An intelligent pain detection and management system

Abstract

This invention relates to the field of pain management technology, specifically to an intelligent pain detection and management system. It includes: a data acquisition module for acquiring facial expression images and physiological parameters of a patient; a preprocessing module for preprocessing the acquired facial expression images and physiological parameters; a recognition module for identifying the patient's pain state based on the facial expression images using a pain state recognition model and outputting a pain coefficient; a processing module for determining personalized drug delivery speed and dosage based on the pain coefficient and the patient's personal information; the processing module also analyzes and evaluates the patient's pain trend, generates a drug delivery control correction value based on the pain trend, and adjusts the drug delivery speed and dosage; the processing module further adjusts the pain trend based on real-time data, evaluates the correction effect, and dynamically adjusts the correction value; and a drug delivery module for executing the drug delivery control plan. This technical solution can improve the timeliness of pain detection and the effectiveness of pain management.

Description

Technical Field

[0001] This invention relates to the field of pain management technology, specifically to an intelligent pain detection and management system. Background Technology

[0002] In the medical field, pain management has always been a key focus and challenge in clinical treatment. Traditional pain detection methods mainly rely on patients' subjective descriptions and the experience and judgment of medical staff, which suffers from high subjectivity and insufficient accuracy. With the continuous advancement of medical technology, especially the rapid development of artificial intelligence and deep learning technologies, new ideas and methods have been provided for pain detection and management.

[0003] However, most existing pain detection systems rely on a single pain assessment method, such as questionnaires or pain rating scales. These methods often fail to comprehensively and accurately reflect a patient's pain status. Furthermore, traditional pain management protocols often employ standardized medication control, lacking personalized adjustments to address individual patient differences, leading to poor treatment outcomes and potentially causing drug side effects. Summary of the Invention

[0004] The purpose of this invention is to propose an intelligent pain detection and management system, which can improve the timeliness of pain detection and the effectiveness of pain management.

[0005] To achieve the above objectives, embodiments of this disclosure provide an intelligent pain detection and management system, comprising:

[0006] The acquisition module is used to acquire images of the patient's facial expressions and physiological parameters;

[0007] The preprocessing module is used to perform preprocessing operations on the acquired facial expression images and physiological parameters;

[0008] The recognition module uses a pain state recognition model built with a neural network to identify the patient's pain state based on the preprocessed facial expression image and outputs the pain coefficient.

[0009] The processing module, based on the pain coefficient output by the recognition module and combined with the patient's personal information, formulates a personalized drug administration control plan, including the drug administration rate and dosage.

[0010] The processing module is also used to analyze and evaluate the patient's pain trend, generate a drug administration control correction value based on the pain trend, and adjust the drug administration rate and dosage; the processing module will also adjust the pain trend based on real-time data, evaluate the correction effect, and dynamically adjust the correction value;

[0011] The drug delivery module is used to achieve precise drug delivery according to the drug delivery control plan formulated by the processing module.

[0012] Beneficial effects of the basic solution: This solution assesses the patient's pain level from multiple dimensions by collecting facial expression images and physiological parameters. The pain state recognition model built using a neural network can efficiently and accurately identify the patient's pain state and output a pain coefficient, thus improving the accuracy of pain detection.

[0013] Based on pain coefficients and patient information, personalized medication control plans are developed to improve treatment effectiveness, reduce unnecessary drug use, and lower the risk of drug side effects. The processing module analyzes and assesses the patient's pain trends, generates medication control correction values, and dynamically adjusts the dosing rate and dosage to better adapt to changes in the patient's pain state, ensuring accurate and timely drug administration. This real-time feedback adjustment mechanism can also intervene to a certain extent based on the patient's physiological parameters to ensure medication safety.

[0014] This system achieves intelligent and automated control from pain detection to drug administration, greatly reducing the workload of medical staff. By developing personalized plans, it enables precise drug delivery, reducing errors caused by human intervention.

[0015] As a feasible preferred solution, the acquisition module employs a camera and physiological parameter sensors; the camera is installed above or to the side of the patient's bed, has the function of adjusting focus and viewing angle, and is surrounded by lighting equipment to eliminate the influence of changes in ambient light; the physiological parameter sensors include a heart rate sensor, a blood pressure sensor, and a blood oxygen saturation sensor, which are connected to the acquisition module via wired or wireless means.

