Infusion flow rate control method and system, readable storage medium and computer

Abstract

The invention provides an infusion flow rate control method and system, a readable storage medium and a computer. The method comprises the steps that patient signs and infusion states of a patient on an infusion seat terminal are continuously and periodically collected; a self-adaptive learning algorithm and a preset long-short-term memory network model are introduced to process the physical signs and the infusion state of the patient so as to predict the future infusion flow rate change trend of the patient; calculating the flow velocity according to the future infusion flow velocity change trend of the patient and the physical signs of the patient, and adjusting the infusion flow velocity of the patient in real time according to the calculation result; the pressure of an infusion bottle on the infusion seat terminal is detected in real time, and when the pressure of the infusion bottle is smaller than a preset pressure threshold value, an alarm instruction is sent so that medical staff can conduct processing. Pressure and flow data in the infusion tube are collected in real time, physical sign data of a patient are combined, the infusion flow speed is dynamically adjusted through an intelligent feedback control algorithm, medical staff are reminded to replace an infusion bottle of the patient in time, and the infusion bottle replacing efficiency of the medical staff is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly relates to an infusion flow rate control method, system, readable storage medium and computer. Background Art

[0002] Intelligent healthcare has become an emerging trend in the medical and health industry, aiming to improve the quality and efficiency of medical services through advanced technological means. Traditional infusion systems usually rely on manual settings or simple timing controls, making it difficult to adapt to individual patient differences and changes in the condition, and there are problems such as inaccurate infusion speeds and poor patient experiences.

[0003] During periods of high incidence of viruses such as the epidemic or influenza, the number of infusion patients in hospitals and clinics often surges. In such cases, it is difficult for nurses and doctors to understand the infusion status of each patient in real time, and thus they cannot replace the infusion bottle for the patient in a timely manner. This not only increases the workload of medical staff, but also may affect the treatment effect and satisfaction of patients. Summary of the Invention

[0004] Based on this, the purpose of the present invention is to provide an infusion flow rate control method, system, readable storage medium and computer to at least solve the deficiencies in the above technologies.

[0005] The present invention proposes an infusion flow rate control method, including: Obtaining the position information of the infusion seat terminal, and continuously and periodically collecting the patient signs and infusion status of the patient on the infusion seat terminal; Introducing an adaptive learning algorithm and a preset long short-term memory network model to process the patient signs and the infusion status to predict the future infusion flow rate change trend of the patient; Calculating the flow rate according to the future infusion flow rate change trend of the patient and the patient signs, and adjusting the infusion flow rate of the patient in real time according to the calculation result; Real-time detecting the pressure of the infusion bottle on the infusion seat terminal, and when the pressure of the infusion bottle is less than a preset pressure threshold, sending a corresponding alarm instruction so that medical staff can process according to the alarm instruction.

[0006] Further, the step of obtaining the position information of the infusion seat terminal and continuously and periodically collecting the patient signs and infusion status of the patient on the infusion seat terminal includes: Real-time collecting the pressure data and flow data in the infusion tube through the pressure sensor and flow sensor on the infusion seat terminal, and real-time collecting the heart rate data and blood pressure data of the patient through the heart rate sensor and blood pressure sensor; Preprocess the pressure data, the flow rate data, the heart rate data, and the blood pressure data to obtain the patient's physical signs and infusion status.

[0007] Further, the steps of introducing an adaptive learning algorithm and a preset long short-term memory network model to process the patient's physical signs and the infusion status to predict the future infusion flow rate change trend of the patient include: Extract key features from the patient's physical signs and the infusion status, and select an adaptive learning algorithm to train the long short-term memory network model; Use the historical pressure data and historical flow rate data at past time points in combination with the long short-term memory network model to predict the future pressure change data of the infusion tube on the infusion seat terminal; Calculate the comprehensive physical sign score of the patient at the current time point based on the patient's physical signs, calculate the corresponding score change data according to the comprehensive physical sign score, and predict the future infusion flow rate of the patient based on the score change data and the future pressure change data.

[0008] Further, the calculation formula for the future pressure change data is: ; In the formula, represents the long short-term memory network model, , , , represent historical pressure data, , , , represent historical flow rate data; The calculation formula for the score change data is: ; ; In the formula, represents the comprehensive physical sign score of the patient at the current time point, represents the comprehensive score function based on heart rate and blood pressure, represents the current time point the comprehensive physical sign score of the patient at the previous time point, represents the current time point the heart rate data of the patient, represents the current time point the blood pressure data of the patient; The calculation formula for the future infusion flow rate of the patient is: ; In the formula, and are adjustment coefficients for balancing the influence of pressure changes and patient sign changes on the flow rate, represents the current time point is the infusion flow rate at the previous time point.

