Wireless infusion pump intelligent regulation and control system and medium

Through the intelligent control system of wireless infusion pump, the prediction model and particle swarm optimization algorithm are used to dynamically adjust the infusion speed, combined with the bed posture and ambient temperature, the problem that the existing technology cannot be adjusted intelligently, and the personalization and accuracy of the infusion are achieved. Improves treatment effect and patient comfort.

CN119055879AInactive Publication Date: 2024-12-03NINGBO FIRST HOSPITAL

Patent Information

Application Number
CN202411569044.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-05
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing medical infusion pumps cannot make intelligent and personalized adjustments based on the patient's real-time situation, which affects the treatment effect and the patient's comfort.

Method used

Adopt a wireless infusion pump intelligent control system, by obtaining patient diagnostic information and real-time vital sign monitoring data, using prediction models and particle swarm optimization algorithm to dynamically adjust the infusion speed, combining bed posture and ambient temperature to achieve personalized and real-time optimization of infusion speed .

Benefits of technology

It significantly improves the accuracy of infusion and the comfort of patients, reduces the risk of overdose, improves the intelligence level of medical infusion, and provides patients with a safer and more efficient treatment experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119055879A_ABST
    Figure CN119055879A_ABST
Patent Text Reader

Abstract

The invention relates to an intelligent regulation and control system for a wireless infusion pump and a medium, relates to the technical field of medical infusion control, and solves the problem of how to accurately adjust the infusion speed in real time according to the physiological status and demand of a patient. At least one processor; the intelligent regulation and control method for the wireless infusion pump comprises the following specific steps that diagnosis information of a patient is obtained, key features are extracted and input into a prediction model, and an infusion scheme is obtained; vital signs of a patient are monitored in real time, and the infusion speed is dynamically adjusted. The intelligent infusion system has the advantages that the intelligent and personalized level of the infusion process is improved, and the safety and the treatment effect of a patient are ensured.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of medical infusion control, and in particular to an intelligent control system for a wireless infusion pump and a medium. Background Art

[0002] Currently, the adjustment of infusion speed during medical infusion mostly depends on the experience and manual operation of medical staff.

[0003] Some existing high-end infusion pumps already have certain automatic adjustment functions, but most of them are adjusted based on preset programs and fixed parameters, and are unable to make intelligent and personalized adjustments based on the patient's real-time situation, which may affect the treatment effect and patient comfort. Summary of the invention

[0004] In order to improve the intelligence and personalization level of the infusion process and ensure patient safety and treatment effect, the present application provides a wireless infusion pump intelligent control system and medium.

[0005] In the first aspect, the present application provides a wireless infusion pump intelligent control system, which adopts the following technical solutions: A wireless infusion pump intelligent control system, comprising: Wireless infusion pump; at least one processor; At least one memory storing a computer program, and at least one processor configured to implement a wireless infusion pump intelligent control method when executing the computer program, the specific steps are as follows: Access to patient diagnostic information or medical records; Extract key features from the patient's diagnostic information or medical records and input them into the trained infusion plan prediction model to output the predicted infusion plan. The key features include but are not limited to symptom description, examination results, and medical history. The infusion plan includes drug selection, dosage, infusion speed, and working mode. Using the predicted infusion plan as the initial infusion plan of the wireless infusion pump, and executing the initial infusion plan of the wireless infusion pump; Real-time monitoring of the patient's real-time vital sign monitoring data during the execution of the initial infusion plan, and selecting key information related to the infusion speed adjustment features, inputting it into a dynamic model for predicting the optimal infusion speed constructed using a particle swarm optimization algorithm, and outputting the predicted optimal infusion speed; The output predicted optimal infusion rate replaces the infusion rate of the initial infusion plan, the infusion plan of the wireless infusion pump is updated, and the updated infusion plan is executed.

[0006] By adopting the above technical solutions, the intelligent control method of the wireless infusion pump significantly improves the accuracy of infusion and the comfort of patients through highly integrated and intelligent technology. First, this method uses a predictive model to accurately formulate an initial infusion plan based on the patient's medical records, ensuring the personalization and pertinence of the treatment. Secondly, through real-time monitoring and dynamic adjustment of the infusion speed, it not only guarantees the treatment effect, but also reduces the risk of drug overdose. In addition, the application of the particle swarm optimization algorithm enables the system to quickly respond to the patient's physiological changes and optimize the infusion speed. Overall, this technology has greatly improved the intelligence level of medical infusion and provided patients with a safer and more efficient treatment experience.

[0007] Optionally, the output of the predicted optimal infusion rate includes: Analyze whether to obtain the patient's bed posture data; If not, maintain the original setting; If yes, then obtain the patient's bed posture data and real-time vital sign monitoring data, and select key information related to the infusion speed adjustment; The key information of the features related to the adjustment of the infusion speed is input into a dynamic model for predicting the optimal infusion speed constructed by using a particle swarm optimization algorithm, and the predicted optimal infusion speed is output, wherein the dynamic model is trained by a data set integrating posture and vital signs.

[0008] By adopting the above technical solutions, integrating the patient's bed posture and vital signs data, and using the particle swarm optimization algorithm, the infusion speed is personalized and dynamically optimized. This innovative method not only improves the accuracy of infusion, but also ensures the patient's comfort and reduces medical risks. By adjusting the bed posture, the system can more intelligently adapt to the patient's real-time needs and improve the treatment effect. The overall technical effect reflects a significant improvement in intelligence, personalization and safety.

[0009] Optionally, the method further includes a step after replacing the infusion speed of the initial infusion plan with the output predicted optimal infusion speed and before updating the infusion plan of the wireless infusion pump, specifically as follows: Obtain the real-time temperature of the drug solution and the rate of change of the external environment temperature; Analyze and obtain the real-time temperature deviation based on the target liquid temperature and the real-time liquid temperature; Analyze and determine the infusion speed adjustment range according to the corresponding relationship between the temperature deviation range that the real-time temperature deviation falls into, the change rate range range that the external environment temperature change rate falls into, and the infusion speed adjustment range; According to the infusion speed adjustment range determined by analysis and the output predicted optimal infusion speed, the optimal infusion speed is re-analyzed and determined, and the original optimal infusion speed determined by analysis is replaced.

[0010] By adopting the above technical solution, this technical link monitors the temperature of the drug solution and the temperature changes of the external environment, and intelligently adjusts the infusion speed to ensure that the drug solution is infused into the patient's body at an appropriate temperature. This dynamic matching of temperature control and infusion speed not only improves the safety and comfort of treatment, but also effectively reduces the side effects of the drug solution caused by temperature discomfort. The overall technical effect demonstrates the combination of intelligence and humanity of medical equipment, bringing patients a better medical service experience.

[0011] Optionally, analyzing and determining the infusion rate adjustment range includes: Analyze whether to obtain the patient's bed posture data; If yes, then analyzing the patient's body position change information based on the acquired bed posture data of the patient; The infusion speed adjustment range is determined based on the correspondence between the patient's body position change information, the temperature deviation range that the real-time temperature deviation falls into, the change rate range that the external environment temperature change rate falls into, and the infusion speed adjustment range.

