An infusion management and control system based on internet of things

Abstract

The application discloses an infusion management and control system based on Internet of Things, and relates to the technical field of automatic control, which comprises the following steps: collecting infusion monitoring data arranged on K beds, obtaining infusion flow rate, medicine liquid level and medicine liquid temperature sequence; jointly analyzing the flow rate and the liquid level to obtain medicine liquid delivery abnormality result; performing interactive correlation analysis on the temperature sequence to obtain temperature abnormality result; triggering an infusion valve, a medicine liquid bag replacement device and a heating device according to the analysis result to realize intelligent control of the infusion process. The application solves the technical problems of low infusion process control reliability and low actual infusion situation fitting degree in the prior art, and achieves the technical effects of improving the infusion process control reliability and the actual infusion situation fitting degree.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automation control, and particularly relates to an infusion management and control system based on Internet of Things. BACKGROUND

[0002] Traditional infusion management usually relies on medical staff to manually monitor the infusion process, such as observing key parameters such as infusion flow rate, liquid volume and drug temperature. However, this approach has the problems of low monitoring efficiency and slow response in the process of multiple beds infusion or long time infusion, especially when the infusion flow rate fluctuates, the drug is close to depletion or the drug temperature changes, it may not be able to take effective measures in time, resulting in interruption of infusion or affecting the quality of infusion. In addition, the existing technology has limited ability in multi-parameter linkage monitoring, usually only single parameter can be monitored independently, lack of comprehensive analysis and collaborative processing ability, difficult to meet the demand of precise control of infusion process. With the increase of the complexity of medical system, the above problems reflect the shortcomings of traditional infusion management in safety and accuracy.

[0003] At present, in the related art, there is the technical problem of low reliability of infusion process control and low fitting degree with actual infusion situation. SUMMARY

[0004] The present application provides an infusion management and control system based on Internet of Things, which solves the technical problem of low reliability of infusion process control and low fitting degree with actual infusion situation in the prior art.

[0005] The present application provides an infusion management and control system based on Internet of Things, comprising:

[0006] The monitoring data collection module is used for collecting monitoring data of K infusion monitoring devices arranged at K beds in a preset monitoring window, and obtaining K infusion monitoring data set sequences, wherein K is an integer greater than or equal to 1; the liquid level sequence acquisition module is used for searching the K infusion monitoring data set sequences with infusion flow rates and medicine bag liquid levels as indexes, and obtaining K infusion flow rate sequences and K medicine bag liquid level sequences; the abnormality analysis module is used for jointly analyzing the K infusion flow rate sequences and the K medicine bag liquid level sequences, and obtaining K medicine liquid delivery abnormality analysis results; the temperature sequence acquisition module is used for searching the K infusion monitoring data set sequences with medicine liquid temperatures as indexes, and obtaining K medicine liquid temperature sequences; the interactive correlation abnormality analysis module is used for performing interactive correlation abnormality analysis on the K medicine liquid temperature sequences, and obtaining K medicine liquid temperature abnormality analysis results; and the infusion control module is used for triggering infusion valves, medicine bag replacement devices and heating devices of the K beds according to the K medicine liquid delivery abnormality analysis results and the K medicine liquid temperature abnormality analysis results, and completing infusion control of the K beds.

[0007] The infusion management and control system based on the Internet of Things provided in the application first collects infusion monitoring data arranged at K beds, obtains infusion flow rate sequences, medicine bag liquid level sequences and medicine liquid temperature sequences, jointly analyzes flow rates and liquid levels, obtains medicine liquid delivery abnormality results, performs interactive correlation analysis on temperature sequences, obtains temperature abnormality results, triggers infusion valves, medicine bag replacement devices and heating devices according to analysis results, realizes intelligent control of the infusion process, and achieves the technical effects of improving the reliability of infusion process control and the fitting degree of actual infusion conditions. BRIEF DESCRIPTION OF DRAWINGS

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. The flowcharts are used to illustrate the operations performed by the system according to the embodiments of the present application in the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, according to the needs, various steps can be processed in reverse order or at the same time. At the same time, other operations can be added to these processes, or one or more steps of the operation can be removed from these processes.

[0009] Figure 1 A structural schematic diagram of the infusion management and control system based on the Internet of Things provided in the embodiments of the present application.

[0010] Figure 2A preset liquid level requirement judgment unit structure schematic diagram of a transfusion management and control system based on Internet of Things is provided for an embodiment of the present application.

[0011] Legend: monitoring data acquisition module 10, liquid level sequence acquisition module 20, abnormality analysis module 30, temperature sequence acquisition module 40, interactive correlation abnormality analysis module 50, transfusion control module 60. DETAILED DESCRIPTION

[0012] The above description is only a summary of the technical solutions of the present application. In order to make the technical means of the present application more clear, the present application can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0013] In order to make the purposes, technical solutions and advantages of the present application more clear, the following will combine the drawings to further describe the present application in detail, and the described embodiments should not be regarded as limiting the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0014] In the following description, "some embodiments" are related to a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subset of all possible embodiments, and can be combined with each other without conflict, the term "first\second" involved only distinguishes similar objects, and does not represent the specific order of the objects. The terms "include" and "have" and any variations, are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0015] The present application provides a transfusion management and control system based on Internet of Things, as shown in Figure 1 The system comprises:

[0016] The monitoring data collection module 10 is configured to collect monitoring data of K infusion monitoring devices arranged at K beds within a preset monitoring window, and obtain K sets of infusion monitoring data sequences, where K is an integer greater than or equal to 1. Specifically, the monitoring data collection module 10 is configured to obtain real-time data within the preset monitoring window from the infusion monitoring devices arranged at the K beds, and generate K sets of infusion monitoring data sequences. Each infusion monitoring device is equipped with a temperature sensor, a liquid level sensor and a flow rate sensor, which are configured to collect the temperature of the medicine liquid, the liquid level of the medicine bag and the current infusion speed, respectively. The monitoring window is set as a fixed time interval (for example, every 5 seconds), to ensure the continuity and stability of data collection. The collected data is preliminarily screened before uploading, and abnormal values and noises are removed, for example, when the temperature data exceeds the reasonable range (below room temperature or above the upper limit of safety) or the liquid level sensor detects unreasonable sharp fluctuations, re-collection will be triggered to ensure the accuracy of the data. Finally, the collected data sequences are uploaded to the central server through the Internet of Things interface, to provide reliable basic data for subsequent anomaly analysis and intelligent control.