[0009] Furthermore, the steps of calculating the flow rate according to the future infusion flow rate change trend of the patient and the patient signs and adjusting the infusion flow rate of the patient in real time according to the calculation result include: Calculating the infusion flow rate of the patient according to the future infusion flow rate change trend of the patient and the patient signs; Constructing a model predictive control MPC infusion flow rate control algorithm, constructing an objective function based on the predicted future infusion flow rate, and rolling and optimizing the infusion flow rate in the future period:

[0010] wherein, and are the weight coefficients of the MPC algorithm, which are dynamically adjusted by an adaptive fuzzy neural network AFNN, represents the current time point is the infusion flow rate.

[0011] Furthermore, the method further includes: Dynamically adjusting the parameters of the MPC algorithm through an adaptive fuzzy neural network, taking the patient's heart rate, patient's blood pressure, and infusion flow rate error as input variables, and dividing them into different fuzzy sets; Taking the weight coefficients and of the objective function of the MPC algorithm and the corresponding adjustment amounts as output variables, and dividing them into fuzzy sets; and Constructing a fuzzy control rule base, specifying fuzzy rules, describing the influence of the patient's heart rate, the patient's blood pressure, and the infusion flow rate error on the weight coefficients and of the objective function of the MPC algorithm, and performing fuzzy inference according to the fuzzy rules to determine the weight coefficients and ; According to the online learning of the neural network, using a 3-layer feedforward neural network, with the input being the triggering intensity of the fuzzy rules and the output being and ; Dynamically adjusting the weight parameters, and dynamically updating the weight coefficients and , to adapt to the personalized needs of patients.

[0012] Furthermore, the method further includes: Obtain the adjustment result of real-time adjustment, and input the adjustment result into the long short-term memory network model, so that the long short-term memory network model is optimized: ; In the formula, represents the optimized long short-term memory network model, represents the long short-term memory network model before optimization, represents the adjustment result, represents the optimization function.

[0013] The present invention also proposes an infusion flow rate control system, including: A data acquisition module, configured to acquire the position information of the infusion seat terminal, and continuously and periodically collect the patient signs and infusion status of the patient on the infusion seat terminal; A flow rate prediction module, configured to introduce an adaptive learning algorithm and a preset long short-term memory network model to process the patient signs and the infusion status, so as to predict the future infusion flow rate change trend of the patient; A flow rate adjustment module, configured to calculate the flow rate according to the future infusion flow rate change trend of the patient and the patient signs, and adjust the infusion flow rate of the patient in real time according to the calculation result; A pressure alarm module, configured to detect the infusion bottle pressure on the infusion seat terminal in real time. When the infusion bottle pressure is less than a preset pressure threshold, send a corresponding alarm instruction, so that medical staff can process according to the alarm instruction.

[0014] Furthermore, the data acquisition module includes: A data acquisition unit, configured to collect the pressure data and flow data in the infusion tube in real time through the pressure sensor and flow sensor on the infusion seat terminal, and collect the heart rate data and blood pressure data of the patient in real time through the heart rate sensor and blood pressure sensor; A data preprocessing unit, configured to preprocess the pressure data, the flow data, the heart rate data, and the blood pressure data to obtain the patient signs and infusion status of the patient.

[0015] Furthermore, the flow rate prediction module includes: A model training unit, configured to extract key features from the patient signs and the infusion status, and select an adaptive learning algorithm to train the long short-term memory network model; A data prediction unit, configured to predict future pressure change data of an infusion tube on the infusion seat terminal by using historical pressure data and historical flow rate data at a plurality of past time points in combination with the long short-term memory network model; A flow rate prediction unit, configured to calculate a comprehensive score of the patient's physical signs at the current time point based on the patient's physical signs, calculate corresponding score change data according to the comprehensive score of the patient's physical signs, and predict the future infusion flow rate of the patient according to the score change data and the future pressure change data.