[0012] By adopting the above technical solutions, the patient's bed posture data, liquid medicine temperature, external ambient temperature and other multiple factors are comprehensively used to intelligently analyze and determine the adjustment range of the infusion speed. This method not only takes into account the patient's real-time position changes, but also combines the impact of ambient temperature on liquid medicine temperature, thereby ensuring the safety and effectiveness of infusion. The innovation of the technology lies in the comprehensive consideration of multiple parameters, realizing the personalization and dynamic adjustment of the infusion speed, and improving the accuracy of medical services and patient experience.

[0013] Optionally, the method further includes steps in parallel with executing the updated infusion plan, specifically as follows: Obtaining ambiguous instructions from patients; According to the mapping relationship between the patient's fuzzy instruction and the adjustment of the infusion amplitude, analyzing and determining the adjustment of the infusion amplitude matched by the patient's fuzzy instruction; Adjust the infusion speed according to the adjusted infusion amplitude matched by the patient's fuzzy instruction determined through analysis; Analyze whether the patient's ambiguous instructions are received again within the preset time range for adjusting the infusion rate; If no, remain unchanged; If yes, the process jumps to the step of analyzing and determining the step of adjusting the infusion amplitude that matches the patient's fuzzy instruction according to the mapping relationship between the patient's fuzzy instruction and the adjustment of the infusion amplitude.

[0014] By adopting the above technical solution, the infusion speed is adjusted intelligently by obtaining and analyzing the patient's fuzzy instructions, achieving the personalization and comfort of infusion therapy. Patients can adjust the infusion speed in real time through simple instructions, which improves the patient's autonomy and treatment experience. The system's continuous monitoring and immediate response mechanism ensures the continuity and safety of treatment. At the same time, the intelligent infusion speed adjustment also reduces the workload of medical staff and improves medical efficiency. The overall technical effect demonstrates the intelligence of medical equipment and the patient-centered design concept.

[0015] Optionally, analyzing and determining the adjustment of the infusion amplitude that matches the patient's ambiguous instruction includes: Obtain patient position change information and real-time temperature deviation; According to the patient's body position change information, the interval range that the real-time temperature deviation falls into, and the mapping relationship between the fuzzy instructions and the adjusted infusion amplitude, the adjusted infusion amplitude that matches the patient's fuzzy instructions is analyzed and determined.

[0016] By adopting the above technical solution, the patient's body position change information and the real-time temperature deviation of the drug solution are comprehensively considered, and combined with fuzzy instruction intelligent analysis, a more accurate infusion adjustment range is determined. This not only ensures the safety and comfort of the infusion process, but also greatly improves the personalization and flexibility of treatment. The patient's fuzzy instructions can be effectively recognized by the system and converted into specific infusion range adjustments, realizing real-time matching of patient needs and treatment plans, and further improving the quality and efficiency of medical services.

[0017] Optionally, the method further includes adjusting the infusion amplitude and the infusion speed in accordance with the patient's fuzzy instruction determined by the analysis, as follows: Obtain the number of times the patient uses fuzzy commands within a preset time range and background information, the background information including the patient's body position change information and real-time temperature deviation; If the number of times a patient with the same background information uses fuzzy instructions exceeds the preset number within the preset time range, the average infusion amplitude actually adjusted by the fuzzy instructions for the same background information is obtained, and the average infusion amplitude is used as the infusion adjustment amplitude on the original basis under the corresponding background information.

[0018] By adopting the above technical solution, the infusion amplitude can be intelligently adjusted by collecting and analyzing the frequency of fuzzy instructions used by patients in specific contexts. This data-driven approach enables the system to dynamically optimize the infusion adjustment strategy based on the patient's actual usage. This not only improves the personalization and accuracy of infusion therapy, but also reduces the number of patient operations in specific situations, thereby improving the patient's comfort and treatment experience. The innovation of the technology lies in combining patient usage habits and background information to achieve intelligent and adaptive infusion adjustment.

[0019] Optionally, the method further includes the following steps after real-time monitoring of the patient's real-time vital sign monitoring data during the execution of the initial infusion plan: Extracting sleep-related parameters from the patient's real-time vital sign monitoring data, and analyzing and determining the sleeping result based on a mapping relationship between the result of the sleep-related parameter range that the sleep-related parameter falls into and the sleeping result; If the result of falling asleep is being awake, maintain the original setting; If the result of falling asleep is that the patient is in a sleeping state, the patient's current physiological parameters are obtained and input into a trained model for predicting the patient's drug needs in a sleeping state, and the predicted drug needs in different time periods at night are output; The infusion speed is determined by analyzing the mapping relationship between the predicted drug demand ranges for different time periods at night and the infusion speed.

[0020] By adopting the above technical solutions, real-time vital signs monitoring is innovatively combined with drug demand prediction to provide patients with personalized treatment plans. By extracting sleep-related parameters and intelligently analyzing the results of falling asleep, the system can automatically adjust the infusion speed after the patient falls asleep to meet the drug needs at different time periods at night. This not only ensures the health and safety of patients, but also improves the treatment effect and reduces unnecessary drug waste. Overall, this technology demonstrates the humanization and intelligent development trend of medical insurance.

[0021] Optionally, the method further includes a step of analyzing and determining the adjustment of the infusion amplitude matched by the patient's fuzzy instruction after obtaining the patient's fuzzy instruction and according to the mapping relationship between the patient's fuzzy instruction and the adjustment of the infusion amplitude, which is specifically as follows: Obtaining the analyzed and determined sleep onset results; If the result of falling asleep is being awake, continue with the next steps; If the result of falling asleep is in the sleeping state, stop the subsequent steps.

[0022] By adopting the above technical solution, the patient's sleeping results are obtained and analyzed, and the subsequent steps are intelligently judged whether to continue to execute, thus realizing the refined and humanized management of infusion therapy. If the patient is awake, the system will continue to analyze the patient's fuzzy instructions and adjust the infusion amplitude accordingly to meet the patient's needs; if the patient has fallen asleep, the subsequent operations will be automatically suspended to avoid disturbing the patient's rest. This intelligent control not only improves the treatment effect, but also reflects the humanistic care of medical technology for patients, and is an important step in improving the quality of medical services.

[0023] Optionally, the method further includes a step of analyzing and determining the infusion speed in parallel with the mapping relationship between the drug demand range and the infusion speed according to the predicted drug demand range in different time periods at night, which is specifically as follows: Analyze whether the patient's bed has posture adjustment function; If yes, then according to the correspondence between the infusion speed interval range that the infusion speed during the sleeping period falls into and the posture adjustment amplitude, the posture adjustment amplitude of the bed on which the patient sleeps is determined by analysis, and the bed on which the patient sleeps is adjusted according to the posture adjustment amplitude; If no, remain unchanged.

[0024] By adopting the above technical solution, this technical solution further improves the personalization of medical services by intelligently analyzing the function of the patient's bed. If the bed has the ability to adjust its posture, the system will intelligently adjust the bed's posture according to the patient's infusion speed, which will not only help the patient's comfort and sleep quality, but also improve the distribution of the drug in the body, thereby improving the treatment effect. If the bed does not have this function, it will remain in its original state to ensure that the patient sleeps soundly. This innovation demonstrates the meticulous care and precision of medical technology for patients.