[0017] The liquid level sequence acquisition module 20 is configured to search the K sets of infusion monitoring data sequences with the infusion flow rate and the liquid level of the medicine bag as indexes, and obtain K infusion flow rate sequences and K medicine bag liquid level sequences. Specifically, the liquid level sequence acquisition module 20 extracts the flow rate and liquid level related data from the infusion monitoring data arranged at the K beds by indexing the monitoring data sequence, with the infusion flow rate and the liquid level of the medicine bag as conditions. Specifically, the module retrieves the flow rate and liquid level fields from the monitoring data set of each bed, aligns the data according to the time stamp, and generates the corresponding infusion flow rate sequence and medicine bag liquid level sequence. For example, within the monitoring window, if the flow rate data of a certain bed contains 60 records and the liquid level data contains 58 records, the module unifies the time stamp by interpolation or truncation to ensure that the flow rate and liquid level sequences correspond one by one. At the same time, the module detects and repairs the extracted data, for example, removes the jump value of the liquid level or the sudden zero value in the flow rate, to ensure the data integrity and quality. Finally, the module outputs the structured K infusion flow rate sequences and K medicine bag liquid level sequences, to provide high-quality time series data input for the anomaly analysis module.

[0018] The abnormality analysis module 30 is configured to jointly analyze the K infusion flow rate sequences and the K drug solution bag liquid level sequences to obtain K drug solution delivery abnormality analysis results. Specifically, the abnormality analysis module 30 identifies abnormal conditions in the infusion process through joint analysis of the infusion flow rate sequences and the drug solution bag liquid level sequences and outputs analysis results. First, the module extracts the last data point of each liquid level sequence and determines whether the minimum liquid level requirement is met. If not, it is directly determined that the drug solution delivery is abnormal and the drug solution bag is replaced. If the requirement is met, the adjacent difference of the liquid level sequence is calculated to generate a liquid level difference sequence, and the flow rate sequence and the liquid level difference sequence are traversed to calculate the fluctuation variance to quantify the stability of the flow rate and the liquid level. On this basis, the module performs stable data filtering on the flow rate and the liquid level difference, and extracts data points with a fluctuation less than a threshold value as stable values. Finally, the module inputs the flow rate fluctuation variance, the liquid level difference fluctuation variance and the stable values thereof into a stability abnormality recognizer for comprehensive analysis, and outputs the abnormality type corresponding to the bed, including normal, liquid level abnormality, flow rate abnormality or joint abnormality, etc., to provide accurate decision basis for subsequent intelligent control. For example, when the liquid level difference fluctuation of a certain bed exceeds the standard and the flow rate fluctuation variance is abnormal, the module will mark it as a joint abnormality and suggest comprehensive treatment.

[0019] In a possible implementation, the abnormality analysis module 30 further includes a preset liquid level requirement judgment unit configured to extract K last drug solution bag liquid levels located at the last positions in the K drug solution bag liquid level sequences respectively, determine whether the K last drug solution bag liquid levels meet a preset liquid level requirement, and if so, perform stability abnormality recognition on the K infusion flow rate sequences and the K drug solution bag liquid level sequences to obtain the K drug solution delivery abnormality analysis results. Specifically, the preset liquid level requirement judgment unit compares the last data point of each bed in the drug solution bag liquid level sequence (i.e., the last drug solution bag liquid level) with a preset liquid level threshold value to determine whether the drug solution bag needs to be replaced. If the last liquid level value is higher than or equal to the liquid level threshold value, the system considers that the liquid level is sufficient and the drug solution bag does not need to be replaced, and proceeds to the subsequent stability abnormality recognition step to further analyze the dynamic stability of the infusion flow rate and the liquid level. For example, when the liquid level value is 40 mm and the threshold value is 20 mm, the system enters the stability analysis process. If the last liquid level value is lower than the liquid level threshold value, the system directly determines that the liquid level is abnormal, does not need to perform stability analysis, immediately outputs the abnormality analysis result and triggers the drug solution bag replacement operation. For example, when the liquid level value is only 10 mm and lower than 20 mm, the system will mark the liquid level as abnormal and issue a replacement instruction. This module ensures the monitoring accuracy of the drug solution bag liquid level while effectively avoiding infusion interruption and improving the reliability of the infusion process.

[0020] an abnormality analysis result output unit configured to output the replacement of the medicine bag as the K liquid delivery abnormality analysis result if the answer is no. Specifically, the function of the abnormality analysis result output unit is to output the replacement of the medicine bag as the abnormality analysis result and output to the control module if the liquid level of the medicine bag does not meet the preset requirement. Specifically, the unit traverses the liquid level judgment result of each bed and identifies the bed whose liquid level is below the threshold. For example, if the last liquid level value of bed 1 is 10 mm, which is lower than the preset threshold of 20 mm, the system will mark it as insufficient liquid level and output the abnormal result of "the medicine bag needs to be replaced". At the same time, for the bed whose liquid level meets the requirement, no abnormal result is generated. Finally, the system outputs the abnormality analysis result of all beds, including the bed number and the corresponding abnormal type, such as {bed 1: replace the medicine bag, bed 3: replace the medicine bag}. Through this module, the system can respond to the liquid level abnormality in real time, accurately locate the problem bed, ensure the timeliness and accuracy of the subsequent medicine bag replacement operation, and effectively avoid the interruption of infusion.

[0021] In a possible implementation, as shown in Figure 2 the preset liquid level requirement judgment unit further includes a liquid level difference calculation sub-unit configured to calculate the liquid level difference of adjacent two medicine bag liquid levels in the K medicine bag liquid level sequence respectively, and obtain K adjacent medicine bag liquid level difference sequences. Specifically, the function of the liquid level difference calculation sub-unit is to calculate the difference between adjacent data points of the medicine bag liquid level sequence of each bed one by one, generate a liquid level difference sequence, and analyze the dynamic change characteristics of the medicine bag liquid level. In the specific implementation process, the liquid level sequence of each bed (for example, [100mm, 90mm, 80mm, 70mm]) is input, the difference between adjacent data points is calculated according to the formula and the difference sequence (such as [10mm, 10mm, 10mm]) is generated. The difference sequence reflects the change rate and trend of the liquid level, and the consistency of the difference indicates that the liquid level drop speed is stable, and the fluctuation of the difference may indicate an abnormal situation. For example, if the difference sequence calculated from the sequence [100mm, 80mm, 70mm, 50mm] is [20mm, 10mm, 20mm], it shows that the liquid level drop rate is uneven, which may need to be further analyzed. Through the accurate calculation of the liquid level difference, the system can dynamically capture the liquid level change and provide data support for the subsequent fluctuation variance calculation and abnormality identification.