[0016] As a feasible preferred embodiment, the preprocessing module performs preprocessing operations on the acquired facial expression images, including cropping, color conversion, histogram equalization, and noise removal; and preprocessing operations on the physiological parameter data, including cleaning, formatting, and normalization.

[0017] As a feasible preferred solution, the recognition module fuses the image features and physiological parameter data output by the neural network model, calculates the pain coefficient, and performs dimensionality reduction processing on the fused feature vector.

[0018] As a feasible and preferred approach, a pre-trained pain state recognition model is used to process the collected patient pain expression images to extract image feature vectors; the collected physiological parameter data is standardized; and the image feature vectors and standardized physiological parameter data are weighted and averaged to obtain the pain coefficient.

[0019] As a feasible preferred solution, the processing module uses a pain coefficient to reflect the patient's pain level, extracts information reflecting the patient's physiological state from physiological parameter data, and extracts factors related to drug metabolism and distribution from personal information; it uses a feature selection method to screen out features that have a significant impact on drug administration control, retaining features closely related to drug administration rate and dosage; based on the extracted features, it constructs a machine learning training module, trains the model using historical data; calculates the initial drug administration rate and dosage, and adjusts the drug administration rate and dosage accordingly.

[0020] As a feasible and preferred option, the drug delivery module also has start-up, stop, and emergency braking functions.

[0021] As a feasible preferred solution, it also includes an alarm module for detecting abnormal states of the patient during drug administration and issuing an alarm signal when an abnormality occurs. The alarm module includes an audible and visual alarm and a communication submodule for issuing audible and visual alarm signals and sending alarm information when an abnormal state is detected.

[0022] As a feasible preferred solution, it also includes a remote monitoring module, which enables medical staff to remotely monitor and guide the patient's pain status through Internet of Things (IoT) technology. The remote monitoring module transmits the patient's pain coefficient, physiological parameter data, and medication records to a cloud server in real time, allowing medical staff to view the patient's real-time data on other terminals and send control signals or conduct real-time communication.

[0023] As a preferred feasible option, the processing module is also used to formulate a dosing strategy, which includes dosing time intervals and drug combination strategies. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of an intelligent pain detection and management system architecture.

[0025] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0026] To make the technical solution and advantages of this application clearer, the technical solution of the present invention will be further described in detail below with reference to the accompanying drawings. It is understood that the specific embodiments described herein are only some embodiments of the present invention, and are only used to explain this application, not to limit it. It should be noted that the technical features or combinations of technical features described in the following embodiments should not be considered isolated; they can be combined with each other to achieve better technical effects. The same reference numerals appearing in the accompanying drawings of the following embodiments represent the same features or components, and can be applied to different embodiments.

[0027] Furthermore, unless otherwise defined, the technical or scientific terms used in this invention description shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains.

[0028] Reference numerals: Electronic device 500, processor 501, communication interface 502, memory 503, bus 504.

[0029] The present invention will now be described in further detail with reference to the accompanying drawings:

[0030] Reference Figure 1 An intelligent pain detection and management system includes a data acquisition module, a preprocessing module, a recognition module, a processing module, a drug delivery module, an alarm module, and a remote monitoring module.

[0031] The acquisition module is used to capture images of the patient's facial expressions and physiological parameters. The module employs a high-precision camera and physiological parameter sensors to ensure data accuracy and real-time performance.

[0032] A high-precision camera is mounted above or to the side of the patient's bed to ensure a clear view of the patient's face. The camera should also be adjustable in focus and angle to accommodate different patients and shooting environments. Lighting, such as LED lights, should surround the camera to eliminate the impact of ambient light variations on image acquisition and ensure image quality.

[0033] Physiological parameter sensors include heart rate sensors, blood pressure sensors, and blood oxygen saturation sensors. These sensors connect to the data acquisition module via wired or wireless connections. Wired connections use standard medical-grade cables to ensure stable and secure data transmission; wireless connections utilize Bluetooth Low Energy or Wi-Fi technology for wireless data transmission. The data acquisition frequency of the physiological parameter sensors should be set according to the patient's specific condition; for example, heart rate and blood pressure can be collected once per minute, and blood oxygen saturation can be collected once every 5 minutes to ensure data accuracy and real-time performance.