[0025] In a second aspect, the present application provides a computer storage medium, which adopts the following technical solution: A computer storage medium includes a memory, a processor, and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement a method for intelligently controlling a wireless infusion pump in an intelligent control system for a wireless infusion pump as described in the first aspect.

[0026] In summary, the beneficial technical effects of this application are: 1. The method of dynamically adjusting the infusion speed according to the patient's diagnostic information and real-time monitoring data makes the infusion process more personalized and intelligent, improving the treatment effect and patient satisfaction; 2. The infusion speed is adjusted based on the patient's fuzzy instructions. This design is more humane and can respond to the patient's real feelings and needs in a timely manner, further improving the safety and comfort of the infusion process. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the overall process of a wireless infusion pump intelligent control method in a wireless infusion pump intelligent control system in an embodiment of the present application.

[0028] Figure 2 It is a flowchart of outputting the predicted optimal infusion rate according to another embodiment of the present application.

[0029] Figure 3It is a flowchart diagram of the steps of another embodiment of the present application after replacing the infusion rate of the initial infusion plan with the output predicted optimal infusion rate and before updating the infusion plan of the wireless infusion pump.

[0030] Figure 4 It is a flowchart of another embodiment of the present application for analyzing and determining the adjustment range of the infusion speed.

[0031] Figure 5 It is a flowchart of the steps of the embodiment of the present application in parallel with the execution of the updated infusion plan. DETAILED DESCRIPTION

[0032] The present application is further described in detail below in conjunction with the accompanying drawings.

[0033] Reference Figure 1 , is a wireless infusion pump intelligent control system disclosed in the present application, comprising a remote pelvic inflammatory disease treatment device, at least one processor and at least one memory storing a computer program, at least one processor is configured to implement a remote pelvic inflammatory disease treatment method when executing the computer program, and the specific steps are as follows: Step S100, obtaining the patient's diagnosis information or medical records.

[0034] Among them, the patient's diagnostic information or medical records include but are not limited to the patient's symptom description, test results, medical history, etc.

[0035] Taking a patient with acute gastritis as an example, in step S100, we need to obtain the following key diagnostic information and medical records: Symptom description: The patient complained of stomach pain, nausea, vomiting and other symptoms, which lasted for about X hours. Examination results: Blood routine showed an increase in white blood cell count, indicating infection; gastroscopy revealed congestion and edema of the gastric mucosa. Medical history: The patient had no similar medical history in the past, but had irregular diet and frequently stayed up late recently. This information will be used in the subsequent infusion plan prediction model to help us more accurately predict the infusion plan suitable for the patient.

[0036] Step S200, extracting key features from the patient's diagnostic information or medical records, and inputting them into the trained infusion scheme prediction model, and outputting the predicted infusion scheme.

[0037] Among them, the patient's diagnosis information or medical records usually exist in text form. Extracting key features from the patient's diagnosis information or medical records mainly uses natural language processing (NLP) technology to automatically identify and extract key features. Through lexical analysis, syntactic analysis, and semantic understanding, NLP can intelligently parse text data and extract key information related to the infusion plan, such as symptom description, examination results, and medical history.

[0038] After extracting the key information related to the infusion plan, the extracted key information will be further processed and handled, including data cleaning, feature selection, feature transformation and feature encoding. The key features obtained subsequently include but are not limited to the description of symptoms, examination results, and medical history. The infusion plan includes the selection of drug solution, dosage, infusion speed, and working mode.

[0039] Suppose there is a patient diagnosed with pneumonia, and his diagnostic information includes key features such as "fever, cough, chest X-ray shows shadows in the lungs". In step S200, NLP technology is first used to extract key symptom descriptions and examination results from these text information. Next, feature engineering technology will further process these features, such as converting "fever" into a specific body temperature value, and converting "chest X-ray shows shadows in the lungs" into corresponding imaging feature codes. Finally, these processed features will be input into the pre-trained infusion plan prediction model, which may output an accurate infusion plan such as "use antibiotic A, the dosage is X mg, infused once every Y hours, for Z days".

[0040] Step S300: Using the predicted infusion plan as the initial infusion plan of the wireless infusion pump, and executing the initial infusion plan of the wireless infusion pump.

[0041] Step S400, real-time monitoring of the patient's real-time vital sign monitoring data during the execution of the initial infusion plan, and selecting key information related to the infusion speed adjustment features, inputting it into a dynamic model for predicting the optimal infusion speed constructed using a particle swarm optimization algorithm, and outputting the predicted optimal infusion speed.

[0042] The real-time data monitoring involved in step S400 continuously collects the patient's vital signs data, such as heart rate, blood pressure, blood oxygen saturation, etc., through medical monitoring equipment, such as electrocardiographs, sphygmomanometers, and oximeters. These data reflect the patient's physiological state during the infusion process and are the basis for subsequent analysis and adjustment.

[0043] The key information for selecting features related to infusion speed adjustment is mainly to select feature information closely related to infusion speed adjustment from real-time monitoring data. For example, when the patient's heart rate is detected to be significantly accelerated, it may mean that the current infusion speed is too fast and needs to be adjusted.

[0044] Dynamic model prediction is a dynamic model built using particle swarm optimization algorithm, which can predict the current optimal infusion rate based on the selected key feature information. Particle swarm optimization algorithm is an optimization algorithm that simulates the foraging behavior of bird flocks. It finds the optimal solution to the problem through information sharing and collaboration among groups. Here, it is used to optimize the infusion rate to achieve a safe and efficient infusion effect.

[0045] For example, suppose a patient is receiving infusion therapy, and the system monitors that his heart rate suddenly increases and his blood pressure also rises. These changes may indicate that the current infusion rate is too fast for the patient. In step S400, the following process is roughly as follows: 1. Data collection: The system first obtains the patient's heart rate and blood pressure data in real time through medical monitoring equipment. 2. Feature selection and preprocessing: The system automatically identifies abnormal changes in heart rate and blood pressure, and preprocesses these data for subsequent analysis. 3. Dynamic model prediction: The dynamic model constructed using the particle swarm optimization algorithm predicts a more appropriate infusion rate based on the preprocessed data, such as adjusting from the original 50ml / h to 30ml / h. 4. Infusion rate adjustment: The system automatically adjusts the infusion rate of the wireless infusion pump to 30ml / h to adapt to the patient's current physiological state.

[0046] Step S500, replacing the infusion speed of the initial infusion plan with the output predicted optimal infusion speed, updating the infusion plan of the wireless infusion pump, and executing the updated infusion plan.

[0047] Reference Figure 2 , the output predicted optimal infusion rate includes: Step S210, analyzing whether the patient's bed posture data is obtained. If not, executing step S220; if yes, executing step S230.

[0048] The bed posture data of the patient mentioned in step S210 may be obtained by detecting a sensor provided on the patient's bed and capable of monitoring the bed posture in real time, or may be obtained by real-time monitoring of changes in the bed posture of the patient through a monitoring device provided.