[0016] Further, the flow rate adjustment module includes: A flow rate calculation unit, configured to calculate the infusion flow rate of the patient according to the future infusion flow rate change trend of the patient and the patient's physical signs; A flow rate adjustment unit, configured to construct a model predictive control infusion flow rate control MPC algorithm, and based on the predicted future infusion flow rate, construct an objective function and roll-optimize the infusion flow rate in a future time period:

[0017] Wherein, and are weight coefficients of the MPC algorithm, dynamically adjusted by an adaptive fuzzy neural network AFNN, represents the current time point of the infusion flow rate.

[0018] Further, the system further includes: A weight coefficient optimization module, configured to dynamically adjust parameters of the MPC algorithm through an adaptive fuzzy neural network, use the patient's heart rate, the patient's blood pressure, and the infusion flow rate error as input variables, and divide them into different fuzzy sets; Take the weight coefficients and of the objective function of the MPC algorithm and the corresponding adjustment amounts as output variables, and divide them into fuzzy sets; and Construct a fuzzy control rule base, specify fuzzy rules, and describe the influence of the patient's heart rate, the patient's blood pressure, and the infusion flow rate error on the weight coefficients and of the objective function of the MPC algorithm, and perform fuzzy inference according to the fuzzy rules to determine the weight coefficients According to neural network online learning, adopt a 3-layer feedforward neural network, with the input being the triggering strength of the fuzzy rules and the output being and ; Dynamically adjust the weight parameters, and dynamically update the weight coefficients according to the fuzzy inference and the neural network output and , to adapt to the personalized needs of patients.

[0019] Furthermore, the system further includes: A model optimization module, configured to obtain the adjustment result of real-time adjustment, and input the adjustment result into the long short-term memory network model, so that the long short-term memory network model is optimized: ; In the formula, represents the optimized long short-term memory network model, represents the long short-term memory network model before optimization, represents the adjustment result, represents the optimization function.

[0020] The present invention also provides a readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned infusion flow rate control method is implemented.

[0021] The present invention also provides a computer, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the above-mentioned infusion flow rate control method is implemented.

[0022] In the infusion flow rate control method, system, readable storage medium and computer of the present invention, by collecting data from the infusion seat terminal, introducing an adaptive learning algorithm and a long short-term memory network model to process the patient signs and infusion status collected by the infusion seat terminal, predicting the future change trend of the patient's infusion flow rate, calculating the flow rate using the patient signs and the future change trend of the infusion flow rate, and adjusting the patient's infusion flow rate in real time according to the calculation result, realizing real-time detection and monitoring of the patient signs and infusion status, dynamically adjusting the infusion flow rate by collecting the pressure and flow data in the infusion tube in real time, combining with the patient sign data, and using the intelligent feedback control algorithm, timely reminding the medical staff to replace the patient's infusion bottle, effectively improving the efficiency of the medical staff in replacing the infusion bottle, significantly improving the safety and effectiveness of the infusion process, and being applicable to various infusion management scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a flowchart of the infusion flow rate control method in the first embodiment of the present invention; Figure 2 is Figure 1 a detailed flowchart of step S101 in Figure 3 is a structural block diagram of the alarm management system in the first embodiment of the present invention; Figure 4 For Figure 1 the detailed flowchart of step S102 in Figure 5 the structural block diagram of the infusion flow rate control system in the second embodiment of the present invention; Figure 6 the structural block diagram of the computer in the third embodiment of the present invention.

[0024] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. Specific Embodiments

[0025] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0027] Embodiment 1 Please refer to Figure 1 , which shows the infusion flow rate control method in the first embodiment of the present invention. The method specifically includes steps S101 to S104: S101, obtaining the position information of the infusion seat terminal, and continuously and periodically collecting the patient signs and infusion status of the patient on the infusion seat terminal; Further, please refer to Figure 2 , the step S101 specifically includes steps S1011~S1012: S1011, collecting the pressure data and flow data in the infusion tube in real time through the pressure sensor and flow sensor on the infusion seat terminal, and collecting the heart rate data and blood pressure data of the patient in real time through the heart rate sensor and blood pressure sensor; S1012, preprocessing the pressure data, the flow data, the heart rate data, and the blood pressure data to obtain the patient signs and infusion status of the patient.

[0028] In this embodiment, the infusion flow rate control method is applied to the alarm management system. Please refer to Figure 3, the alarm management system includes an infusion seat terminal, a medical staff terminal, and a background management control system. Among them, the infusion seat terminal detects the patient and the infusion status in real time, controls the infusion flow rate, and transmits the detected relevant data to the medical staff terminal and the background management control system through a communication module; the medical staff terminal is an intelligent terminal device used to manage and monitor the infusion seat terminal, capable of managing and processing the information and data of the infusion seat terminal; the background management control system is a background management control interface used to manage and monitor the medical staff terminal, including a communication module, capable of managing and processing various information and data in the infusion seat terminal, and having blockchain technology to ensure the security and immutability of the data.