[0022] a fluctuation variance calculation subunit configured to traverse the K infusion flow rate sequences and the K adjacent medicine liquid level difference sequences to calculate fluctuation variances, and determine K infusion flow rate fluctuation variances and K adjacent medicine liquid level difference fluctuation variances. Specifically, the fluctuation variance calculation subunit calculates fluctuation variances one by one by traversing the infusion flow rate sequence and the adjacent medicine liquid level difference sequence of each bed, and quantifies the stability degree of the flow rate and the liquid level difference. In specific implementation, the mean value of each sequence is first calculated, then the square of the difference between each data point and the mean value is calculated, and the average value is taken as the fluctuation variance. For example, for the infusion flow rate sequence [4.2, 4.1, 4.3, 4.0], the mean value is 4.15 ml / s, and the fluctuation variance is 0.0125 ml² / s²; for the liquid level difference sequence [15, 14, 16, 14], the mean value is 14.75 mm, and the fluctuation variance is 0.6875 mm². These fluctuation variances reflect the stability of the flow rate and the liquid level change in the infusion process, and the smaller the fluctuation variance, the more stable the change; if the fluctuation variance is large, there may be abnormal fluctuation. The fluctuation variance set is calculated and output one by one for the flow rate and liquid level difference sequences of all beds, which provides accurate quantitative data support for subsequent stability analysis and abnormality identification.

[0023] a stable data screening subunit configured to traverse the K infusion flow rate sequences and the K adjacent medicine liquid level difference sequences to screen stable data, and obtain K infusion flow rate stable values and K adjacent medicine liquid level difference stable values. Specifically, the stable data screening subunit traverses the infusion flow rate sequence and the adjacent medicine liquid level difference sequence of each bed one by one, and constructs a flow rate data space and a liquid level difference data space. In the flow rate data space, the minimum value of the infusion flow rate is first determined, and a screening starting straight line parallel to the time axis is drawn, and a preset screening bandwidth (such as ±0.1 ml / s) is set to construct a screening neighborhood range. Then, the screening straight line is moved upward step by step, the neighborhood range is dynamically adjusted, the data in the neighborhood is screened, and the number of data points in each neighborhood is recorded. When the number of data points in the new neighborhood is less than that in the previous neighborhood, the iteration is stopped, and the mean value of the data in the final screening neighborhood is calculated to obtain the flow rate stable value of the bed. The same method is applicable to the liquid level difference data space, and the stable value of the liquid level difference is finally calculated by dynamically screening the neighborhood. For example, for the flow rate sequence [4.2, 4.1, 4.3, 4.0, 4.2] and the liquid level difference sequence [15, 14, 16, 14, 15], the flow rate stable value is 4.1 ml / s and the liquid level difference stable value is 14.5 mm after screening. These stable values can accurately reflect the stable state of the infusion process, and provide a reliable basis for subsequent abnormality detection.

[0024] The fluctuation variance anomaly identification subunit is configured to identify anomalies in the K infusion flow rate fluctuation variances and K infusion flow rate stability values, and the K adjacent medicine bag liquid level difference value fluctuation variances and K adjacent medicine bag liquid level difference value fluctuation variances using the stability anomaly identifier, to obtain the K medicine liquid delivery anomaly analysis results. Specifically, the fluctuation variance anomaly identification subunit identifies anomalies in the infusion process by jointly analyzing the infusion flow rate fluctuation variance, the infusion flow rate stability value, the liquid level difference value fluctuation variance, and the liquid level difference value stability value of each bed using a neural network model (stability anomaly identifier). First, the system combines the data of each bed into a feature vector, such as [0.012, 4.1, 0.8, 14.5], which respectively represent the flow rate fluctuation variance, the flow rate stability value, the liquid level difference value fluctuation variance, and the liquid level difference value stability value. The feature vector is input into the neural network model for anomaly probability calculation. The model uses a multi-layer structure, including an input layer, a hidden layer, and an output layer, and finally outputs an anomaly probability value. If the probability value exceeds a predetermined threshold (such as 0.7), the system determines that there is an anomaly in the infusion. For example, the feature vector of bed 1 outputs an anomaly probability of 0.85, which is determined to be abnormal and requires replacement of the medicine bag or other intervention operations; while the anomaly probability of bed 2 is 0.4, which is determined to be normal and does not require adjustment. Through this process, the system can quickly and efficiently identify anomalies in the infusion, ensuring the stability and reliability of the infusion process.

[0025] In a possible implementation, the stable data screening subunit further includes a liquid level difference value space construction micro-unit, configured to construct K infusion flow rate spaces and K adjacent medicine-water bag liquid level difference spaces based on the K infusion flow rate sequences and the K adjacent medicine-water bag liquid level difference value sequences, where the horizontal coordinate axis of each infusion flow rate space is time and the vertical coordinate axis is infusion flow rate, and the horizontal coordinate axis of each adjacent medicine-water bag liquid level difference space is time and the vertical coordinate axis is adjacent medicine-water bag liquid level difference. Specifically, the main function of the liquid level difference value space construction micro-unit is to construct infusion flow rate spaces and liquid level difference value spaces respectively by processing the infusion flow rate sequence and the adjacent medicine-water bag liquid level difference value sequence of each bed, to provide an intuitive data model for subsequent stable value screening and anomaly analysis. Specifically, the infusion flow rate space takes time as the horizontal axis and flow rate as the vertical axis, and represents the dynamic change of flow rate with time as a curve, for example, flow rate values [4.1, 4.2, 4.0] corresponding to time points [t1, t2, t3] can be plotted as points (t1, 4.1), (t2, 4.2), (t3, 4.0) and form a curve; similarly, the liquid level difference value space takes time as the horizontal axis and liquid level difference value as the vertical axis, and plots the liquid level change as a curve, for example, difference values [15, 14, 16] corresponding to points (t1, 15), (t2, 14), (t3, 16). These space models dynamically visualize the rules of infusion rate and liquid level change, providing intuitive support on the time axis for subsequent data screening and anomaly detection, and improving the system's ability to analyze dynamic characteristics.