[0049] Specifically, whether the bed posture data of the patient has been obtained can be determined by checking whether the system has successfully obtained the bed posture data of the patient through an appropriate data interface.

[0050] Step S220, maintaining the original settings.

[0051] Step S230, obtaining the patient's bed posture data and real-time vital sign monitoring data, and selecting key information related to the infusion speed adjustment features.

[0052] The acquisition of the patient's bed posture data and real-time vital signs monitoring data first requires the use of high-precision, non-invasive sensors to continuously monitor the patient's bed posture and vital signs, and then the bed posture data and vital signs data collected from different sensors are fused and processed to obtain a data set that fully reflects the patient's current status. Finally, advanced algorithms are used to extract key features related to infusion speed adjustment from the fused data. These features may include the patient's position change frequency, heart rate change trend, etc.

[0053] In step S240, key information of features related to infusion speed adjustment is input into a dynamic model for predicting the optimal infusion speed constructed using a particle swarm optimization algorithm, and the predicted optimal infusion speed is output, wherein the dynamic model is trained by a data set integrating posture and vital signs.

[0054] The specific technologies involved in the implementation are as follows: 1. Particle swarm optimization algorithm: This is an intelligent optimization algorithm that searches for the optimal solution to the problem by simulating the foraging behavior of bird flocks. In step S240, the algorithm is used to adjust the parameters of the dynamic model to make the infusion rate prediction output by the model more accurate. 2. Dynamic modeling technology: The system uses dynamic modeling technology to build a model that can predict the optimal infusion rate based on real-time data. This model not only takes into account the patient's bed posture and vital signs data, but also can dynamically adjust the prediction results according to changes in these data. 3. Machine learning technology: The construction and training process of the dynamic model involves machine learning technology. The system uses historical data to train the model so that it can learn the relationship between the infusion rate and factors such as patient posture and vital signs. 4. Data input and output technology: Step S240 also needs to accurately input the extracted key feature information into the dynamic model and receive the prediction results output by the model.

[0055] For example, suppose that during the infusion process, a patient's bed posture changes from lying flat to lying on the side, and his heart rate increases. In step S230, the system has successfully extracted these key information related to the adjustment of the infusion speed. In step S240, this information will be input into the dynamic model constructed using the particle swarm optimization algorithm. The model will predict a new optimal infusion speed based on the patient's real-time posture and vital signs data, as well as the rules and patterns learned from historical data. For example, if the model determines that the side-lying posture and increased heart rate may lead to slower drug absorption, it may recommend increasing the infusion speed to ensure effective absorption of the drug. This prediction result will serve as the basis for the system to adjust the infusion speed, thereby ensuring that the patient can receive the drug treatment that best suits the current state.

[0056] Reference Figure 3A wireless infusion pump intelligent control system further includes steps after replacing the infusion speed of the initial infusion plan with the output predicted optimal infusion speed and before updating the infusion plan of the wireless infusion pump, which are as follows: Step SA00, obtaining the real-time temperature of the drug solution and the rate of change of the external environment temperature.

[0057] The real-time temperature of the drug solution can be obtained by setting a high-precision temperature sensor in the drug delivery pipeline, and the external environment temperature change rate can be obtained by setting a high-precision temperature sensor in the external environment of the infusion pump to detect the external environment temperature.

[0058] Step SB00, analyzing and obtaining the real-time temperature deviation based on the target liquid medicine temperature and the real-time liquid medicine temperature.

[0059] Specifically, after the processor receives the real-time liquid temperature data, it will compare it with the preset target liquid temperature. Through a specific algorithm, the system will calculate the difference between the two, that is, the real-time temperature deviation.

[0060] Step SC00, analyzing and determining the infusion speed adjustment range according to the corresponding relationship between the temperature deviation interval into which the real-time temperature deviation falls, the change rate interval range into which the external environment temperature change rate falls, and the infusion speed adjustment range.

[0061] Specifically, step SC00 involves interval matching technology. Specifically, the system can set a series of temperature deviation intervals and ambient temperature change rate intervals based on previous data and medical standards. When the real-time monitored data falls into these intervals, the system will automatically match the corresponding infusion speed adjustment range.

[0062] For example, assume that at a certain moment, the system detects that the real-time temperature of the liquid medicine is 2°C lower than the target temperature, while the external ambient temperature is decreasing at a rate of 0.5°C / minute. According to the preset range, this temperature deviation and the rate of change of the external ambient temperature correspond to a specific increase in the infusion rate, such as 5%.

[0063] Through intelligent analysis and calculation, the system determines that in the current situation, in order to maintain the target temperature of the liquid medicine, the infusion speed needs to be increased by 5%. In this way, even if the external ambient temperature continues to drop, the temperature of the liquid medicine can reach and remain within the target temperature range faster due to the increase in the infusion speed, thereby ensuring the treatment effect and patient comfort.

[0064] Step SD00, based on the analyzed and determined infusion speed adjustment range and the output predicted optimal infusion speed, re-analyze and determine the optimal infusion speed, and replace the original analyzed and determined optimal infusion speed.

[0065] For example, assuming that in a certain environment, the system analyzes steps SA00 to SC00 and concludes that the real-time temperature of the liquid medicine is 25°C, the external environment temperature change rate is +1°C / minute, the real-time temperature deviation is -2°C, and the corresponding infusion speed adjustment range is increased by 5%. In step SD00, the system will recalculate an optimal infusion speed through a dynamic optimization algorithm based on these data and the originally predicted optimal infusion speed (for example, 60ml / h). After calculation, the system may conclude that in the current environment, in order to maintain the stability of the liquid medicine temperature and reach the target temperature as soon as possible, the infusion speed needs to be increased to 63ml / h (i.e., an increase of 5% on the original basis). Subsequently, the system will immediately send this optimized speed instruction to the wireless infusion pump, and the infusion pump will adjust its infusion speed to 63ml / h accordingly after receiving the instruction.

[0066] Reference Figure 4 , the analysis and determination of the infusion speed adjustment range include: Step SC10, analyzing whether the patient's bed posture data is obtained.

[0067] Specifically, the patient's bed posture data can be obtained by using sensors integrated on the bed, such as tilt sensors, position sensors, etc. to monitor and record the bed's posture in real time.

[0068] Step SC20: if yes, analyzing the patient's body position change information based on the acquired bed posture data of the patient.

[0069] The patient's position change information is obtained as follows: First, by comparing the bed posture data at different time points, the system can determine whether the patient changes from a supine position to a lateral position, or stands up from a supine position, etc. Then, after identifying the position change, the system will further extract the characteristics of these changes and quantify them. This includes but is not limited to the amplitude, speed, and frequency of the change.

[0070] Step SC30, analyzing and determining the infusion speed adjustment range according to the correspondence between the patient's body position change information, the temperature deviation range into which the real-time temperature deviation falls, the change rate range into which the external environment temperature change rate falls, and the infusion speed adjustment range.