[0029] In this embodiment, the infusion seat terminal refers to an intelligent terminal device installed on the patient's seat, including a patient detection module, an infusion pipeline, and an intelligent control module, which can accurately detect the infusion status and interact with the medical staff terminal. Among them, the patient detection module includes a heart rate sensor and a blood pressure sensor, which are responsible for real-time monitoring of the patient's vital signs data; the infusion pipeline includes an infusion bottle, an infusion tube, a pressure sensor, and a flow sensor, which are responsible for real-time monitoring of the status of the infusion process; the intelligent control module includes a data acquisition unit, a processing unit, and a control unit, which are responsible for real-time acquisition and processing of the pressure and flow data in the infusion tube, and controlling the infusion flow rate according to the processing results.

[0030] The above-mentioned medical staff terminal refers to an intelligent terminal device used in a hospital to manage and monitor the infusion seat terminal, capable of managing and processing the information and data of the infusion seat terminal, providing convenient management and monitoring functions for medical staff, including adding and deleting patient information, viewing the location information of the infusion seat, monitoring the status of the infusion bottle, and receiving warnings from the infusion seat terminal.

[0031] The above-mentioned background management control system is a background management control interface used to manage and monitor the medical staff terminal, including a communication module, capable of managing and processing various information and data in the infusion seat terminal, and having blockchain technology to ensure the security and immutability of the data.

[0032] In specific implementation, when a patient uses the infusion seat terminal, obtain the location information of the infusion seat terminal, and real-time collect the pressure data in the infusion tube through the pressure sensor and the flow sensor and the flow data , and real-time collect the patient's heart rate data and blood pressure data through the heart rate sensor and the blood pressure sensor; preprocess the data, perform denoising and outlier processing on the collected data to ensure the accuracy and reliability of the data; S102. Introduce an adaptive learning algorithm and a preset long short-term memory network model to process the patient's physical signs and the infusion status, so as to predict the future change trend of the patient's infusion flow rate; Further, please refer to Figure 4 , the step S102 specifically includes steps S1021 to S1023: S1021. Extract key features from the patient's physical signs and the infusion status, and select an adaptive learning algorithm to train the long short-term memory network model; S1022. Use the historical pressure data and historical flow data at past time points to combine with the long short-term memory network model to predict the future pressure change data of the infusion tube at the infusion seat terminal; S1023. Calculate the comprehensive physical sign score of the patient at the current time point, calculate the corresponding score change data according to the comprehensive physical sign score of the patient, and predict the future infusion flow rate of the patient according to the score change data and the future pressure change data.

[0033] In specific implementation, extract key features from the above preprocessed data, select an adaptive learning algorithm, and train the long short-term memory (LSTM) network model; use the pressure and flow historical data at the past n time points to combine with the long short-term memory (LSTM) to predict the future pressure change , and its pressure change prediction formula is: ; In the formula, represents the long short-term memory network model, , , , represent historical pressure data, , , , represent historical flow data; Further, calculate the comprehensive physical sign score of the patient at the current time point , and calculate the change in the comprehensive physical sign change score of the patient : ; ; In the formula, represents the comprehensive physical sign score of the patient at the current time point, represents the comprehensive score function according to heart rate and blood pressure, represents the current time point the comprehensive physical sign score of the patient at the previous time point; According to the predicted pressure change and the change in patient signs , calculate the future infusion flow rate of the patient : ; In the formula, and are adjustment coefficients for balancing the influence of pressure change and patient sign change on the flow rate respectively, represents the current time point The infusion flow rate at the previous time point.

[0034] S103. Calculate the flow rate according to the future infusion flow rate change trend of the patient and the patient signs, and adjust the infusion flow rate of the patient in real time according to the calculation result; Further, the step S103 specifically includes steps S1031 to S1032: S1031. Calculate the infusion flow rate of the patient according to the future infusion flow rate change trend of the patient and the patient signs; S1032. Construct a model predictive control infusion flow rate control MPC algorithm, and based on the predicted future infusion flow rate, construct an objective function and roll and optimize the infusion flow rate in the future time period:

[0035] Among them, and are the weight coefficients of the MPC algorithm, dynamically adjusted by the adaptive fuzzy neural network AFNN, represents the current time point The infusion flow rate of.

