Intelligent infusion monitoring method and system
By combining weight and flow sensors in an intelligent monitoring method, the remaining amount of medication is dynamically analyzed and a priority treatment group is generated, which solves the problem of inaccurate monitoring of the remaining amount of medication in the infusion pump, and achieves precise control of the infusion process and improves nursing efficiency.
Patent Information
- Application Number
- CN202511709130.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-20
- Publication Date
- 2026-01-02
AI Technical Summary
The existing infusion pumps lack drug metering and time monitoring functions, resulting in significant differences between the calculated drug remaining amount and infusion completion time and the actual results. This increases the nursing workload and affects the accurate execution of the treatment plan, especially causing inconvenience to patients who need to be discharged after the infusion is completed.
An intelligent monitoring method combining weight and flow sensors is used to achieve precise monitoring of the chemotherapy drug infusion process by dynamically analyzing the changing trend of the remaining drug amount, combined with the abnormal fluctuation judgment and priority treatment group generation mechanism. This includes real-time monitoring of the remaining drug amount and accurate determination of the infusion end time.
It enables precise monitoring of the chemotherapy drug infusion process, improves nursing efficiency, ensures treatment safety, and enhances the patient's medical experience.
Smart Images
Figure CN121243542A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of medical devices, and particularly relates to an intelligent infusion monitoring method and system. BACKGROUND
[0002] In clinical chemotherapy treatment, accurate infusion of drugs is crucial for treatment effect and patient safety. Taking 5-fluorouracil as an example, the drug is usually configured into an infusion pump by a static preparation center and is administered by a 48-hour continuous pumping method. However, in actual application, due to individual differences of patients (such as body position changes, activity frequency, etc.) and influences of the surrounding environment (such as temperature, pressure changes, etc.), the actual pumping speed of the infusion pump often deviates from the preset value, with the longest deviation reaching several hours. Such deviation makes it difficult for medical staff to accurately grasp the remaining amount and infusion remaining time of the drug, thereby affecting the accurate execution of the treatment plan.
[0003] At present, infusion pumps on the market generally lack drug liquid metering and time monitoring functions, and the monitoring of the remaining amount and the remaining time of the drug in the clinic mainly relies on medical staff to manually check and speculate at regular intervals. This method not only increases the nursing workload, but also is prone to errors due to human factors, resulting in a significant difference between the calculated result and the actual infusion completion time. In addition, for patients who need to be discharged after the infusion is completed, the accurate infusion completion time is an important basis for their subsequent travel arrangements (such as purchasing a train ticket), and the existing technology cannot meet this demand, bringing many inconveniences to patients.
[0004] To solve the above problems, an intelligent monitoring device that can be compatible with existing infusion pumps and accurately monitor the remaining amount and the remaining time of the drug in real time is urgently needed. The device should have high-precision monitoring capability and be easy for patients to carry, so as to improve chemotherapy nursing efficiency, ensure treatment safety and improve the patient's medical experience. However, there is no mature solution to the above-mentioned needs in the existing technology, especially the research on combining weight sensors and flow sensors to achieve double monitoring and improve monitoring reliability is still insufficient. Therefore, it is of great clinical significance and application value to develop an intelligent infusion monitoring method and system based on weight sensors and flow sensors. SUMMARY
[0005] In order to make up for the deficiencies of the prior art and solve at least one technical problem proposed in the background art.
[0006] In a first aspect, the application provides an intelligent infusion monitoring method, comprising the following steps: obtaining drug liquid information and infusion information, wherein the drug liquid information includes an initial drug weight and a pump body out-drug pipeline flow, and the infusion information includes an infusion start time and a real-time pumping speed; Based on the patient's initial medication weight and real-time infusion rate, combined with a preset monitoring period, the remaining medication amount within the monitoring period is dynamically analyzed to determine whether the remaining medication amount conforms to the expected trend. If the remaining drug level conforms to the expected trend, the remaining infusion time is calculated based on the current pumping rate, and the infusion end time is determined. If the remaining drug level does not conform to the expected trend, an abnormal fluctuation analysis is performed on the remaining drug level to determine whether there is a continuous abnormal fluctuation. If it is a non-continuous abnormal fluctuation, the abnormal data points are removed, the valid data points are retained, and the remaining infusion time is recalculated to determine the infusion end time. Based on the infusion end time of all patients, patients were grouped for optimized treatment, and priority treatment groups were generated according to priority order. Patients in the priority treatment groups received priority intervention.