[0071] Among them, the implementation process of step SC30 is as follows: the system will first integrate the patient's body position change information obtained in step SC20 with real-time temperature deviation, external environment temperature change rate and other data. These data are dynamically changing, so they need to be updated and monitored in real time. Next, the system will analyze the correspondence between these data and the infusion speed adjustment range. For example, certain changes in body position may require speeding up the infusion speed to ensure that the drug is delivered in time, while certain temperature deviations may require slowing down the infusion speed to prevent the drug from entering the body too quickly and causing adverse reactions. Based on the above analysis, the system will determine a specific infusion speed adjustment range.

[0072] Suppose a patient is receiving intravenous infusion therapy, and the system monitors that the patient changes from supine to lateral position, and the real-time temperature deviation shows that the patient's body temperature has risen slightly, while the external environment temperature is also gradually rising. In step SC30, the system will take these factors into consideration: the change in body position from supine to lateral may mean that some of the patient's blood vessels are compressed, and the infusion rate needs to be adjusted appropriately to prevent the accumulation or poor flow of the drug solution. The increase in body temperature may indicate that the patient has a slight reaction to the drug, which may require slowing down the infusion rate to reduce the impact of the drug on the patient. The increase in external environmental temperature may also affect the patient's physiological state, such as speeding up metabolism, which also needs to be considered in the adjustment of the infusion rate. Based on the above considerations, the system may decide to slightly reduce the infusion rate to ensure the safety of the patient's treatment in the current state.

[0073] Reference Figure 5 , a wireless infusion pump intelligent control system also includes steps in parallel with executing the updated infusion plan, specifically as follows: Step SE00, obtaining the patient's fuzzy instructions.

[0074] In step SE00, the system captures the fuzzy instructions issued by the patient through specific interaction methods, such as voice recognition, touch sensing or gesture recognition, etc. These fuzzy instructions are usually the patient's willingness to adjust the infusion speed based on his or her own feelings or needs, expressed in a non-specific and imprecise way.

[0075] Suppose a patient who is using a wireless infusion pump feels uncomfortable and wants to reduce the infusion rate slightly. He may verbally issue vague instructions such as "slow down" or "too fast", or express his need for speed adjustment through gestures.

[0076] Step SF00, analyzing and determining the adjusted infusion amplitude matched by the patient's fuzzy instruction according to the mapping relationship between the patient's fuzzy instruction and the adjusted infusion amplitude.

[0077] Specifically, the mapping relationship between the patient's fuzzy instructions and the adjustment of the infusion amplitude is a pre-built detailed mapping rule library, which contains the correspondence between various fuzzy instructions and specific infusion amplitude adjustment schemes. This rule library is continuously optimized and improved based on medical knowledge, clinical experience and patient feedback.

[0078] When the system receives ambiguous instructions from patients, it first uses natural language processing technology to parse these instructions. This includes speech-to-text conversion, keyword extraction, semantic analysis and other steps, aiming to accurately capture the patient's true intentions.

[0079] After parsing the patient's fuzzy instructions, the system matches them with entries in the mapping rule base. Once the best match is found, the system immediately determines the corresponding infusion amplitude adjustment plan.

[0080] Assume that the patient feels uncomfortable during the infusion process and issues a vague instruction such as "slow down". In step SF00, the system will quickly capture this instruction and parse it using natural language processing technology.

[0081] Through keyword extraction and semantic analysis, the system determines that the patient's intention is to reduce the infusion rate. Next, the system searches the mapping rule library for the infusion adjustment scheme corresponding to "slower". Based on the preset rules, the system may decide to reduce the infusion rate by 10% or 20%, depending on the patient's current condition and the previous infusion rate.

[0082] Step SG00, adjust the infusion speed according to the adjusted infusion amplitude matched by the patient's fuzzy instruction determined through analysis.

[0083] Assume that in step SF00, the system analyzes that the fuzzy instruction "hurry up" issued by the patient means that the infusion speed needs to be increased by 20%. In step SG00, the system performs the following operations: Calculate the new speed: The system first obtains the current infusion speed, for example, 60mL / h. Then, based on the 20% increase, the new infusion speed is calculated to be 72mL / h (60mL / h *1.2). Adjust the infusion pump: The system sends an instruction to the infusion pump, requiring it to smoothly adjust the infusion speed to 72mL / h. At the same time, the system monitors the speed change of the infusion pump in real time to ensure that the actual speed is consistent with the calculated value. Safety monitoring: During the adjustment process, if the system detects that the infusion pump speed is abnormal, such as too fast or too slow, it will immediately trigger a safety warning and take corresponding measures, such as pausing the infusion pump, notifying medical staff, etc., to ensure the safety of the patient.

[0084] Step SH00, analyzing whether the patient's fuzzy instruction is received again within the preset time range for adjusting the infusion speed. If not, execute step SI00; if yes, execute step SJ00.

[0085] Specifically, after the infusion speed is adjusted according to the patient's fuzzy instruction, the system will enter a preset monitoring period. During this period, the system will continue to detect whether there is a new fuzzy instruction input. If the patient's fuzzy instruction is received again during the monitoring period, the system will determine whether the instruction requires further infusion speed adjustment.

[0086] Assume that the patient feels uncomfortable during the infusion process and issues a vague instruction of "slow down", and the system reduces the infusion speed according to this instruction. After that, the system will enter the monitoring stage of step SH00.

[0087] No new instructions: If the patient does not issue any new instructions within the next few minutes (preset time range), the system will consider that the current infusion speed meets the patient's needs, and then execute step SI00 to maintain the current infusion speed unchanged.

[0088] New instruction: If the patient feels uncomfortable again during the monitoring period and issues a new fuzzy instruction, such as "slow down", the system will immediately capture this instruction and execute step SJ00. This means that the system will re-analyze the mapping relationship between the new instruction and the adjustment of the infusion amplitude, and further adjust the infusion speed accordingly to meet the patient's real-time needs. This design ensures the flexibility of the infusion process and the comfort of the patient.

[0089] Step SI00, remain unchanged.

[0090] Step SJ00, jump to the step of analyzing and determining the step of adjusting the infusion amplitude that matches the patient's fuzzy instruction according to the mapping relationship between the patient's fuzzy instruction and the adjustment of the infusion amplitude.

[0091] Step SJ00 is not a one-time operation, but is performed continuously during the infusion process. As long as the patient continues to issue ambiguous instructions within the preset time range, the system will continue to repeat step SJ00 to ensure that the infusion speed is consistent with the patient's needs.

[0092] Suppose that when a patient uses a wireless infusion pump, he or she first issues a vague instruction of "slow down" because of discomfort, and the system reduces the infusion speed according to this instruction. However, after a period of time, the patient may issue a new instruction of "faster" due to changes in his or her condition or personal feelings.

[0093] After receiving this new instruction, step SH00 will determine that the patient's ambiguous instruction has been received again within the preset time range, and trigger the execution of step SJ00. At this time, the system will re-analyze the mapping relationship between the instruction "hurry up" and the adjustment of the infusion amplitude, and determine the infusion amplitude that needs to be increased.

[0094] The adjustment of the infusion range matched by the patient's ambiguous instructions is analyzed and determined, including: Step SF10, obtaining the patient's body position change information and real-time temperature deviation.