[0036] In specific implementation, The new flow rate calculated according to the prediction result and the patient sign data , adjust the infusion flow rate in real time through the MPC algorithm :

[0037] Solve the optimal flow rate sequence through quadratic programming and execute the first-step control quantity; Dynamically adjust the MPC parameters through the adaptive fuzzy neural network (AFNN); Take the patient's heart rate (HR), patient's blood pressure (BP), and the infusion flow rate error as input variables and divide them into different fuzzy sets; Take the adjustment amount of the MPC objective function weight coefficient flow rate weight and the sign weight of and As an output variable, divide it into fuzzy sets; Construct a fuzzy control rule base, formulate fuzzy rules, and describe the influence of the patient's heart rate, the patient's blood pressure, and the infusion flow rate error on the weight coefficients of the MPC objective function and ; In this embodiment, first, divide the patient's heart rate (HR) and the patient's blood pressure (BP) into low (L), normal (N), and high (H); divide the infusion flow rate error (ΔF) into negative large (NB), negative small (NS), zero (ZO), positive small (PS), and positive large (PB). The output variables and are divided into large reduction (LB), small reduction (SB), maintenance (M), small increase (SI), and large increase (LI). Then, set the fuzzy control rules: Rule 1: If HR = H and BP = H and = NB, then = LB, = LI; Rule 2: If HR = H and BP = N and = NS, then = SB, = SI; Rule 3: If HR = N and BP = H and = ZO, then = SB, = SI; Rule 4: If = PB and BP = LR, then = LI, = LB; Rule 5: If HR = L and BP = L and = ZO, then = M, = M; Rule 6: If BP = H and = NB, then = LB, = LI; Rule 7: If HR = N and BP = N and = ZO, then = M, = M; Rule 8: If = PB and HR = N, then = LI, = SB; Rule 9: If HR = H and = PB, then = LB, = LI; Rule 10: If BP = H and = NS, then = SB, = LI; Rule 11: If = NB and HR = N, then = LB, = SI; Rule 12: If = ZO and BP = H, then = LB, = LI; Rule 13: If HR = L and BP = H, then = SB, = LI; Rule 14: If = PS and HR = H, then = SB, = LI; Rule 15: If = NB and BP = N, then = SB, = SI; Rule 16: If HR = N and BP = H and = PS, then = SB, = SI; Rule 17: If = PB and BP = N, then = LI, = LB; Rule 18: If BP = L and = ZO, then = SI, = SB; Rule 19: If HR = H and B = L, then = LB, = LI; Rule 20: If = NS and HR = L, then = SB, = SI; Rule 21: If BP = L and = PB, then = SI, = LB; Rule 22: If = PS and BP = N, then = SI, = SB; Rule 23: If HR = H and = ZO, then = LB, = LI; Rule 24: If = NS and BP = H, then = SB, = LI; Rule 25: If HR = N and BP = N and = PS and BP = H, then = SI, = SB; Fuzzy inference is carried out according to the fuzzy rules, and the weighted average method is used for defuzzification output to determine the weight coefficients of the MPC objective function. and 。

[0038] According to the online learning of the neural network, a three-layer feedforward neural network is adopted, with the input being the triggering intensity of fuzzy rules and the output being and ; Dynamically adjust the weight parameters, and update the weight coefficients dynamically according to the fuzzy inference and the output of the neural network and , to adapt to the personalized needs of patients. S104, real-time detection of the infusion bottle pressure on the infusion seat terminal, when the infusion bottle pressure is less than the preset pressure threshold, send a corresponding alarm instruction, so that the medical staff can process according to the alarm instruction.

[0039] In specific implementation, the pressure of the infusion bottle is collected in real time through a pressure sensor , when the intelligent control module detects that the infusion bottle pressure is less than the set threshold , the control unit will immediately trigger the corresponding alarm device, the infusion seat terminal emits a red light warning and vibrates, and through voice prompts the medical staff that the infusion bottle needs to be replaced, and sends the instruction to the medical terminal and the background management control system through the wireless communication module, allowing the remote doctor to conduct real-time monitoring and guidance through the video call function.

[0040] In some alternative embodiments, the method further includes: Obtain the adjustment result of the real-time adjustment, and input the adjustment result into the long short-term memory network model, so that the long short-term memory network model is optimized: ; In the formula, represents the optimized long short-term memory network model, represents the long short-term memory network model before optimization, represents the adjustment result, represents the optimization function.