[0034] The preprocessing module is used to preprocess the acquired facial expression images and physiological parameter data, providing a high-quality data foundation for subsequent analysis.

[0035] Specifically, the acquired facial expression images are cropped to remove unnecessary background information, retaining only the patient's facial area. This cropping operation is automatically performed using image recognition algorithms. Color conversion is then performed, transforming the image from the RGB color space to the grayscale color space to reduce data volume and computational complexity. Histogram equalization and other processing methods are applied to enhance image contrast and clarity. Noise removal is then performed using algorithms such as Gaussian filtering and median filtering to remove noise and interference from the image, improving image quality.

[0036] The collected physiological parameter data is cleaned to remove outliers and duplicates. Outliers are detected and removed by setting thresholds or using statistical methods; duplicates are processed using data deduplication algorithms. The cleaned physiological parameter data is then formatted into a uniform standard format to facilitate subsequent analysis and processing. Simultaneously, the data is normalized to eliminate differences in data units and dimensions between different sensors.

[0037] The recognition module utilizes a pain state recognition model built with a U-Net neural network embedded with VGGNet-A to identify the patient's pain state based on preprocessed facial expression images, outputs a pain coefficient, and combines physiological parameter data to assist in the judgment of pain state, thereby improving the accuracy of the assessment.

[0038] VGGNet-A is a deep convolutional neural network with powerful image feature extraction capabilities. In this embodiment, VGGNet-A is used as a feature extractor to extract key features from facial expression images. U-Net is a convolutional neural network based on an encoder-decoder structure, suitable for image segmentation tasks. In this embodiment, U-Net is used as a classifier to classify and recognize the extracted image features. A large number of facial expression images and corresponding pain state labels are used as training data and input into the neural network model for training. By adjusting the model parameters and optimizing the algorithm, the model can accurately identify the patient's pain state.

[0039] The pain coefficient is calculated by fusing image features and physiological parameter data output from the pain state recognition model. Specifically, the pre-trained pain state recognition model processes the collected images of the patient's pain expression to extract image feature vectors. The collected physiological parameter data is standardized to eliminate the influence of different units and value ranges on subsequent calculations. The image feature vectors and the standardized physiological parameter data are then fused by weighted averaging according to their weights to obtain the fused feature vector, i.e., the pain coefficient, which represents the patient's current level of pain.

[0040] The fused feature vectors are then subjected to dimensionality reduction. In this embodiment, Principal Component Analysis (PCA) is used for dimensionality reduction. The covariance matrix of the feature vector Z is calculated, and eigenvalue decomposition is performed on the covariance matrix to obtain eigenvalues ​​and their corresponding eigenvectors. Then, based on the set dimensionality reduction dimension k, the eigenvectors corresponding to the k largest eigenvalues ​​are selected to form the dimensionality-reduced feature space. Finally, the feature vector Z is projected onto this feature space to obtain the dimensionality-reduced feature vector Z'. The dimensionality-reduced feature vector Z' is used as input data for subsequent pain coefficient calculation or other analyses. Because Z' has a lower dimension, the computational complexity is effectively reduced while retaining the main information in the original data.

[0041] The processing module, based on the pain coefficient output by the recognition module and combined with the patient's personal information (such as age, weight, gender, pain threshold, etc.), formulates a personalized drug administration control plan, including the administration rate and dosage. Specifically, based on the pain coefficient and the patient's personal information (such as age, weight, gender, pain threshold, etc.), an intelligent algorithm calculates the administration rate and dosage. The algorithm should consider factors such as the drug's half-life and metabolic rate to ensure the drug's effectiveness and safety.

[0042] The processing module obtains the patient's real-time pain coefficient from the recognition module. It acquires physiological parameter data such as heart rate, blood pressure, and respiratory rate from physiological parameter monitoring devices. It retrieves the patient's personal information from the patient's file, including age, weight, gender, and pain threshold. It consults drug instructions or databases to obtain characteristic data such as the drug's half-life and metabolic rate, and preprocesses the data.

