and handwriting input functions for convenient text input; if using voice input, patients only need to speak the relevant content, and the system uses advanced speech recognition technology to convert speech into text information. At the medical institution level, doctors view patient monitoring data, medical records, and treatment suggestions generated by the artificial intelligence algorithm module through the doctor-patient communication interaction terminal 500. Based on professional knowledge and clinical experience, doctors evaluate and adjust the treatment suggestions, inputting the adjusted treatment plan through input module 503, including increases or decreases in drug dosage and changes in treatment methods. This process achieves two-way information flow between doctors and patients, enabling doctors to understand the patient's actual situation in a timely manner, and allowing patients to accurately obtain professional guidance from doctors.

[0147] After the patient's input information is encoded and encrypted by the processor 505, the communication module 504 transmits the information to the data analysis cloud platform 400 at a high speed via a 4G / 5G or Wi-Fi network. Upon receiving the patient information, the data analysis cloud platform 400 optimizes and adjusts the treatment plan based on the latest monitoring data and analysis results. The doctor's adjusted treatment plan is also processed by the processor 505 and transmitted to the data analysis cloud platform 400 by the communication module 504. The data analysis cloud platform 400 sends the adjusted information to the intelligent control chip 307 of the personalized drug delivery device 300 to control the relevant parameters of drug delivery, and simultaneously feeds back the adjusted treatment plan to the patient's doctor-patient communication terminal 500. Through this information transmission and feedback mechanism, the treatment plan is dynamically adjusted and optimized, ensuring that the treatment process can be adjusted in a timely manner according to the patient's actual situation, thereby improving treatment effectiveness and patient satisfaction.

[0148] Beneficial effects of this embodiment

[0149] This embodiment establishes a rheumatoid arthritis management system that simultaneously integrates multi-dimensional symptom monitoring, personalized drug delivery, and real-time doctor-patient interaction. Through the collaboration of multi-dimensional pain sensors and wearable devices, it comprehensively captures joint pain and physiological changes, enabling an objective reflection of the condition and avoiding subjective bias. A data analysis cloud platform integrates multi-source data to generate personalized treatment plans, which, combined with a precise drug delivery device, facilitates on-demand medication, reduces side effects, and enhances treatment targeting. A doctor-patient communication and interaction terminal promotes real-time information exchange, allowing for timely adjustments to the treatment plan and improving patient compliance. The synergy of these components provides patients with convenient and efficient full-cycle management, ensuring treatment safety and improving treatment outcomes and patient experience.

[0150] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A personalized injection system for multidimensional symptom monitoring of rheumatoid arthritis, characterized in that: The system includes a wearable continuous monitoring device fixedly mounted on the outside of a patient's joint. A multi-dimensional pain sensor, also fixedly attached to the outside of the joint, is connected via a wire to the wearable continuous monitoring device. A personalized drug delivery device and a doctor-patient communication terminal are wirelessly connected to the wearable continuous monitoring device via a data analysis cloud platform. The multi-dimensional pain sensor includes a sensor housing. A pressure-sensing chip for sensing changes in joint surface pressure is located at the center of the bottom of the sensor. Temperature-sensitive resistors for measuring local joint temperature are evenly distributed around the pressure-sensing chip. A vibration acceleration sensor is fixedly mounted on the outside of the sensor housing. A main control circuit is fixedly mounted in the center of the inside of the sensor housing. A wireless transmission module is fixedly mounted on one side of the main control circuit. A battery is fixedly mounted at the bottom of the sensor housing and connected to the main control circuit via a wire. The personalized drug delivery device includes a main support frame. A micro-drug reservoir is fixedly mounted on the top of the main support frame. A high-precision infusion pump is fixedly mounted at the bottom of the micro-drug reservoir. An intelligent control chip is fixedly mounted on the side of the high-precision infusion pump. The intelligent control chip uses a proprietary... A microcontroller is mounted on a circuit board next to the high-precision injection pump to control the drug injection volume. A power module is fixedly installed at the bottom of the main support, and the power module is connected to each electrical component via wires. The high-precision injection pump includes a stepper motor, a lead screw, a piston, and an injection syringe. The stepper motor is connected to the lead screw via a coupling. The piston is located inside the injection syringe, and the lead screw drives the piston to move within the injection syringe to achieve precise drug injection. A drug concentration sensor is fixedly installed near the outlet of the injection syringe. A device is installed on the pipe between the injection syringe and the injection needle. The wearable continuous monitoring device includes: a wearable body, an accelerometer fixedly mounted near the joint area of ​​the wearable body, a gyroscope fixedly mounted on the side of the accelerometer, a heart rate sensor fixedly mounted on the inner side of the wearable body in contact with the skin, a sleep monitoring sensor fixedly mounted on the side of the heart rate sensor, a main control unit fixedly mounted in the middle of the wearable body, a Bluetooth module fixedly mounted on the side of the main control unit, and a rechargeable battery fixedly mounted at the bottom of the wearable body, the rechargeable battery being connected to various power-consuming components via wires.

2. The personalized injection system for multidimensional symptom monitoring of rheumatoid arthritis according to claim 1, characterized in that: The data analysis cloud platform consists of a server cluster, data processing software, artificial intelligence algorithm modules, a database management system, and a network interface for communicating with external devices and connecting the server cluster to the Internet.

3. The personalized injection system for multidimensional symptom monitoring of rheumatoid arthritis according to claim 2, characterized in that... The doctor-patient communication and interaction terminal includes a terminal control panel. A display screen for displaying the doctor's replies and treatment suggestions is fixedly installed on the top of the terminal control panel. An input module is fixedly installed below the display screen. A communication module, a processor, and a power supply are fixedly installed inside the terminal control panel.

4. The method of using the personalized injection system for multidimensional symptom monitoring of rheumatoid arthritis according to any one of claims 1 to 3, characterized in that, Includes the following steps: S1 attaches a multi-dimensional pain sensor to the patient's painful joint. It collects joint pain-related data through its internal pressure sensing chip, temperature-sensitive resistor, and vibration acceleration sensor, and uses a wireless transmission module to send the data to a wearable continuous monitoring device or directly to a data analysis cloud platform. S2, the patient wears a wearable continuous monitoring device that collects various physiological parameters through an accelerometer, gyroscope, heart rate sensor and sleep monitoring sensor, and transmits the data to a data analysis cloud platform via a Bluetooth module; S3, the data analysis cloud platform receives data from multi-dimensional pain sensors and wearable continuous monitoring devices, and uses data processing software and artificial intelligence algorithm modules to analyze and process the data to generate personalized treatment suggestions; S4. Doctors view information sent by the data analysis cloud platform through the doctor-patient communication interaction terminal, and input the adjusted treatment plan through the input module. The plan is then processed by the processor and transmitted to the data analysis cloud platform through the communication module. The data analysis cloud platform sends the adjustment information to the intelligent control chip of the personalized drug delivery device to control the drug delivery. At the same time, the drug concentration sensor and flow sensor of the personalized drug delivery device feed back the drug status data to the data analysis cloud platform.

Images