[0041] In specific implementation, record the actual effect of each adjustment, and according to the actual adjustment effect, input the feedback data into the long short-term memory (LSTM) network model, continuously optimize the long short-term memory (LSTM) network model, and perform model optimization.

[0042] In summary, the infusion flow rate control method in the above embodiments of the present invention collects data from the infusion seat terminal, and introduces an adaptive learning algorithm and a long short-term memory network model to process the patient's physical signs and infusion status collected by the infusion seat terminal, so as to predict the future change trend of the patient's infusion flow rate. The infusion flow rate is calculated using the patient's physical signs and the future change trend of the infusion flow rate, and the patient's infusion flow rate is adjusted in real time according to the calculation result, realizing real-time detection and monitoring of the patient's physical signs and infusion status. By collecting the pressure and flow data in the infusion tube in real time, combining with the patient's physical sign data, and dynamically adjusting the infusion flow rate using an intelligent feedback control algorithm, it timely reminds the medical staff to replace the patient's infusion bottle, effectively improving the efficiency of the medical staff in replacing the infusion bottle, significantly improving the safety and effectiveness of the infusion process, and being applicable to various infusion management scenarios.

[0043] Embodiment 2 On the other hand, the present invention also proposes an infusion flow rate control system. Please refer to Figure 5 , which shows the infusion flow rate control system in the second embodiment of the present invention. The system includes: A data acquisition module 11, configured to acquire the position information of the infusion seat terminal, and continuously and periodically collect the patient's physical signs and infusion status of the patient on the infusion seat terminal; Further, the data acquisition module 11 includes: A data acquisition unit, configured to collect the pressure data and flow data in the infusion tube in real time through the pressure sensor and flow sensor on the infusion seat terminal, and collect the heart rate data and blood pressure data of the patient in real time through the heart rate sensor and blood pressure sensor; A data preprocessing unit, configured to preprocess the pressure data, the flow data, the heart rate data, and the blood pressure data to obtain the patient's physical signs and infusion status.

[0044] A flow rate prediction module 12, configured to introduce an adaptive learning algorithm and a preset long short-term memory network model to process the patient's physical signs and the infusion status, so as to predict the future change trend of the patient's infusion flow rate; Further, the flow rate prediction module 12 includes: A model training unit, configured to extract key features from the patient's physical signs and the infusion status, and select an adaptive learning algorithm to train the long short-term memory network model; A data prediction unit, configured to use the historical pressure data and historical flow data at past time points in combination with the long short-term memory network model to predict the future pressure change data of the infusion tube on the infusion seat terminal; A flow rate prediction unit, configured to calculate a comprehensive score of the patient's physical signs at the current time point based on the patient's physical signs, calculate corresponding score change data according to the comprehensive score of the patient's physical signs, and predict the future infusion flow rate of the patient according to the score change data and the future pressure change data.

[0045] A flow rate adjustment module 13, configured to calculate a flow rate according to the future infusion flow rate change trend of the patient and the patient's physical signs, and adjust the infusion flow rate of the patient in real time according to the calculation result; Further, the flow rate adjustment module 13 includes: A flow rate calculation unit, configured to calculate the infusion flow rate of the patient according to the future infusion flow rate change trend of the patient and the patient's physical signs; A flow rate adjustment unit, configured to construct a model predictive control (MPC) algorithm for controlling the infusion flow rate, construct an objective function based on the predicted future infusion flow rate, and roll-optimize the infusion flow rate for a future period:

[0046] Wherein, and are the weight coefficients of the MPC algorithm, dynamically adjusted by an adaptive fuzzy neural network (AFNN), represents the current time point of the infusion flow rate.

[0047] A pressure alarm module 14, configured to detect the pressure of the infusion bottle on the infusion seat terminal in real time, and when the pressure of the infusion bottle is less than a preset pressure threshold, send a corresponding alarm instruction so that medical staff can process according to the alarm instruction.