[0043] The pain coefficient is used to reflect the patient's pain level; information reflecting the patient's physiological state, such as the degree of heart rate abnormality and blood pressure fluctuation range, is extracted from physiological parameter data; factors related to drug metabolism and distribution, such as age, weight, and gender, are extracted from personal information.

[0044] Feature selection methods were used to identify features that significantly impacted dosing control, retaining those closely related to dosing rate and dosage while removing redundant or irrelevant features. Based on the extracted features, a machine learning training module was constructed, and the model was trained using historical data (including patients' pain coefficients, physiological parameter data, personal information, and dosing records).

[0045] The patient's pain level, physiological parameters, personal information, and drug characteristics are input into a trained model. Based on the input data, the model calculates the initial dosing rate and dosage. Factors such as the drug's half-life and metabolic rate are considered to appropriately adjust the dosing rate and dosage to ensure the drug's effectiveness and safety.

[0046] The drug delivery module, based on the drug delivery control plan developed by the processing module, uses precise mechanical transmission devices and speed sensors to achieve accurate drug delivery. The module has start-up, stop, and emergency braking functions to ensure the safety and controllability of the drug delivery process.

[0047] The alarm module is used to detect abnormal states of the patient during drug administration, such as abnormal facial expressions or abnormal physiological parameters, and to issue an alarm signal when an abnormality occurs. Specifically, the image recognition algorithm monitors changes in the patient's facial expressions in real time, and will trigger an alarm when it detects abnormal expressions (such as pain or fear) or abnormal values ​​(such as excessively fast heart rate or excessively low blood pressure).

[0048] The alarm module also includes an audible and visual alarm, which will emit an audible and visual alarm signal when an abnormal state is detected, reminding medical staff to take timely measures; it also includes a communication submodule, which is used to send alarm information to medical staff or family members so that timely response measures can be taken.

[0049] The remote monitoring module utilizes IoT technology to enable healthcare professionals to remotely monitor and guide patients' pain levels. Specifically, it transmits information such as the patient's pain level, physiological parameters, and medication records to a cloud server in real time. Healthcare professionals can view this real-time data, including pain levels, physiological parameters, and medication records, on computers, mobile phones, or other devices. The data is preferably displayed in charts or graphs for a more intuitive understanding of the patient's pain status. The remote monitoring module can also send control signals to the system to manually adjust the medication administration rate or communicate with the patient in real time to answer questions or provide assistance.

[0050] Example 2

[0051] The technical feature that distinguishes this embodiment from the above embodiments is that the processing module is also used to analyze pain trends and predict the patient's future pain status through historical data, providing a basis for early intervention.

[0052] Specifically, information such as the patient's past pain index, physiological parameters, and medication records is collected to establish a historical database. The historical data is analyzed and predicted to assess the patient's pain trend. Based on the pain trend, a medication control correction value is generated to adjust the administration rate and dosage.

[0053] For example, if the patient's pain is predicted to worsen, the administration rate can be increased to relieve pain earlier; if the patient's pain is predicted to lessen or stabilize, the administration rate can be decreased to reduce side effects. If the patient's pain is predicted to take longer to subside, the dosage per administration can be increased to prolong the duration of action. If the patient's pain is predicted to subside quickly, the dosage per administration can be reduced to avoid overdose.

[0054] The processing module will also adjust pain trends based on real-time data, assess the correction effect, and dynamically adjust the correction value to better suit individual patients and improve the correction effect.

[0055] Example 3

[0056] The technical feature that distinguishes this embodiment from the above embodiments is that the processing module is also used to formulate a drug administration strategy, which includes the drug administration time interval and drug combination strategy. The drug administration strategy is optimized by an improved optimization algorithm, specifically including the following contents.

[0057] The drug administration strategy is encoded, and a set of candidate solutions is randomly generated as the initial population.

[0058] Define the fitness function The formula is as follows:

[0059]

[0060] in, For the effectiveness of pain relief, This is a weighting coefficient for the effectiveness of pain relief. Due to the severity of drug side effects, This is a weighting factor for the severity of drug side effects. Due to physiological parameter instability, This is a weighting coefficient for the instability of physiological parameters. For drug costs, This is the weighting coefficient for drug costs.