[0007] As a further aspect of the present invention: the process of dynamically analyzing the remaining amount of drug within the monitoring period is as follows: The monitoring period is divided into several sampling time points. The remaining amount of drug at each sampling time point is obtained and analyzed to obtain the standard deviation and mean of the remaining amount of drug within the monitoring period. The fluctuation index of the remaining amount of drug is calculated using the coefficient of variation formula. If the fluctuation index is greater than or equal to the fluctuation threshold, it indicates that the remaining amount of drug within the monitoring period does not conform to the expected trend. Otherwise, it indicates that the remaining amount of drug within the monitoring period conforms to the expected trend.
[0008] As a further aspect of the present invention: the standard deviation and mean of the remaining drug amount during the monitoring period are obtained as follows: The remaining drug levels at all sampling time points within the monitoring period are integrated into a remaining drug level sequence. Extract all data from the drug remaining quantity sequence, calculate the mean of the drug remaining quantity within the monitoring period using the mean formula, and extract all data from the drug remaining quantity sequence, calculate the standard deviation of the drug remaining quantity within the monitoring period using the standard deviation formula.
[0009] As a further aspect of the present invention: the process for obtaining the infusion end time when the remaining drug volume conforms to the expected trend is as follows: The amount of drug that has been pumped in is obtained by multiplying the duration of the monitoring cycle with the current pumping rate. The remaining drug volume is calculated by comparing it with the current infusion rate to obtain the remaining infusion time. The time when the remaining infusion time ends is recorded as the infusion end time.
[0010] As a further aspect of the present invention: if the remaining amount of drug does not conform to the expected trend, the specific process for performing abnormal fluctuation analysis on the remaining amount of drug is as follows: Analyze the remaining amount of drug at each sampling time point to identify the abrupt change points within the sampling time points; Traverse all mutation time points on the time axis, count the number of consecutive occurrences of each mutation time point on the time axis, and compare the result with the maximum value of consecutive occurrences of a preset mutation time point to obtain the mutation continuity index. If the mutation continuity index is less than the mutation continuity threshold, it indicates that the remaining amount of drug exhibits discontinuous abnormal fluctuations.
[0011] As a further aspect of the present invention: the process of obtaining the abrupt change time point in the sampling time point is as follows: The difference between the remaining amount of drug at the sampling time point and the preset remaining amount of drug is calculated, and the absolute value of the difference is taken to obtain the abnormal fluctuation deviation value. If the abnormal fluctuation deviation value is not within the preset fluctuation deviation range, the corresponding sampling time point will be recorded as the sudden change time point.
[0012] As a further aspect of the present invention: the process of acquiring the effective data points is as follows: Calculate the percentage of mutation time points in the sampling time points. If the percentage of mutation time points is less than the threshold, remove the data points corresponding to the mutation time points from the drug remaining quantity sequence of the monitoring period, and retain the data points of normal time points. The data points of the retained normal time points are recorded as valid data points.
[0013] As a further aspect of the present invention: the process for obtaining the infusion end time when the remaining drug amount does not conform to the expected trend is as follows: All valid data points at normal time points are integrated into a valid data sequence. All data in the valid data sequence are extracted and the mean is calculated to obtain the current effective infusion rate of the drug. The remaining drug amount is compared with the current effective infusion rate to obtain the remaining infusion time. The time point at which the remaining infusion time ends is recorded as the infusion end time.
[0014] As a further aspect of the present invention: the process of obtaining the priority processing group is as follows: The infusion end times of all patients' current medications were integrated into an infusion end time dataset, and the DBSCAN algorithm was used for clustering to determine the infusion end time clusters. All infusion end-of-infusion clusters are sorted in chronological order to obtain an infusion end-of-infusion cluster ranking table. The infusion end-of-infusion cluster that ranks first in the ranking table is selected as the priority treatment group.