[0048] Further, the system further includes: A weight coefficient optimization module, configured to dynamically adjust the parameters of the MPC algorithm through an adaptive fuzzy neural network, use the patient's heart rate, the patient's blood pressure, and the infusion flow rate error as input variables, and divide them into different fuzzy sets; Take the weight coefficients and of the objective function of the MPC algorithm and the corresponding adjustment amounts as output variables, and divide them into fuzzy sets; and Construct a fuzzy control rule base, specify fuzzy rules, describe the influence of the patient's heart rate, the patient's blood pressure, and the infusion flow rate error on the weight coefficients and ; According to the online learning of the neural network, a three-layer feedforward neural network is adopted, with the input being the triggering intensity of fuzzy rules and the output being and ; Dynamically adjust the weight parameters, and dynamically update the weight coefficients according to the fuzzy inference and the output of the neural network and to adapt to the personalized needs of patients.

[0049] Furthermore, the system further includes: A model optimization module, configured to obtain the adjustment result of real-time adjustment and input the adjustment result into the long short-term memory network model, so as to optimize the long short-term memory network model: ; In the formula, represents the optimized long short-term memory network model, represents the long short-term memory network model before optimization, represents the adjustment result, represents the optimization function.

[0050] The functions or operation steps implemented when the above-mentioned modules and units are executed are substantially the same as those in the above method embodiments, and will not be elaborated here.

[0051] The infusion flow rate control system provided by the embodiments of the present invention has the same implementation principle and technical effects as those in the foregoing method embodiments. For a brief description, for the parts not mentioned in the system embodiments, reference can be made to the corresponding content in the foregoing method embodiments.

[0052] Embodiment III The present invention also proposes a computer. Please refer to Figure 6 , which shows the computer in the third embodiment of the present invention, including a memory 10, a processor 20, and a computer program 30 stored on the memory 10 and executable on the processor 20. When the processor 20 executes the computer program 30, the above-mentioned infusion flow rate control method is implemented.

[0053] Among them, the memory 10 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disc, etc. The memory 10 can be an internal storage unit of a computer in some embodiments, such as the hard disk of the computer. The memory 10 can also be an external storage device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Further, the memory 10 can also include both an internal storage unit of a computer and an external storage device. The memory 10 can be used not only to store application software installed in the computer and various types of data, but also to temporarily store data that has been output or will be output.

[0054] Among them, the processor 20 can be an Electronic Control Unit (ECU, also known as a vehicle computer), a Central Processing Unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments, and is used to run the program code stored in the memory 10 or process data, such as executing an access restriction program, etc.

[0055] It should be noted that Figure 6 The structure shown does not constitute a limitation on the computer. In other embodiments, the computer may include fewer or more components than shown in the figure, or combine certain components, or have a different component arrangement.

[0056] An embodiment of the present invention also proposes a readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the infusion flow rate control method as described above.

[0057] Those skilled in the art can understand that the logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing a logical function, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or used in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0058] More specific examples (a non-exhaustive list) of computer-readable media include the following: electrical connections (electronic devices) having one or more wirings, portable computer disk cartridges (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber devices, and portable compact disc read-only memory (CDROM). Additionally, the computer-readable media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0059] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having suitable combinational logic gate circuits, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0060] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as falling within the scope described in this specification.

[0061] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for controlling an infusion flow rate, characterized in that: include: Acquire the location information of the infusion chair terminal, and continuously and periodically collect the patient's vital signs and infusion status of the patient on the infusion chair terminal; Introducing an adaptive learning algorithm and a preset long short-term memory network model to process the patient's vital signs and the infusion status to predict the future infusion flow rate change trend of the patient; Calculating the flow rate according to the future infusion flow rate change trend of the patient and the patient's physical signs, and adjusting the infusion flow rate of the patient in real time according to the calculation result; The pressure of the infusion bottle on the infusion chair terminal is detected in real time. When the pressure of the infusion bottle is less than a preset pressure threshold, a corresponding alarm instruction is sent so that medical staff can handle it according to the alarm instruction.

2. The infusion flow rate control method according to claim 1, characterized in that: The steps of obtaining the position information of the infusion chair terminal and continuously and periodically collecting the patient's vital signs and infusion status of the patient on the infusion chair terminal include: The pressure sensor and flow sensor on the infusion chair terminal are used to collect the pressure data and flow data in the infusion tube in real time, and the heart rate sensor and blood pressure sensor are used to collect the heart rate data and blood pressure data of the patient in real time; The pressure data, the flow data, the heart rate data and the blood pressure data are preprocessed to obtain the patient's vital signs and infusion status.