[0026] A starting straight line determination micro-unit is configured to construct K screening starting straight lines respectively from the minimum values of infusion flow rate in the K infusion flow rate spaces and straight lines parallel to the horizontal coordinate axis. Specifically, the main function of the starting straight line determination micro-unit is to find the minimum value of flow rate from the infusion flow rate space of each bed, and to construct a starting screening straight line parallel to the time axis based on the value, as the initial reference for subsequent data screening and iteration. The specific process is as follows: the system traverses the infusion flow rate space of each bed, for example, for flow rate data [4.1, 4.0, 4.2, 3.8] and time points [t1, t2, t3, t4], the minimum flow rate value 3.8 and the corresponding time range [t1, t4] are determined, and then the straight line y=3.8 with the horizontal coordinate range [t1, t4] is constructed. This process is repeated in the infusion flow rate spaces of all beds, and the starting screening straight line of each bed is output, for example, the minimum value of the flow rate data [3.5, 3.6, 3.4, 3.8] of another bed is 3.4, and its corresponding straight line is y=3.4. These straight lines serve as the initial reference point for the screening operation, ensuring that the screening range fits the actual flow rate change, improving the efficiency of subsequent screening, and laying the foundation for dynamically capturing flow rate changes and screening stable values.

[0027] The starting straight line neighborhood construction micro unit is used to construct K screening starting straight line neighborhoods of the K screening starting straight lines in the K infusion flow rate spaces according to a preset screening bandwidth. Specifically, the function of the starting straight line neighborhood construction micro unit is to generate a neighborhood range based on the infusion flow rate space and the screening starting straight line of each bed in combination with the preset screening bandwidth, so as to provide reasonable upper and lower limits for subsequent screening. The specific process is as follows: the screening starting straight line (for example, the starting straight line of bed 1 is y=3.8, and the starting straight line of bed 2 is y=3.4) is extracted from the infusion flow rate space of each bed, the neighborhood range is formed by expanding around the starting straight line value according to the preset bandwidth (for example, ±0.5 flow rate unit), for example, the neighborhood of bed 1 is [3.3, 4.3], and the neighborhood of bed 2 is [2.9, 3.9]. These neighborhoods construct the upper and lower boundaries in the infusion flow rate space with the time axis as the horizontal coordinate and the flow rate value as the vertical coordinate. The dynamic construction of the neighborhood ensures that the screening range fits the actual flow rate change, effectively excludes noise data, and at the same time provides accurate initial reference points for subsequent iteration operations, thereby improving the accuracy of abnormality recognition and flow rate stability judgment.

[0028] The iterative straight line neighborhood construction micro unit is used to move the K screening starting straight lines upward according to the preset screening bandwidth, obtain K screening first iteration straight lines, and construct K screening first iteration straight line neighborhoods. Specifically, the iterative straight line neighborhood construction micro unit dynamically adjusts the screening straight line position and the neighborhood range to realize the optimization of flow rate data screening. Specifically, starting from the initial screening straight line of each bed (for example, y=3.8 represents the starting flow rate straight line of bed 1), the straight line is moved upward by the preset bandwidth (for example, ±0.5) to generate a new iteration straight line (for example, y=4.3 after one iteration, and y=4.8 after two iterations). In each iteration, the neighborhood is constructed according to the upper and lower floating ranges of the straight line (for example, the neighborhood range of y=4.3 is [3.8, 4.8]), and the number of flow rate data points contained in the neighborhood is counted. The system verifies whether the data quantity of the current iteration neighborhood is more than that of the previous iteration neighborhood, and if the data quantity is insufficient, the iteration is stopped, and the flow rate mean value in the last valid neighborhood is taken as the screening result (for example, the final screening straight line is y=4.8). This method is independently executed for each bed to dynamically adapt to the infusion flow rate characteristics of different beds, so as to ensure that the data screening range not only comprehensively covers the flow rate change, but also excludes noise data, thereby providing a high-precision flow rate stability basis for subsequent analysis.

[0029] The infusion flow rate stable value obtaining micro-unit is configured to determine whether the K screening one-iteration straight line neighborhood data quantity of the K screening one-iteration straight line neighborhood is greater than or equal to the K screening starting straight line neighborhood data quantity of the K screening starting straight line neighborhood. If not, the iteration is stopped, the mean value of the infusion flow rate in the K screening starting straight line neighborhood is calculated, and K infusion flow rate stable values are obtained. Specifically, the infusion flow rate stable value obtaining micro-unit extracts the flow rate stable value of each bed through multiple iteration screening, and provides accurate reference for subsequent analysis. Specifically, the system constructs a straight line neighborhood for the flow rate data of each bed in each iteration, counts the data quantity, and compares the data quantity of the current iteration neighborhood with the data quantity of the starting straight line neighborhood. If the data quantity of the iteration straight line neighborhood is less than or equal to the data quantity of the starting straight line neighborhood, it is determined that the iteration reaches the convergence condition, and the further iteration is stopped. Subsequently, the system calculates the mean value of the data in the starting straight line neighborhood as the flow rate stable value of the bed. For example, if the starting straight line neighborhood contains 50 data points, and the data quantity of the neighborhood is reduced to 48 points after iteration, the system stops iteration, and calculates the flow rate mean value of the 50 points as the stable value. This process is independently performed on all beds, and finally K flow rate stable values are generated, which comprehensively reflect the flow rate stable state of the infusion process, and avoid calculation redundancy caused by excessive iteration, thereby providing reliable data support for subsequent anomaly detection and control.

[0030] The liquid level difference value stable value obtaining micro-unit is configured to perform stable data screening in the K adjacent medicine bag liquid level difference value spaces, and obtain K adjacent medicine bag liquid level difference value stable values. Specifically, the liquid level difference value stable value obtaining micro-unit extracts the liquid level difference value stable value of each bed by performing stable data screening in the K adjacent medicine bag liquid level difference value spaces, and provides reliable data support for infusion control and anomaly detection. Specifically, the abscissa of the liquid level difference value space represents time, and the ordinate represents the difference value of the adjacent medicine bag liquid level. The system defines a stability screening window in the space, and analyzes the fluctuation range and mean value change of the data in the window. For the region with small liquid level difference value fluctuation range and stable mean value change, the system determines it as a stable data region, and calculates the mean value as the liquid level difference value stable value. For example, if the liquid level difference value of a bed fluctuates in a certain time period within the range of 0.5 to 0.6, and the mean value is 0.55, the system takes 0.55 as the liquid level difference value stable value of the bed. This process is independently performed on all beds, ensures the removal of noise and abnormal data, and extracts stable values that can accurately reflect the actual state of the infusion process.