[0015] Secondly, the present invention provides an intelligent infusion monitoring system, the system comprising: Data acquisition module: acquires drug solution information and infusion information. The drug solution information includes the initial drug weight and the flow rate of the pump's outflow pipe, while the infusion information includes the infusion start time and the real-time pumping speed. Dynamic analysis module: Based on the patient's initial drug weight and real-time infusion rate, combined with the preset monitoring period, the remaining drug amount within the monitoring period is dynamically analyzed to determine whether the remaining drug amount conforms to the expected trend. Abnormal fluctuation judgment module: If the remaining drug amount conforms to the expected trend, the remaining infusion time is calculated based on the current pumping rate, and the infusion end time is determined; if the remaining drug amount does not conform to the expected trend, abnormal fluctuation analysis is performed on the remaining drug amount to determine whether there is continuous abnormal fluctuation. If it is non-continuous abnormal fluctuation, abnormal data points are removed, valid data points are retained, and the remaining infusion time is recalculated to determine the infusion end time. Priority treatment group generation module: Based on the infusion end time of all patients, patients are grouped and optimized for treatment, priority treatment groups are generated according to priority order, and priority intervention is given to patients in the priority treatment groups.
[0016] In the above method, the formula for the coefficient of variation is:
[0017] in, Indicates volatility index, The standard deviation of the remaining amount of drug This represents the average amount of drug remaining.
[0018] The formula for the mutation continuity index is:
[0019] in, Indicates the continuous mutation index. This indicates the number of consecutive occurrences of the mutation time point. This represents the maximum number of consecutive occurrences of a preset mutation time point.
[0020] Through the above methods and systems, this invention achieves precise monitoring of the chemotherapy drug infusion process, significantly improves nursing efficiency, ensures treatment safety, and enhances the patient's medical experience. Attached Figure Description
[0021] Appendix Figure 1 Core workflow diagram of intelligent infusion monitoring system Detailed Implementation
[0022] This invention provides an intelligent infusion monitoring method and system. Its core lies in achieving precise monitoring of the chemotherapy drug infusion process by dynamically analyzing the changing trend of remaining drug levels, combined with an abnormal fluctuation judgment and priority treatment group generation mechanism. The specific implementation of this invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
[0023] In practical applications, as shown in the appendix Figure 1 As shown, the intelligent infusion monitoring method of the present invention first acquires drug information and infusion information through a data acquisition module. Drug information includes the initial drug weight and the flow rate of the pump's dispensing pipeline, while infusion information includes the infusion start time and real-time pumping speed. This data acquisition can be accomplished using devices such as sensors, electronic scales, and flow meters. For example, in a specific implementation scenario, the initial drug weight can be measured and recorded using a high-precision electronic scale, the flow rate of the pump's dispensing pipeline is monitored in real-time by a flow meter, the infusion start time is automatically recorded by the system, and the real-time pumping speed is fed back to the system through the pump controller. This data forms the basis for subsequent dynamic analysis.
[0024] Next, the dynamic analysis module dynamically analyzes the remaining drug amount within the monitoring period based on the patient's initial medication weight and real-time infusion rate, combined with a preset monitoring cycle. The core of this process is determining whether the remaining drug amount conforms to the expected trend. To achieve this, the system divides the monitoring cycle into several sampling time points and acquires the remaining drug amount at each point. Subsequently, the system integrates the remaining drug amounts from all sampling time points into a remaining drug amount sequence, extracts all data from this sequence, and calculates the mean and standard deviation. The formula for the mean is... ,in Indicates the number of sampling time points. Indicates the first The remaining amount of drug at each sampling time point; the standard deviation formula is: Based on the mean and standard deviation, the system uses the coefficient of variation formula. Calculate the fluctuation index of the remaining drug quantity. If the fluctuation index is greater than or equal to the preset fluctuation threshold, the remaining drug quantity is determined to be inconsistent with the expected trend; otherwise, it is determined to be consistent with the expected trend.
[0025] When the remaining drug level conforms to the expected trend, the system calculates the remaining infusion time based on the current pumping rate and determines the infusion end time. Specifically, the system multiplies the duration of the monitoring period by the current pumping rate to obtain the amount of drug already pumped. Then, the system ratios the remaining drug level to the current pumping rate to obtain the remaining infusion time, and records the end time of this remaining infusion time as the infusion end time. For example, assuming the current pumping rate is 5 ml / min and the remaining drug level is 100 ml, the remaining infusion time is 20 minutes, and the infusion end time is the current time plus 20 minutes.