3. The infusion flow rate control method according to claim 1, characterized in that: The steps of introducing an adaptive learning algorithm and a preset long short-term memory network model to process the patient's physical signs and the infusion status to predict the future infusion flow rate change trend of the patient include: Extracting key features from the patient's vital signs and the infusion status, and selecting an adaptive learning algorithm to train the long short-term memory network model; Using the historical pressure data and historical flow data at several past time points in combination with the long short-term memory network model, the future pressure change data of the infusion tube on the infusion chair terminal is predicted; A comprehensive score of the patient's vital signs at the current time point is calculated based on the patient's vital signs, and corresponding score change data is calculated based on the comprehensive score of the patient's vital signs, and the patient's future infusion flow rate is predicted based on the score change data and the future pressure change data.

4. The infusion flow rate control method according to claim 3, characterized in that: The calculation formula for the future pressure change data is: ； In the formula, represents the long short-term memory network model, , , , Represents historical pressure data, , , , Represents historical traffic data; The calculation formula of the score change data is: ； ； In the formula, Represents the comprehensive score of the patient's physical signs at the current time point, represents the comprehensive scoring function based on heart rate and blood pressure, Indicates the current time point The comprehensive score of the patient's physical signs at the previous time point, Indicates the current time point Patient heart rate data, Indicates the current time point Blood pressure data of patients; The calculation formula for the patient's future infusion flow rate is: ； In the formula, and are adjustment coefficients used to balance the effects of pressure changes and patient vital signs changes on flow rate, Indicates the current time point Infusion flow rate at the previous time point.

5. The infusion flow rate control method according to claim 4, characterized in that: The steps of calculating the flow rate according to the future infusion flow rate change trend of the patient and the patient's physical signs, and adjusting the infusion flow rate of the patient in real time according to the calculation result include: Calculating the infusion flow rate of the patient according to the future infusion flow rate change trend of the patient and the patient's physical signs; Construct a model predictive control infusion flow rate control MPC algorithm, build an objective function based on the predicted future infusion flow rate, and optimize the infusion flow rate in the future period in a rolling manner: in, and is the weight coefficient of the MPC algorithm, which is dynamically adjusted by the adaptive fuzzy neural network AFNN. Indicates the current time point infusion flow rate.

6. The infusion flow rate control method according to claim 5, characterized in that: The method further comprises: Dynamically adjusting the parameters of the MPC algorithm through an adaptive fuzzy neural network, taking the patient's heart rate and the patient's blood pressure as well as the infusion flow rate error as input variables, and dividing them into different fuzzy sets; The weight coefficient of the objective function of the MPC algorithm is and The corresponding adjustment amount and As output variables, they are divided into fuzzy sets; Construct a fuzzy control rule base, specify fuzzy rules, describe the weight coefficients of the patient's heart rate, the patient's blood pressure and the infusion flow rate error on the objective function of the MPC algorithm and The influence of fuzzy reasoning is performed according to the fuzzy rules to determine the weight coefficient of the objective function of the MPC algorithm and ; According to the neural network online learning, a 3-layer feedforward neural network is used, the input is the fuzzy rule trigger intensity, and the output is and ; Dynamically adjust weight parameters and dynamically update weight coefficients based on fuzzy reasoning and neural network output and , adapted to the individual needs of patients.

7. The infusion flow rate control method according to claim 1, characterized in that: The method further comprises: Acquire the adjustment result of the real-time adjustment, and input it into the long short-term memory network model according to the adjustment result, so as to optimize the long short-term memory network model: ； In the formula, represents the optimized long short-term memory network model, represents the long short-term memory network model before optimization, Indicates the adjustment result. Represents an optimization function.

8. An infusion flow rate control system, characterized in that: include: A data acquisition module, used to acquire the position information of the infusion chair terminal, and continuously and periodically collect the patient's vital signs and infusion status of the patient on the infusion chair terminal; A flow rate prediction module, used to introduce an adaptive learning algorithm and a preset long short-term memory network model to process the patient's vital signs and the infusion status, so as to predict the future infusion flow rate change trend of the patient; A flow rate adjustment module, used to calculate the flow rate according to the future infusion flow rate change trend of the patient and the patient's physical signs, and adjust the infusion flow rate of the patient in real time according to the calculation result; The pressure alarm module is used to detect the infusion bottle pressure on the infusion chair terminal in real time. When the infusion bottle pressure is less than a preset pressure threshold, a corresponding alarm instruction is sent so that medical staff can handle it according to the alarm instruction.

9. A readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the infusion flow rate control method as described in any one of claims 1 to 7 is implemented.

10. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the infusion flow rate control method as described in any one of claims 1 to 7 is implemented.

Images