[0031] In a possible implementation, the infusion flow rate stability value acquisition micro-unit further includes: an upward iteration unit, configured to, if no, perform upward iteration based on the K screening one-time iteration straight lines, obtain K screening N-time iteration straight lines through N iterations, and construct a K screening N-time iteration straight line neighborhood, where N is an integer greater than or equal to 2. Specifically, the upward iteration unit constructs and optimizes the screening straight line neighborhood by gradually adjusting the position of the screening straight line, to finally extract the stable infusion flow rate value. Specifically, the unit starts from the screening one-time iteration straight line, gradually moves the straight line position upward according to a fixed preset bandwidth, and reconstructs the straight line neighborhood and counts the number of data points in the straight line neighborhood after each adjustment. In each iteration, the system compares the data quantity of the current neighborhood with the data quantity of the last round neighborhood, and if the data quantity of the current neighborhood is not less than that of the last round, iteration is continued; otherwise, iteration is stopped, and the data mean value of the last round neighborhood is taken as the stable value. Meanwhile, when the number of iterations reaches a preset maximum value, the system forcibly stops iteration regardless of whether the data quantity continues to increase, and takes the neighborhood mean value of the last iteration as the final result. For example, in a certain infusion monitoring, if the data quantity of the third iteration neighborhood decreases, the system will back to the second iteration and take the mean value thereof as the stable value. This mechanism effectively balances data optimization and calculation reliability, and provides an accurate basis for subsequent analysis and control of the stability of the infusion flow rate.

[0032] The flow rate stable value output unit is configured to determine whether N is greater than or equal to a preset maximum iteration number when the K screening N iteration linear neighborhood data quantity of the K screening N iteration linear neighborhood is greater than or equal to the corresponding K screening N-1 iteration linear neighborhood data quantity, and if yes, stop iteration and take the mean value of the infusion flow rate of the K screening N iteration linear neighborhood as the K infusion flow rate stable value. Specifically, the main task of the flow rate stable value output unit is to accurately obtain the infusion flow rate stable value corresponding to the K beds through step-by-step iteration calculation. In the iteration process, the unit compares the data quantity of the current N iteration linear neighborhood with the data quantity of the previous (N-1) iteration linear neighborhood: if the data quantity of the N iteration linear neighborhood is greater than or equal to the data quantity of the N-1 iteration linear neighborhood, continue to iterate upwards to construct the N+1 iteration linear neighborhood, until one of the following two conditions is met: 1) the preset maximum iteration number is reached, at which time the iteration is stopped and the mean value of the infusion flow rate of the N iteration linear neighborhood is taken as the stable value; 2) if the data quantity of the linear neighborhood is less than the data quantity of the N-1 iteration linear neighborhood at the N iteration, the iteration is immediately stopped and the mean value of the infusion flow rate of the N-1 iteration linear neighborhood is taken as the stable value. For example, in the calculation of the infusion flow rate of a certain bed, if the neighborhood data quantity is less than the 4th iteration at the 5th iteration, the system will automatically back up and take the mean value of the 4th iteration neighborhood as the final stable value. Through this iteration and dynamic adjustment mechanism, the unit can effectively process the infusion flow rate data of the K beds, ensure that reliable stable values can be obtained in the case of data fluctuations and noise interference, and thus provide basic data support for subsequent precise infusion control.

[0033] The temperature sequence acquisition module 40 is configured to search the K infusion monitoring data set sequences with the infusion fluid temperature as the index, and obtain K infusion fluid temperature sequences. Specifically, the temperature sequence acquisition module 40 is configured to extract the infusion fluid temperature sequence corresponding to each bed from the infusion monitoring device arranged at the K beds. The module groups and traverses the infusion monitoring data set sequences by taking the infusion fluid temperature as the search index. The temperature sensor of each infusion monitoring device records the infusion fluid temperature in real time and uploads these data to the system. The module locks the data set of each bed one by one, extracts the temperature-related records therefrom, and arranges them in time sequence to form K independent infusion fluid temperature sequences. For example, if the monitoring data of a certain bed include three dimensions of flow rate, liquid level and temperature, the module filters out the temperature data points and sorts them in time sequence to generate the temperature sequence of the bed. In this way, the temperature sequence acquisition module ensures that the infusion fluid temperature data of all beds are complete and accurate, and provides reliable data support for subsequent temperature anomaly analysis and control.

[0034] The interactive correlation anomaly analysis module 50 is used for interactive correlation anomaly analysis on the K liquid temperature sequences, and K liquid temperature anomaly analysis results are obtained. Specifically, the interactive correlation anomaly analysis module 50 analyzes the K liquid temperature sequences, aiming to identify the abnormal situation of the liquid temperature of each bed and generate the corresponding anomaly analysis result. First, the module receives the K liquid temperature sequences provided by the temperature sequence acquisition module, which are arranged in chronological order and reflect the dynamic changes of the liquid temperature of each bed. The module evaluates the overall consistency and abnormal trend of the liquid temperature change by calculating the correlation and difference between each temperature sequence and other sequences, combined with the geographical distribution information of each bed. For example, when the liquid temperature of a certain bed deviates significantly from the average value of other beds, or there is an abnormal fluctuation in time, the bed will be marked as abnormal. In addition, the module also identifies local anomalies or changes in overall patterns by dynamically comparing the fluctuation range, change rate and difference value of the liquid temperature of each bed at adjacent time points. The analysis result is output in the form of K liquid temperature anomaly analysis results, each bed corresponding to a result, which clearly indicates the abnormal state of the liquid temperature, and provides an accurate basis for subsequent temperature adjustment or control decisions. For example, if the liquid temperature of a certain bed suddenly drops below the critical value, the module will immediately mark the bed as abnormal, facilitating the timely triggering of the temperature adjustment mechanism.