[0095] The patient's position change information is obtained as follows: The patient's position change is monitored in real time through the intelligent sensor integrated in or connected to the wireless infusion pump system. For example, when the patient changes from lying to sitting or standing, the position change data will be captured and recorded immediately. This information helps the system understand the patient's activity status and adjust the infusion speed more accurately.

[0096] The real-time temperature deviation of the patient is obtained as follows: the temperature of the patient's surrounding environment is monitored by a temperature sensor, and compared with the preset comfortable temperature range to calculate the real-time temperature deviation.

[0097] Step SF20, analyzing and determining the adjusted infusion amplitude matched by the patient's fuzzy instruction according to the patient's body position change information, the interval range into which the real-time temperature deviation falls, and the mapping relationship between the fuzzy instruction and the adjusted infusion amplitude.

[0098] Specifically, the system first integrates the patient's position change information and real-time temperature deviation obtained in step SF10. The system analyzes the integrated data based on the mapping relationship between the pre-set fuzzy instructions and the adjustment of the infusion amplitude. Through analysis, the system can accurately match the patient's fuzzy instructions (such as "faster", "slower", etc.) to the specific infusion adjustment amplitude. This ensures that the infusion speed can accurately reflect the patient's real-time wishes and needs. Finally, the system outputs the determined infusion amplitude adjustment plan for use in the subsequent infusion speed adjustment steps.

[0099] Assume that a patient is receiving intravenous infusion and issues a vague instruction of "a little faster". At this time, the system has obtained through step SF10 that the patient's body position has changed from supine to semi-recumbent, and the real-time temperature deviation shows that the ambient temperature has slightly decreased.

[0100] In step SF20, the system first integrates this information and recognizes that changes in the patient's body position may lead to changes in venous return pressure, and a drop in ambient temperature may affect the stability of the drug and the patient's comfort. Then, based on the preset mapping relationship, the system analyzes that the instruction "slightly faster" should correspond to a moderate increase in the infusion speed under the current body position and temperature conditions.

[0101] Ultimately, the system outputs a specific adjustment plan, such as increasing the infusion speed by 10%, to ensure that the patient's needs are met while ensuring the safety and comfort of the infusion process.

[0102] A wireless infusion pump intelligent control system also includes the steps of adjusting the infusion amplitude and adjusting the infusion speed in parallel according to the patient fuzzy instructions determined by analysis, which are specifically as follows: Step Sa00, obtaining the number of times the patient uses fuzzy instructions within a preset time range and background information, the background information including the patient's body position change information and real-time temperature deviation.

[0103] Specifically, the system will track and collect the number of times the patient uses fuzzy instructions (such as "faster", "slower", etc.) within a preset time range in real time. In addition to recording the use of fuzzy instructions, the system will also collect relevant background information, including changes in the patient's body position (such as from lying to sitting) and real-time temperature deviation (the difference between the ambient temperature and the preset comfort temperature).

[0104] Step Sb00, if the number of times the patient with the same background information uses fuzzy instructions exceeds the preset number within the preset time range, the average infusion amplitude actually adjusted by the fuzzy instructions for the same background information is obtained, and the average infusion amplitude is used as the infusion adjustment amplitude on the original basis under the corresponding background information.

[0105] Specifically, the system will look for historical records of patients using fuzzy instructions to adjust the infusion speed under the same background information (such as specific body position and temperature deviation) within a preset time range. Then, the system will count the number of times the patient uses fuzzy instructions under these background information and compare it with the preset number threshold. If the threshold is exceeded, it means that the patient's infusion needs have a certain regularity and stability under this background. After confirming that the use of fuzzy instructions under specific background information has reached a certain frequency, the system will calculate the average infusion amplitude actually adjusted by these instructions. This average value reflects the average demand of patients for infusion speed adjustment under this background. Finally, the system sets this average infusion amplitude as the infusion adjustment amplitude on the original basis under the corresponding background information.

[0106] For example, suppose the system presets a time range of 24 hours and a preset number of times of 3. In the past 24 hours, the system detected that the patient used the "speed up" fuzzy command more than 3 times to adjust the infusion speed when the patient was in a supine position and the ambient temperature was moderate (such as 22-24°C). The system traced back the data within these 24 hours and found that the patient used the "speed up" command 5 times in total under the above background. These 5 commands increased the infusion speed by 5%, 7%, 6%, 5% and 7% respectively. The system calculated the average adjustment range to be (5+7+6+5+7) / 5 = 6%. Therefore, in the future, when the patient is in similar background information again (supine position, moderate ambient temperature) and issues the "speed up" command, the system will default to increasing the infusion speed by 6%, and there is no need to analyze the adjustment range from scratch. This not only improves the response speed of the system, but also increases the accuracy of the adjustment.

[0107] A wireless infusion pump intelligent control system further includes steps after real-time monitoring of the patient's real-time vital sign monitoring data during the execution of the initial infusion plan, specifically as follows: Step Sc00, extracting sleep-related parameters from the patient's real-time vital sign monitoring data, and analyzing and determining the sleeping result based on the mapping relationship between the result of the sleep-related parameter range that the sleep-related parameters fall into and the sleeping result.

[0108] The process of step Sc00 is as follows: First, the system extracts parameters closely related to sleep from the patient's real-time vital signs monitoring data. These parameters may include but are not limited to heart rate, respiratory rate, blood oxygen level, and body movement, which can effectively reflect the patient's sleep status and sleep quality.

[0109] Next, the system will analyze the extracted sleep-related parameters based on the mapping relationship between the pre-set sleep-related parameter range and the sleeping result. This mapping relationship is based on a large amount of medical research and clinical experience, and can accurately judge the patient's sleeping condition based on the parameter value. By comparing the mapping relationship, the system can determine the patient's current sleeping result, that is, whether he is awake or asleep.

[0110] Step Sd00: if the result of falling asleep is being awake, the original setting is maintained.

[0111] Step Se00, if the result of falling asleep is in a sleeping state, the patient's current physiological parameters are obtained and input into a trained model for predicting the patient's drug needs in a sleeping state, and the predicted drug needs in different time periods at night are output.

[0112] First, the system obtains the patient's current physiological parameters, such as heart rate, blood pressure, blood oxygen saturation, etc. Next, these physiological parameters are input into a trained machine learning model. This model focuses on predicting the patient's medication needs when they are asleep. The model analyzes these parameters and outputs predicted values ​​for medication requirements at different time periods at night. The results of the model's predictions will be used for subsequent infusion rate adjustment decisions. These predicted values ​​reflect the amount of medication that the patient may need at different time periods at night, thereby helping the system to regulate the infusion pump more intelligently.