[0026] However, when the remaining drug quantity does not conform to the expected trend, the system needs to further analyze the abnormal fluctuations in the remaining drug quantity. The key to this process is determining whether continuous abnormal fluctuations exist. First, the system analyzes the remaining drug quantity at each sampling time point to identify abrupt change points. Specifically, the system subtracts the remaining drug quantity at each sampling time point from a preset remaining drug quantity and takes the absolute value of the difference to obtain the abnormal fluctuation deviation value. If the abnormal fluctuation deviation value is not within the preset fluctuation deviation range, the corresponding sampling time point is recorded as the abrupt change point. Subsequently, the system iterates through all abrupt change points on the time axis, counts the number of consecutive occurrences of each abrupt change point on the time axis, and compares this number with the maximum value of the preset consecutive occurrences of each abrupt change point to obtain the abrupt change continuity index. The formula for the abrupt change continuity index is: in This indicates the number of consecutive occurrences of the mutation time point. This represents the maximum number of consecutive occurrences of a preset mutation time point. If the mutation continuity index is less than the mutation continuity threshold, it indicates that the remaining drug quantity exhibits discontinuous and abnormal fluctuations.
[0027] For cases of discontinuous abnormal fluctuations, the system needs to remove abnormal data points and retain valid data points. Specifically, the system calculates the percentage of abrupt changes in the number of sampling time points. If the percentage of abrupt changes is less than a threshold, the data points corresponding to the abrupt changes are removed from the remaining drug quantity sequence of the monitoring period, while the data points from normal time points are retained and recorded as valid data points. Subsequently, the system integrates all valid data points from normal time points into a valid data sequence and extracts all data from the valid data sequence to calculate the mean, thus obtaining the current effective drug infusion rate. Finally, the system calculates the ratio of the remaining drug quantity to the current effective drug infusion rate to obtain the remaining infusion duration, and records the end time of the remaining infusion duration as the infusion end time.
[0028] After calculating the infusion end time for a single patient, the system enters the priority treatment group generation module. This module integrates the infusion end times of all patients' current medications into an infusion end time dataset and uses the DBSCAN algorithm to perform cluster analysis to determine the infusion end segment clusters. Subsequently, the system sorts all infusion end segment clusters in chronological order, obtaining an infusion end segment cluster ranking table, and extracts the infusion end segment cluster at the top of the ranking table as the priority treatment group. By prioritizing intervention for patients in the priority treatment group, the system can significantly improve nursing efficiency and ensure treatment safety.
[0029] The operating principle and process of the system can be explained in conjunction with specific application scenarios. For example, in a chemotherapy ward of a hospital, nurses need to manage the infusion process of multiple patients simultaneously. Traditional manual monitoring methods are not only inefficient but also prone to human error, leading to the failure to detect infusion abnormalities in a timely manner. However, with the intelligent infusion monitoring system of this invention, nurses only need to input data such as the initial drug weight, the flow rate of the pump's delivery tubing, the infusion start time, and the real-time pumping speed into the system to achieve fully automated monitoring of the infusion process. The system will automatically determine whether there are any abnormalities based on the trend of changes in the remaining drug amount and, if necessary, remove abnormal data points and recalculate the infusion end time. In addition, the system will generate priority processing groups based on the infusion end times of all patients, helping nurses to rationally arrange the work sequence, thereby significantly improving nursing efficiency.
[0030] It is worth noting that the intelligent infusion monitoring system of this invention can also be combined with various hardware devices to achieve more efficient data acquisition and processing. For example, the system can be equipped with a high-precision electronic scale to measure the initial drug weight, a flow meter to monitor the flow rate of the pump's drug delivery pipeline in real time, a time recording module to record the infusion start time, and a pump controller to provide feedback on the real-time pumping speed. These hardware devices are connected to the system via wired or wireless means to ensure the real-time nature and accuracy of the data. In addition, the system can also store and analyze data through a cloud server, allowing medical staff to view the infusion status and make remote interventions at any time.
[0031] In summary, this invention achieves precise monitoring of the chemotherapy drug infusion process by dynamically analyzing the changing trend of remaining drug levels, combined with abnormal fluctuation judgment and priority treatment group generation mechanisms. The system not only significantly improves nursing efficiency and ensures treatment safety but also enhances the patient's medical experience. In practical applications, the system can flexibly adjust parameter settings according to different needs, such as the length of the monitoring period, the magnitude of the fluctuation threshold, and the setting of continuous mutation thresholds, thereby adapting to the infusion monitoring requirements of different scenarios.