[0035] In a possible implementation, the interactive correlation anomaly analysis module 50 further includes a mean value calculation unit for calculating the mean value of the K liquid temperature sequences respectively, and obtaining K liquid temperature mean values. Specifically, the mean value calculation unit is used to process the K liquid temperature sequences, and the time mean value of each sequence is calculated by summing the temperature data at each time point in each sequence and dividing by the total number of time points, thereby generating K liquid temperature mean values. For example, for the liquid temperature sequence of a certain bed [36.5, 36.6, 36.4, 36.5, 36.7, 36.6, 36.5, 36.4, 36.6, 36.5], the unit divides the sum 365.4 by the number of time points 10, and obtains the mean value of 36.54°C. This process is performed for all beds one by one, and the generated set of K liquid temperature mean values comprehensively reflects the temperature conditions of different beds, providing data support for subsequent neighborhood anomaly analysis and system optimization. The unit ensures the accuracy of temperature data processing, and lays a foundation for quickly identifying temperature deviations and timely adjusting the liquid temperature.

[0036] The abnormality analysis result acquisition unit is configured to acquire K bed position information of the K beds, perform neighborhood abnormality analysis on the K average liquid temperature values based on the K bed position information, and obtain K liquid temperature abnormality analysis results. Specifically, the abnormality analysis result acquisition unit combines the position information of the K beds and the corresponding average liquid temperature values to perform neighborhood abnormality analysis and generate K liquid temperature abnormality analysis results. Specifically, the system first acquires the position information of all beds, which is used to determine the spatial relationship between beds, such as the logical or physical position of adjacent beds. Then, for each bed, the unit extracts the average liquid temperature value set of its neighborhood beds and calculates the average value of the set as the neighborhood reference value. Subsequently, the difference between the average liquid temperature value of the bed and the neighborhood reference value is calculated. If the difference exceeds a preset abnormality threshold, for example, 0.5°C, it is determined that the bed has a temperature abnormality; if the difference is within the threshold range, it is considered that the temperature is normal. In this way, the abnormality analysis result acquisition unit can effectively identify possible temperature abnormalities based on the comprehensive comparison of individual bed and neighborhood temperature data. For example, if the average liquid temperature value of a bed is 37.2°C and the average value of its neighborhood is 36.5°C, the bed will be marked as abnormal to prompt the system that there may be a heating device failure or environmental factor interference. This process ensures the accuracy and reliability of the system detection results, providing an important basis for subsequent regulation.

[0037] In one possible implementation, the abnormality analysis result acquisition unit further includes a first average liquid temperature value extraction sub-unit configured to randomly extract a first average liquid temperature value from the K average liquid temperature values. Specifically, the first average liquid temperature value extraction sub-unit randomly selects one value from the K average liquid temperature values as a reference for subsequent neighborhood analysis. This process first loads the average liquid temperature value set of all beds and assigns equal selection probability to each value through a random selection algorithm to ensure fairness and randomness of selection. The sub-unit randomly extracts an average liquid temperature value, named "first average liquid temperature value", and records and marks it for subsequent analysis. For example, in a 5-bed infusion system, the average liquid temperature values are 36.7°C, 36.9°C, 37.0°C, 37.1°C, and 36.8°C, and the sub-unit may randomly select 37.1°C as the first average liquid temperature value for subsequent neighborhood matching and abnormality analysis. Through this random extraction process, the system provides a diversity starting point for subsequent analysis, ensuring the robustness and reliability of the results.

[0038] The bed position matching subunit is configured to match the K bed positions according to the bed position corresponding to the first liquid temperature mean value, and extract the K liquid temperature means according to the matching result to obtain a first liquid temperature mean neighborhood. Specifically, the bed position matching subunit loads the first liquid temperature mean value and the corresponding bed position to determine the target position that needs to be matched. For example, if the first liquid temperature mean value is 37.1°C and the corresponding bed position is bed position 3, the system loads the liquid temperature mean value data and the position of all bed positions. Then, according to the matching rule (such as the neighborhood containing the previous and next 1 bed position), the neighborhood data related to the target bed position is extracted. Taking bed position 3 as an example, its neighborhood may include bed position 2 and bed position 4, and the corresponding liquid temperature mean values are 36.9°C and 37.0°C. Finally, the subunit outputs the extracted neighborhood liquid temperature data (such as 36.9°C, 37.1°C, and 37.0°C) as the first liquid temperature mean neighborhood for subsequent abnormality analysis. This matching process ensures that the system can accurately extract relevant neighborhood data, improving the relevance and accuracy of the analysis.

[0039] The liquid temperature abnormality judgment subunit is configured to calculate the difference between the mean value of the first liquid temperature mean neighborhood and the first liquid temperature mean value. If the calculation result is greater than or equal to a preset liquid temperature mean difference value, the first liquid temperature abnormality analysis result is a liquid temperature abnormality. Specifically, the liquid temperature abnormality judgment subunit calculates the difference between the mean value of the first liquid temperature mean neighborhood and the first liquid temperature mean value to determine whether the liquid temperature is abnormal. Specifically, first, the temperature values are extracted from the neighborhood data and the mean value is calculated, for example, the neighborhood mean value is 37.0°C. Then, the difference between the mean value and the first liquid temperature mean value (such as 36.5°C) is calculated to obtain a difference of 0.5°C. Then, the difference is compared with a preset temperature abnormality threshold (such as 0.4°C). If the difference is greater than or equal to the threshold, it is determined that the liquid temperature is abnormal; if the difference is less than the threshold, it is determined that the temperature is normal. For example, when the difference is 0.5°C and exceeds the threshold of 0.4°C, the system records the analysis result as "liquid temperature abnormality" and feeds back to the control module to trigger subsequent adjustment operations. This judgment process can quickly and accurately detect temperature abnormality, ensuring the reliability and stability of the system.

[0040] The drug solution temperature anomaly analysis result obtaining sub-unit is configured to perform neighborhood anomaly analysis on the remaining K-1 drug solution temperature means and obtain K-1 drug solution temperature anomaly analysis results. Specifically, the function of the drug solution temperature anomaly analysis result obtaining sub-unit is to perform neighborhood anomaly analysis on each of the remaining K-1 drug solution temperature means one by one, and finally obtain K-1 drug solution temperature anomaly analysis results. Specifically, this unit extracts each drug solution temperature mean from the analysis queue in turn, determines the neighborhood range based on its corresponding bed position, and the neighborhood range can be set by spatial or logical distance, for example, selecting adjacent beds or beds within a certain range as the neighborhood. Then, the mean of all drug solution temperature data in the neighborhood is calculated, and the difference between the extracted drug solution temperature mean is calculated. If the difference is greater than or equal to the preset anomaly threshold (for example, 0.5 ), the temperature data is marked as abnormal; otherwise, it is marked as normal. In this way, K-1 drug solution temperature means are analyzed one by one, and the anomaly analysis results of each time are recorded. Finally, these analysis results will be integrated and passed to the control module of the system for further decision-making and control operation, to ensure that the system can effectively identify and handle drug solution temperature anomaly situations. For example, when the drug solution temperature of a bed is 37.4 , the neighborhood mean is 36.8 , the difference is 0.6 , which exceeds the preset threshold 0.5 , the bed is marked as temperature anomaly, thereby triggering related processing.