[0113] The training process of the model for predicting the patient's drug demand in the sleep state is as follows: Before model training, the patient's historical physiological parameter data is first preprocessed. This includes data cleaning, normalization, and possible feature engineering steps to extract features related to drug demand. The model architecture uses a recurrent neural network (RNN) in deep learning. Training data set construction: The training set is constructed using the patient's historical data, where the input features are the patient's physiological parameter sequence (such as heart rate, blood pressure, blood oxygen saturation, etc.), and the label is the drug demand in the corresponding time period. The data set should cover the changes in the patient's physiological parameters in different sleep states. Model training is as follows: The model is trained using a back propagation algorithm and an optimizer (such as Adam or RMSprop). The training goal is to minimize the error between the predicted drug demand and the actual drug demand, usually using the mean square error (MSE) as the loss function. Use the validation set to evaluate the performance of the model and perform model tuning as needed. Tuning may include adjusting the network structure, increasing or decreasing the number of hidden layer nodes, changing the activation function, adjusting the learning rate, etc. After training, the model is deployed to the wireless infusion pump system. During the real-time monitoring of the initial infusion plan, if the system detects that the patient has fallen asleep, the patient's current physiological parameters are obtained and input into the trained model. The model predicts the drug requirements at different time periods at night based on these parameters.

[0114] Assume that a patient falls asleep at 10 o'clock in the evening. At this time, the wireless infusion pump intelligent control system has already known that the patient is in a sleeping state through the judgment of steps Sc00 and Sd00.

[0115] In step Se00, the system first obtains the patient's current physiological parameters such as heart rate, blood pressure and blood oxygen saturation. These data show that the patient's heart rate is relatively stable, blood pressure is within the normal range, and blood oxygen saturation is also maintained well.

[0116] The system then inputs these physiological parameters into a trained machine learning model. After analysis, the model predicts the patient's medication requirements at different times of the night. For example, the model may predict that the patient's medication requirements will increase slightly between 1 a.m. and 3 a.m., and that the medication requirements will gradually decrease between 4 a.m. and 6 a.m.

[0117] Step Sf00, analyzing and determining the infusion speed according to the mapping relationship between the drug demand range into which the drug demand in different time periods at night falls and the infusion speed.

[0118] First, the system maps the predicted drug demand with the preset drug demand range. Different drug demand ranges correspond to different infusion speeds. Based on the mapping relationship, the system determines the optimal infusion speed for each time period at night. This speed can meet the patient's drug needs while avoiding overdose or underdose. Finally, the wireless infusion pump will automatically adjust according to the infusion speed determined by the system. In this way, even after the patient falls asleep, the infusion pump can be intelligently regulated according to actual needs.

[0119] Suppose the system predicts that a patient's drug demand is moderate between midnight and 2 a.m., slightly higher between 2 a.m. and 4 a.m., and gradually decreases between 4 a.m. and 6 a.m.

[0120] In step Sf00, the system maps these predicted values ​​to a preset range of drug requirements. For example, a medium drug requirement may correspond to a medium-speed infusion, a slightly higher drug requirement may correspond to a fast infusion, and a gradually decreasing requirement may correspond to a gradually slowing infusion speed.

[0121] Based on these mapping relationships, the system will automatically adjust the infusion pump to medium speed between midnight and 2 a.m., to fast speed between 2 a.m. and 4 a.m., and gradually reduce the infusion speed between 4 a.m. and 6 a.m.

[0122] A wireless infusion pump intelligent control system also includes steps after obtaining the patient's fuzzy instruction and before analyzing and determining the adjustment of the infusion amplitude matched by the patient's fuzzy instruction according to the mapping relationship between the patient's fuzzy instruction and the adjustment of the infusion amplitude, specifically as follows: Step Sg00, obtaining the analyzed and determined sleeping result. Step Sh00, if the sleeping result is awake, continuing to the subsequent steps.

[0123] Step Si00, if the result of falling asleep is in the sleeping state, stop the subsequent steps.

[0124] A wireless infusion pump intelligent control system also includes a mapping relationship between the drug demand range and the infusion speed according to the predicted drug demand in different time periods at night, and a step of analyzing and determining the infusion speed in parallel, specifically as follows: analyzing whether the bed where the patient sleeps has a posture adjustment function; If yes, then analyze and determine the posture adjustment amplitude of the patient's bed based on the correspondence between the infusion speed interval range that the infusion speed during sleep falls into and the posture adjustment amplitude, and adjust the patient's bed based on the posture adjustment amplitude; if no, maintain unchanged.

[0125] First, we analyze whether the bed the patient sleeps on has a posture adjustment function. This posture adjustment bed can adjust the height and angle as needed to help patients find the most comfortable sleeping position.

[0126] If the bed the patient sleeps on has a posture adjustment function, we will further determine the adjustment range of the bed's posture based on the infusion rate. Specifically, we will divide the infusion rate into different intervals and set the corresponding posture adjustment range for each range. For example, when the infusion rate is in a lower range, the bed's posture may only need to be fine-tuned to ensure the patient's comfort; when the infusion rate increases, the bed's posture may need to be adjusted more significantly to adapt to the patient's medication needs and physical condition.

[0127] Further, environmental factors and the patient's autonomous control options can also be considered. A wireless infusion pump intelligent control system also includes a mapping relationship between the drug demand range and the infusion speed according to the predicted drug demand in different time periods at night, and a step of analyzing and determining the infusion speed in parallel, which is as follows: Step 1: Analyze whether the patient's bed has a posture adjustment function. If it has a posture adjustment function, execute step 2; if not, maintain it unchanged. Step 2: Obtain the infusion speed during the patient's sleep and determine the infusion speed range to which it belongs. At the same time, monitor environmental factors such as the temperature, humidity and noise level of the ward in real time. Step 3: According to the correspondence between the preset infusion speed range, environmental factors and posture adjustment amplitude, analyze and calculate the posture adjustment amplitude of the patient's bed, and adjust the patient's bed according to the amplitude. For example, when the infusion speed is in a higher range, and the temperature, humidity and noise level of the ward are too high, the posture adjustment amplitude is increased accordingly to improve the patient's sleeping comfort and infusion effect. Step 4: Set a safety warning mechanism. Determine the safety threshold range of the infusion speed and posture adjustment amplitude. When the infusion speed or posture adjustment range approaches or exceeds the safety threshold, the system automatically sends out an early warning signal to remind medical staff to evaluate and intervene. Step 5: On the basis of strictly ensuring medical safety, a certain degree of autonomous control authority is granted to specific control command sources. Specifically, the specific control command source sends control commands to the system through a specific operation interface, such as a mobile application or a bedside control panel. The system receives and analyzes the commands. If the commands are within the pre-set safety range, the infusion speed or bed posture is adjusted accordingly. To ensure safety, the system sends the current adjustable parameter range and the corresponding safety prompt information to the operation interface for display. For example, the upper and lower limits of the infusion speed adjustment and the maximum angle of the bed posture adjustment are clearly informed. During the control command sending process, the system monitors each command received in real time. If the analysis determines that the command is about to approach or exceed the safety range, the system will promptly send visual and sound dual warning signals to the operation interface. If the command eventually exceeds the safety range, the system will automatically stop the corresponding operation and quickly restore to the safety setting value. At the same time, a notification signal is sent to the receiving terminal of the medical staff so that they can understand the situation in time and make necessary interventions.