[0061] Pain relief effectiveness The formula is as follows:

[0062]

[0063] in, It is the pain score at time t. It is the number of points used to evaluate the total time. It is the time point at which medication is started. It is a decay factor used to account for the decay effect of pain relief over time.

[0064] Severity of drug side effects The formula is as follows:

[0065]

[0066] in, It refers to the number of types of side effects. It is the first The severity rating of each side effect. It is the first The duration of these side effects.

[0067] physiological parameter instability The formula is as follows:

[0068]

[0069] in, It is the number of physiological parameters being assessed. It is the set of values ​​for the j-th physiological parameter during the evaluation period. and These are the maximum and minimum values ​​of the parameter, respectively. It is the average value of this parameter.

[0070]

[0071] in, It refers to the number of drugs used. It is k The unit price of the drug It is k Dosage of the drug.

[0072] In this embodiment, the weighting coefficients are set in a personalized manner based on the patient's specific situation and treatment goals. For example, the weighting coefficients for pain relief effectiveness, severity of drug side effects, and instability of physiological parameters are adjusted according to treatment expectations. If pain management is the primary focus, the weighting coefficient for pain relief effectiveness is increased. The weighting coefficient for drug cost is adjusted according to the patient's economic situation. For example, if the patient's economic situation is poor, the coefficient for drug cost is increased.

[0073] Individuals are randomly selected as parents. A random crossover point is chosen to split the two parent individuals into two parts. The gene segments after the crossover point are then swapped to generate two offspring individuals. One or more gene loci on a chromosome are randomly selected and flipped (from 0 to 1, or from 1 to 0) to increase the diversity of the intermediate high-population group. The resulting offspring are ranked according to their fitness. The number of offspring to be selected is determined. 50% of the offspring are selected from those ranked from highest to lowest fitness, and the remaining offspring are selected from those ranked from lowest to highest fitness. The proportion of offspring selected from highest to lowest fitness is gradually increased with the number of iterations, thus avoiding getting trapped in local optima in the early stages of iteration and accelerating convergence in the later stages.

[0074] Once the convergence condition is met, the iteration ends. The optimal chromosome after the iteration ends is decoded to obtain the optimal drug administration strategy. The optimal solution is output and used to guide the actual drug administration operation.

[0075] This disclosure also provides an intelligent pain detection and management method, which utilizes an intelligent pain detection and management system.

[0076] This disclosure also provides a storage medium storing a computer program, which, when executed by a processor, can implement all the steps of the above-described intelligent pain detection and management method.

[0077] Those skilled in the art will understand that implementing all or part of the processes in an intelligent pain detection and management method can be accomplished by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes of various embodiments of the intelligent pain detection and management method. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0078] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the aforementioned intelligent pain detection and management method. In this application embodiment, the processor is the control center of the computer system; it can be a physical machine processor or a virtual machine processor.

[0079] Reference Figure 2 The electronic device 500 includes at least one processor 501, at least one communication interface 502, at least one memory 503, and at least one bus 504. The bus 504 is used for communication between these components, the communication interface 502 is used for signaling or data communication with other node devices, and the memory 503 stores machine-readable instructions executable by the processor 501. When the electronic device 500 is running, the processor 501 communicates with the memory 503 via the bus 504, and when the machine-readable instructions are invoked by the processor 501, they execute the steps of the intelligent pain detection and management method described above.