[0041] The drug solution temperature anomaly analysis result determining sub-unit is configured to take the first drug solution temperature anomaly analysis result and the K-1 drug solution temperature anomaly analysis results as K drug solution temperature anomaly analysis results. Specifically, the function of the drug solution temperature anomaly analysis result determining sub-unit is to integrate and output the drug solution temperature anomaly analysis results of all beds to form a complete set of K drug solution temperature anomaly analysis results. Specifically, this sub-unit first receives the first drug solution temperature anomaly analysis result generated by the first drug solution temperature mean extraction and analysis process, and then combines the neighborhood anomaly analysis results of the remaining K-1 drug solution temperature means to organize these results in the order of bed number or index, generating a structured analysis result set { , ,..., } where each indicates whether there is an abnormal state of the corresponding bed, such as "abnormal" or "normal". This integration process not only ensures the integrity of the analysis results of all beds, but also supports the subsequent decision and control module calls. For example, in an infusion monitoring scenario, if the first and fourth beds are detected to have abnormal drug solution temperature, and the remaining beds are normal, the sub-unit will generate a result set {abnormal, normal, normal, abnormal, normal} and pass it to the control module to trigger the corresponding heating or alarm operation. Through this mechanism, the system realizes comprehensive monitoring and timely response to the abnormal state of the drug solution temperature.

[0042] In a possible implementation, the drug solution temperature abnormality judgment sub-unit further includes: calculating the difference between the mean value of the first drug solution temperature mean value neighborhood and the first drug solution temperature mean value, and if the calculation result is less than the preset drug solution temperature mean value difference, the first drug solution temperature abnormality analysis result is normal drug solution temperature. Specifically, the difference between the mean value of the first drug solution temperature mean value neighborhood and the first drug solution temperature mean value is calculated to determine whether the current drug solution temperature is within the normal range. Specifically, this micro-unit first receives the first drug solution temperature mean value and the mean value of its neighborhood, and calculates the difference between the two through subtraction. Then, the difference is compared with the system preset drug solution temperature mean value difference threshold. If the difference is less than the preset threshold, it is determined that the current drug solution temperature state is "normal", and the judgment result is recorded as the first drug solution temperature abnormality analysis result. For example, if the first drug solution temperature mean value is 37.5°C, the neighborhood mean value is 37.6°C, and the difference is 0.1°C, which is less than the preset threshold of 0.5°C, the system will mark the result as "normal temperature". This determination mechanism can effectively filter out false positives caused by slight fluctuations, and provide accurate and reliable basic data for subsequent drug solution temperature monitoring and control.

[0043] The infusion control module 60 is used to trigger the infusion valves, the medicine bag replacement device and the heating device of the K beds according to the K infusion abnormality analysis results and the K temperature abnormality analysis results, and complete the infusion control of the K beds. Specifically, the infusion control module 60 realizes the infusion control of the K beds by comprehensively analyzing the K infusion abnormality analysis results and the K temperature abnormality analysis results. Specifically, the module judges the situation of insufficient liquid level or abnormal flow rate according to the infusion abnormality analysis results, triggers the medicine bag replacement device or adjusts the opening degree of the infusion valve of the corresponding bed to ensure sufficient liquid level and stable flow rate. At the same time, if the temperature abnormality analysis result shows that the temperature exceeds the preset range, the module starts the heating device to adjust the temperature of the medicine to the target range. For example, when the liquid level of a bed is detected to be insufficient, the system immediately triggers the medicine bag replacement device to complete the replacement of the medicine bag; if the flow rate is detected to be unstable, the opening degree of the infusion valve is adjusted until the flow rate returns to normal; if the temperature of the medicine is lower than 36°C (the target range is 36°C to 38°C), the system activates the heating device to raise the temperature to 37°C. The whole module realizes real-time monitoring, instruction triggering and closed-loop adjustment to ensure the stability of the infusion process, no interruption, and maintain the flow rate and temperature of the medicine in the best state, and comprehensively guarantee the accuracy and reliability of the infusion control.

[0044] The embodiment of the present application adopts the infusion monitoring data arranged on the K beds to obtain the infusion flow rate, the medicine bag liquid level and the medicine temperature sequence; performs joint abnormality analysis on the flow rate and the liquid level to obtain the infusion abnormality result; performs interactive correlation analysis on the temperature sequence to obtain the temperature abnormality result; triggers the infusion valve, the medicine bag replacement device and the heating device according to the analysis results to realize the intelligent control of the infusion process, and achieves the technical effects of improving the reliability of the infusion process control and the fitting degree of the actual infusion situation.

[0045] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or the server, and each unit and module included is only divided according to the functional logic, but is not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy mutual distinction, and do not limit the protection scope of the present application.