[0128] Based on the same inventive concept, an embodiment of the present invention provides a computer storage medium, including a computer program product that can be loaded and executed by a processor to implement the following Figures 1 to 5 Procedures for each step involved in the process.

[0129] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A wireless infusion pump intelligent control system, characterized in that: include: Wireless infusion pump; at least one processor; At least one memory storing a computer program, and at least one processor configured to implement a wireless infusion pump intelligent control method when executing the computer program, the specific steps are as follows: Access to patient diagnostic information or medical records; Extract key features from the patient's diagnostic information or medical records and input them into the trained infusion plan prediction model to output the predicted infusion plan. The key features include but are not limited to symptom description, examination results, and medical history. The infusion plan includes drug selection, dosage, infusion speed, and working mode. Using the predicted infusion plan as the initial infusion plan of the wireless infusion pump, and executing the initial infusion plan of the wireless infusion pump; Real-time monitoring of the patient's real-time vital sign monitoring data during the execution of the initial infusion plan, and selecting key information related to the infusion speed adjustment features, inputting it into a dynamic model for predicting the optimal infusion speed constructed using a particle swarm optimization algorithm, and outputting the predicted optimal infusion speed; The output predicted optimal infusion rate replaces the infusion rate of the initial infusion plan, updates the infusion plan of the wireless infusion pump, and executes the updated infusion plan; The outputs predict the optimal infusion rate including: Analyze whether to obtain the patient's bed posture data; If not, maintain the original setting; If yes, then obtain the patient's bed posture data and real-time vital sign monitoring data, and select key information related to the infusion speed adjustment; The key information of the features related to the adjustment of the infusion speed is input into a dynamic model for predicting the optimal infusion speed constructed by using a particle swarm optimization algorithm, and the predicted optimal infusion speed is output, wherein the dynamic model is trained by a data set integrating posture and vital signs.

2. A wireless infusion pump intelligent control system according to claim 1, characterized in that: The method further includes steps after replacing the infusion rate of the initial infusion plan with the output predicted optimal infusion rate and before updating the infusion plan of the wireless infusion pump, which are specifically as follows: Obtain the real-time temperature of the drug solution and the rate of change of the external environment temperature; Analyze and obtain the real-time temperature deviation based on the target liquid temperature and the real-time liquid temperature; Analyze and determine the infusion speed adjustment range according to the corresponding relationship between the temperature deviation range that the real-time temperature deviation falls into, the change rate range range that the external environment temperature change rate falls into, and the infusion speed adjustment range; According to the infusion speed adjustment range determined by analysis and the output predicted optimal infusion speed, the optimal infusion speed is re-analyzed and determined, and the original optimal infusion speed determined by analysis is replaced.

3. A wireless infusion pump intelligent control system according to claim 2, characterized in that: Analysis to determine the infusion rate adjustment range includes: Analyze whether to obtain the patient's bed posture data; If yes, then analyzing the patient's body position change information based on the acquired bed posture data of the patient; The infusion speed adjustment range is determined based on the correspondence between the patient's body position change information, the temperature deviation range that the real-time temperature deviation falls into, the change rate range that the external environment temperature change rate falls into, and the infusion speed adjustment range.

4. A wireless infusion pump intelligent control system according to claim 3, characterized in that: Also included are steps in parallel with executing the updated infusion protocol, as follows: Obtaining ambiguous instructions from patients; According to the mapping relationship between the patient's fuzzy instruction and the adjustment of the infusion amplitude, analyzing and determining the adjustment of the infusion amplitude matched by the patient's fuzzy instruction; Adjust the infusion speed according to the adjusted infusion amplitude matched by the patient's fuzzy instruction determined through analysis; Analyze whether the patient's ambiguous instructions are received again within the preset time range for adjusting the infusion rate; If no, remain unchanged; If yes, the process jumps to the step of analyzing and determining the step of adjusting the infusion amplitude that matches the patient's fuzzy instruction according to the mapping relationship between the patient's fuzzy instruction and the adjustment of the infusion amplitude.

5. A wireless infusion pump intelligent control system according to claim 4, characterized in that: The adjustment of the infusion range matched by the patient's ambiguous instructions is analyzed and determined, including: Obtain patient position change information and real-time temperature deviation; According to the patient's body position change information, the interval range that the real-time temperature deviation falls into, and the mapping relationship between the fuzzy instructions and the adjusted infusion amplitude, the adjusted infusion amplitude that matches the patient's fuzzy instructions is analyzed and determined.

6. A wireless infusion pump intelligent control system according to claim 5, characterized in that: The method also includes adjusting the infusion amplitude and the infusion speed in parallel with the fuzzy instructions of the patient determined by the analysis, which are specifically as follows: Obtain the number of times the patient uses fuzzy commands within a preset time range and background information, the background information including the patient's body position change information and real-time temperature deviation; If the number of times a patient with the same background information uses fuzzy instructions exceeds the preset number within the preset time range, the average infusion amplitude actually adjusted by the fuzzy instructions for the same background information is obtained, and the average infusion amplitude is used as the infusion adjustment amplitude on the original basis under the corresponding background information.

7. The wireless infusion pump intelligent control system according to claim 6, characterized in that: The method further includes the following steps after real-time monitoring of the patient's real-time vital sign monitoring data during the execution of the initial infusion plan: Extracting sleep-related parameters from the patient's real-time vital sign monitoring data, and analyzing and determining the sleeping result based on a mapping relationship between the result of the sleep-related parameter range that the sleep-related parameter falls into and the sleeping result; If the result of falling asleep is being awake, maintain the original setting; If the result of falling asleep is that the patient is in a sleeping state, the patient's current physiological parameters are obtained and input into a trained model for predicting the patient's drug needs in a sleeping state, and the predicted drug needs in different time periods at night are output; The infusion speed is determined by analyzing the mapping relationship between the predicted drug demand ranges for different time periods at night and the infusion speed.

8. The wireless infusion pump intelligent control system according to claim 7, characterized in that: The method further includes the following steps after obtaining the patient's fuzzy instruction and before analyzing and determining the infusion amplitude adjustment matched by the patient's fuzzy instruction according to the mapping relationship between the patient's fuzzy instruction and the infusion amplitude adjustment, which are as follows: Obtaining the sleep results determined by the analysis; If the result of falling asleep is being awake, continue with the next steps; If the result of falling asleep is in the sleeping state, stop the subsequent steps.

9. A computer storage medium, characterized in that The invention comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement a method for intelligently controlling a wireless infusion pump in an intelligently controlling system for a wireless infusion pump as claimed in any one of claims 1 to 8.

Citation Information

Patent Citations

  • Intelligent infusion control system

    CN105641770A

  • Delivery device with noise reducing component

    CN109260545A

  • Non-contact medical infusion tube liquid outlet amount control method

    CN113855924A

  • Automatic infusion management device

    CN115154756A

  • Infusion system and storage medium

    CN116077765A

Cited By

  • Remote control method of ECMO equipment

    CN119717907A

  • A remote control method for an ECMO device

    CN119717907B

  • Intelligent cleaning control method and system for emulsion pump

    CN119982482A

  • Intelligent infusion monitoring method and system

    CN120154778A