[0080] The above content is merely an embodiment of the present invention. Commonly known structures and characteristics of the solutions are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can improve and implement this solution based on the guidance provided in this application and their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. An intelligent pain detection and management system, characterized in that: The acquisition module is used to acquire images of the patient's facial expressions and physiological parameters; The preprocessing module is used to perform preprocessing operations on the acquired facial expression images and physiological parameters; The recognition module uses a pain state recognition model built with a neural network to identify the patient's pain state based on the preprocessed facial expression image and outputs the pain coefficient. The processing module, based on the pain coefficient output by the recognition module and combined with the patient's personal information, formulates a personalized drug administration control plan, including the drug administration rate and dosage. The processing module is also used to analyze and evaluate the patient's pain trend, generate a drug administration control correction value based on the pain trend, and adjust the drug administration rate and dosage; the processing module will also adjust the pain trend based on real-time data, evaluate the correction effect, and dynamically adjust the correction value; The drug delivery module is used to achieve precise drug delivery according to the drug delivery control plan formulated by the processing module. The processing module is also used to formulate dosing strategies, including dosing intervals and drug combination strategies. The dosing strategies are optimized using improved optimization algorithms, specifically including the following: The drug administration strategy is encoded, and a set of candidate solutions is randomly generated as the initial population. Define the fitness function The formula is as follows: in, For the effectiveness of pain relief, This is a weighting coefficient for the effectiveness of pain relief. Due to the severity of drug side effects, This is a weighting factor for the severity of drug side effects. Due to physiological parameter instability, This is a weighting coefficient for the instability of physiological parameters. For drug costs, This is a weighting factor for drug costs; Pain relief effectiveness The formula is as follows: in, It is the pain score at time t. It is the number of points used to evaluate the total time. It is the time point at which medication is started. It is a decay factor used to account for the decay effect of pain relief over time; Severity of drug side effects The formula is as follows: in, It refers to the number of types of side effects. It is the first The severity rating of each side effect. It is the first The duration of the side effects; physiological parameter instability The formula is as follows: in, It is the number of physiological parameters being assessed. It is the set of values ​​for the j-th physiological parameter during the evaluation period. and These are the maximum and minimum values ​​of the parameter, respectively. It is the average value of this parameter; in, It refers to the number of drugs used. It is k The unit price of the drug It is k Dosage of the drug.

2. The intelligent pain detection and management system according to claim 1, characterized in that: The acquisition module uses a camera and physiological parameter sensors. The camera is installed above or to the side of the patient's bed and has the function of adjusting the focus and viewing angle. The camera is also equipped with lighting devices to eliminate the influence of changes in ambient light. The physiological parameter sensors include a heart rate sensor, a blood pressure sensor, and a blood oxygen saturation sensor, which are connected to the acquisition module via wired or wireless means.

3. The intelligent pain detection and management system according to claim 1, characterized in that: The preprocessing module performs preprocessing operations on the acquired facial expression images, including cropping, color conversion, histogram equalization, and noise removal. Preprocessing of physiological parameter data includes cleaning, formatting, and normalization.

4. The intelligent pain detection and management system according to claim 3, characterized in that: The recognition module fuses the image features and physiological parameter data output by the neural network model, calculates the pain coefficient, and performs dimensionality reduction processing on the fused feature vector.

5. The intelligent pain detection and management system according to claim 1, characterized in that: The pre-trained pain state recognition model was used to process the collected images of patients' pain expressions and extract image feature vectors; the collected physiological parameter data were standardized. The pain coefficient is obtained by weighted averaging and fusing the image feature vector and standardized physiological parameter data according to their respective weights.

6. The intelligent pain detection and management system according to claim 1, characterized in that: The processing module uses a pain coefficient to reflect the patient's pain level, extracts information reflecting the patient's physiological state from physiological parameter data, and extracts factors related to drug metabolism and distribution from personal information. Feature selection methods were used to screen out features that had a significant impact on dosing control, and features closely related to dosing rate and dosage were retained. Based on the extracted features, a machine learning training module was constructed, and the model was trained using historical data. The initial dosing rate and dosage are calculated, and then adjusted.

7. The intelligent pain detection and management system according to claim 1, characterized in that: The drug delivery module also has start, stop, and emergency braking functions.

8. The intelligent pain detection and management system according to claim 1, characterized in that: It also includes an alarm module for detecting abnormal states of patients during drug administration and issuing alarm signals when abnormalities occur. The alarm module includes an audible and visual alarm and a communication submodule for issuing audible and visual alarm signals and sending alarm information when an abnormal state is detected.

9. The intelligent pain detection and management system according to claim 1, characterized in that: It also includes a remote monitoring module, which uses Internet of Things (IoT) technology to enable medical staff to remotely monitor and guide patients' pain status. The remote monitoring module transmits the patient's pain coefficient, physiological parameter data, and medication records to the cloud server in real time. Medical staff can view the patient's real-time data on other terminals and send control signals or conduct real-time communication.

Images