[0046] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. An infusion management and control system based on the Internet of Things, characterized in that, The system includes: The monitoring data acquisition module is used to collect monitoring data from K infusion monitoring devices deployed at K beds within a preset monitoring window, and obtain K infusion monitoring data set sequences, where K is an integer greater than or equal to 1; A liquid level sequence acquisition module is used to retrieve the K sets of infusion monitoring data sequences using infusion flow rate and drug bag liquid level as indexes, and obtain K infusion flow rate sequences and K drug bag liquid level sequences. An anomaly analysis module is used to perform joint anomaly analysis on the K infusion flow rate sequences and the K drug package level sequences to obtain K drug delivery anomaly analysis results. A temperature sequence acquisition module is used to retrieve the K infusion monitoring data sets using the drug solution temperature as an index, and obtain the K drug solution temperature sequences. An interactive correlation anomaly analysis module is used to perform interactive correlation anomaly analysis on the K drug solution temperature sequences to obtain K drug solution temperature anomaly analysis results. An infusion control module is used to trigger the infusion valves, medicine pack replacement devices, and heating devices of the K beds based on the abnormal analysis results of the K drug delivery and the K abnormal drug temperature, thereby completing the infusion control of the K beds. The interaction-related anomaly analysis module includes: The mean calculation unit is used to calculate the mean of the K drug solution temperature sequences respectively to obtain the mean of the K drug solution temperature; An anomaly analysis result acquisition unit is used to acquire the K bed locations of the K beds, and perform neighborhood anomaly analysis on the average temperature of the K medicine solutions based on the K bed locations to obtain the K medicine solution temperature anomaly analysis results. The anomaly analysis result acquisition unit includes: The first average temperature extraction subunit of the liquid medicine is used to randomly extract the average temperature of the first liquid medicine from the K average temperatures of the liquid medicine. A bed location matching subunit is used to match the bed location corresponding to the first average drug temperature with the K bed locations, and extract the K average drug temperatures based on the matching results to obtain the neighborhood of the first average drug temperature. The medicine liquid temperature abnormality judgment subunit is used to calculate the difference between the mean of the neighborhood of the first medicine liquid temperature mean and the first medicine liquid temperature mean. If the calculation result is greater than or equal to the preset medicine liquid temperature mean difference, the first medicine liquid temperature abnormality analysis result is medicine liquid temperature abnormality. The subunit for obtaining the results of the drug solution temperature anomaly analysis is used to perform neighborhood anomaly analysis on the remaining K-1 average drug solution temperatures to obtain K-1 drug solution temperature anomaly analysis results. The sub-unit for determining the abnormal analysis results of liquid medicine temperature is used to take the first abnormal analysis result of liquid medicine temperature and the K-1 abnormal analysis results of liquid medicine temperature as K abnormal analysis results of liquid medicine temperature. Calculate the difference between the mean of the neighborhood of the first liquid temperature and the mean of the first liquid temperature. If the calculated result is less than the preset difference of the mean of the liquid temperature, the result of the first liquid temperature anomaly analysis is that the liquid temperature is normal.

2. The infusion management and control system based on the Internet of Things as described in claim 1, characterized in that, The anomaly analysis module includes: A preset liquid level requirement judgment unit is used to extract the liquid levels of the last K liquid bags in the K liquid level sequences, and determine whether the liquid levels of the last K liquid bags meet the preset liquid level requirements. If so, the unit performs stability anomaly identification on the K infusion flow rate sequences and the K liquid bag liquid level sequences to obtain the K liquid delivery anomaly analysis results. An anomaly analysis result output unit is used to, if not, take the replacement of the medicine pack as K medicine delivery anomaly analysis results.

3. The infusion management and control system based on the Internet of Things as described in claim 2, characterized in that, The preset liquid level requirement judgment unit includes: A liquid level difference calculation subunit is used to calculate the liquid level difference between two adjacent liquid levels in the K liquid level sequences of liquid packs, and obtain K adjacent liquid level difference sequences of liquid packs. A fluctuation variance calculation subunit is used to traverse the K infusion flow rate sequences and the K adjacent drug pack level difference value sequences to calculate the fluctuation variance, and determine the fluctuation variance of the K infusion flow rate sequences and the fluctuation variance of the K adjacent drug pack level difference value sequences. A stable data filtering subunit is used to traverse the K infusion flow rate sequences and the K adjacent drug pack level difference sequences to perform stable data filtering, and obtain K stable values ​​of infusion flow rate and K stable values ​​of adjacent drug pack level difference. The fluctuation variance anomaly identification subunit is used to identify anomalies in K infusion flow rate fluctuation variances and K infusion flow rate stability values, as well as the fluctuation variances of the K adjacent drug pack level difference values ​​and the K adjacent drug pack level difference value fluctuation variances, by using a stability anomaly identifier, and to obtain the K drug delivery anomaly analysis results.

4. The infusion management and control system based on the Internet of Things as described in claim 3, characterized in that, The stable data filtering subunit includes: A micro-unit for constructing liquid level difference space is used to construct K infusion flow rate spaces and K adjacent drug pack liquid level difference spaces based on the K infusion flow rate sequences and the K adjacent drug pack liquid level difference sequences. In each infusion flow rate space, the horizontal axis is time and the vertical axis is infusion flow rate. In each adjacent drug pack liquid level difference space, the horizontal axis is time and the vertical axis is the liquid level difference between adjacent drug packs. The starting line determination micro-unit is used to select K starting lines for screening, which are lines that pass through the minimum infusion flow rate in K infusion flow rate spaces and are parallel to the horizontal axis. A starting line neighborhood construction micro-unit is used to construct K screening starting line neighborhoods of the K screening starting lines in the K infusion flow rate spaces according to a preset screening bandwidth. The iterative line neighborhood construction micro-unit is used to move the K initial screening lines upward according to the preset screening bandwidth to obtain K screening one-time iteration lines and construct the neighborhood of the K screening one-time iteration lines. The micro-unit for obtaining stable infusion flow rate is used to determine whether the data volume of the K linear neighborhoods of the K linear neighborhoods of the first iteration of the screening is greater than or equal to the data volume of the K linear neighborhoods of the corresponding initial screening linear neighborhoods. If not, the iteration is stopped, the average infusion flow rate in the K linear neighborhoods of the initial screening linear neighborhoods is calculated, and K stable infusion flow rate values ​​are obtained. A micro-unit for obtaining stable liquid level difference values ​​is used to filter stable data in the liquid level difference space of K adjacent medicine bags to obtain stable liquid level difference values ​​of K adjacent medicine bags.

5. The infusion management and control system based on the Internet of Things as described in claim 4, characterized in that, The micro-unit for obtaining the stable infusion flow rate includes: An upward iterative unit is used to perform upward iteration based on the K filtered one-time iteration lines if no, and after N iterations, obtain K filtered N-time iteration lines and construct the neighborhood of the K filtered N-time iteration lines, where N is an integer greater than or equal to 2; The flow rate stability output unit is used to determine whether N is greater than or equal to the preset maximum number of iterations when the data volume of the K filtered N-iteration linear neighborhoods is greater than or equal to the corresponding data volume of the K filtered N-1-iteration linear neighborhoods. If so, the iteration is stopped, and the average infusion flow rate of the K filtered N-iteration linear neighborhoods is taken as the K infusion flow rate stability value.

Images