Claims
1. An intelligent infusion monitoring system, characterized in that: Includes the following modules: Data acquisition module: acquires drug solution information and infusion information. The drug solution information includes the initial drug weight and the flow rate of the pump's outflow pipe, while the infusion information includes the infusion start time and the real-time pumping speed. Dynamic analysis module: Based on the patient's initial drug weight and real-time infusion rate, combined with the preset monitoring period, the remaining drug amount within the monitoring period is dynamically analyzed to determine whether the remaining drug amount conforms to the expected trend. Abnormal fluctuation judgment module: If the remaining drug amount conforms to the expected trend, the remaining infusion time is calculated based on the current pumping rate, and the infusion end time is determined; if the remaining drug amount does not conform to the expected trend, abnormal fluctuation analysis is performed on the remaining drug amount to determine whether there is continuous abnormal fluctuation. If it is non-continuous abnormal fluctuation, abnormal data points are removed, valid data points are retained, and the remaining infusion time is recalculated to determine the infusion end time. Priority treatment group generation module: Based on the infusion end time of all patients, patients are grouped and optimized for treatment, priority treatment groups are generated according to priority order, and priority intervention is given to patients in the priority treatment groups.
2. The intelligent infusion monitoring system according to claim 1, characterized in that: The process of dynamically analyzing the remaining amount of drug during the monitoring period is as follows: The monitoring period is divided into several sampling time points. The remaining amount of drug at each sampling time point is obtained and analyzed to obtain the standard deviation and mean of the remaining amount of drug within the monitoring period. The fluctuation index of the remaining amount of drug is calculated using the coefficient of variation formula. If the fluctuation index is greater than or equal to the fluctuation threshold, it indicates that the remaining amount of drug within the monitoring period does not conform to the expected trend. Otherwise, it indicates that the remaining amount of drug within the monitoring period conforms to the expected trend.
3. The intelligent infusion monitoring system according to claim 2, characterized in that: The standard deviation and mean of the remaining drug amount during the monitoring period were obtained as follows: The remaining drug levels at all sampling time points within the monitoring period are integrated into a remaining drug level sequence. Extract all data from the drug remaining quantity sequence, calculate the mean of the drug remaining quantity within the monitoring period using the mean formula, and extract all data from the drug remaining quantity sequence, calculate the standard deviation of the drug remaining quantity within the monitoring period using the standard deviation formula.
4. The intelligent infusion monitoring system according to claim 3, characterized in that: The process for obtaining the infusion end time when the remaining drug volume conforms to the expected trend is as follows: The amount of drug that has been pumped in is obtained by multiplying the duration of the monitoring cycle with the current pumping rate. The remaining drug volume is calculated by comparing it with the current infusion rate to obtain the remaining infusion time. The time when the remaining infusion time ends is recorded as the infusion end time.
5. The intelligent infusion monitoring system according to claim 4, characterized in that: The remaining drug quantity did not conform to the expected trend. The specific process for performing abnormal fluctuation analysis on the remaining drug quantity is as follows: Analyze the remaining amount of drug at each sampling time point to identify the abrupt change points within the sampling time points; Traverse all mutation time points on the time axis, count the number of consecutive occurrences of each mutation time point on the time axis, and compare the result with the maximum value of consecutive occurrences of a preset mutation time point to obtain the mutation continuity index. If the mutation continuity index is less than the mutation continuity threshold, it indicates that the remaining amount of drug exhibits discontinuous abnormal fluctuations.
6. The intelligent infusion monitoring system according to claim 5, characterized in that: The process for obtaining the abrupt change time points in the sampling time points is as follows: The difference between the remaining amount of drug at the sampling time point and the preset remaining amount of drug is calculated, and the absolute value of the difference is taken to obtain the abnormal fluctuation deviation value. If the abnormal fluctuation deviation value is not within the preset fluctuation deviation range, the corresponding sampling time point will be recorded as the sudden change time point.
7. The intelligent infusion monitoring system according to claim 5, characterized in that: The process of obtaining the valid data points is as follows: Calculate the percentage of mutation time points in the sampling time points. If the percentage of mutation time points is less than the threshold, remove the data points corresponding to the mutation time points from the drug remaining quantity sequence of the monitoring period, and retain the data points of normal time points. The data points of the retained normal time points are recorded as valid data points.
8. The intelligent infusion monitoring system according to claim 5, characterized in that: The process for obtaining the priority processing group is as follows: The infusion end times of all patients' current medications were integrated into an infusion end time dataset, and the DBSCAN algorithm was used for clustering to determine the infusion end time clusters. All infusion end-of-infusion clusters are sorted in chronological order to obtain an infusion end-of-infusion cluster ranking table. The infusion end-of-infusion cluster that ranks first in the ranking table is selected as the priority treatment group.