Sensor and intelligent identification-based extravasation early warning system and method in injection process

Abstract

The invention relates to the technical field of intelligent identification, and particularly discloses an extravasation early warning system and method in an injection process based on a sensor and intelligent identification. The system is provided with an individualized initial monitoring threshold determination module, an intelligent monitoring patch index analysis module, an injection pump linkage index analysis module, an in-injection-process extravasation risk analysis module, a multi-level alarm and linkage intervention response module and an injection process closed-loop tracking module. The problems that a traditional infusion pump block alarm is not sensitive and unclear to exosmosis detection, an existing exosmosis detection device is tedious in operation, difficult to detect early-stage exosmosis or prone to generating false alarms, mainly depends on manual observation of skin surface signs, is low in accuracy, has technical shortages and is poor in safety and the like are solved. And non-contact optical detection equipment is possibly interfered by skin pigments, ambient light and the like, so that the detection stability is poor.

Description

Technical Field

[0001] This invention relates to the field of intelligent identification technology, specifically to an extravasation early warning system and method for injection processes based on sensors and intelligent identification. Background Technology

[0002] In the medical field, intravenous injection is a common treatment method, but the problem of extravasation during injection cannot be ignored. For example, extravasation of iodine contrast agents can cause a series of pathophysiological changes due to alterations in the osmotic pressure gradient. When administered via high-pressure syringe, extravasation can cause strong irritation to subcutaneous tissues, potentially leading to swelling and pain. In severe cases, it can result in complications such as tissue necrosis, ulceration, and interfascial syndrome, affecting limb function and diagnostic results. Extravasation of contrast agents during contrast-enhanced CT scans can affect image quality, causing patient discomfort and examination failure. With the development of medical imaging technology, the use of radioactive contrast agents has increased significantly. Although the incidence of intravenous contrast agent extravasation is less than 1%, it can lead to serious complications. In prolonged infusion scenarios such as surgery and monitoring, the risks of extravasation are even more severe.

[0003] Currently, research on extravasation warning during injection still has some shortcomings. Specifically, traditional infusion pump occlusion alarms are neither sensitive nor clear in detecting extravasation, and existing extravasation detection devices suffer from problems such as cumbersome operation, difficulty in detecting early extravasation, or susceptibility to false alarms. Current extravasation detection mainly relies on manual observation of skin surface signs, which has low accuracy. Early extravasation detection devices may have technical limitations, and non-contact optical detection equipment may be affected by skin pigmentation, ambient light, etc., leading to poor detection stability. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an extravasation warning system and method for injection processes based on sensors and intelligent identification, which can effectively solve the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of the present invention provides an extravasation warning system during injection based on sensors and intelligent recognition, comprising: an individualized initial monitoring threshold determination module, used to collect basic information of the monitored object and information of the injected drug, and combine this information with a drug characteristic database and historical data of the monitored object to determine the individualized initial monitoring threshold parameters for the monitored object; an intelligent monitoring patch index analysis module, used to collect intelligent monitoring patch data of the monitored object, and combine this data with the individualized initial monitoring threshold parameters of the monitored object to analyze and obtain intelligent monitoring patch indicators; and an infusion pump linkage index analysis module, used to collect data of the monitored object... The system analyzes the infusion pump linkage data, combined with the individualized initial monitoring threshold parameters of the monitored object, to obtain the infusion pump linkage index. The extravasation risk analysis module analyzes the extravasation risk during injection by combining the intelligent monitoring patch index and the infusion pump linkage index, and outputs the extravasation risk assessment result. The multi-level alarm and linkage intervention response module determines the multi-level alarm and linkage intervention response plan based on the extravasation risk assessment result. The closed-loop tracking module for the injection process archives the intelligent monitoring patch data, the infusion pump linkage data, and the multi-level alarm and linkage intervention response plan for extravasation, achieving closed-loop tracking of the injection process.

[0006] As a further step, basic information and injected drug information of the monitored subjects are collected. Combined with a drug characteristic database and historical data of the monitored subjects, individualized initial monitoring threshold parameters are determined. The specific analysis process is as follows: Basic information of the monitored subjects is collected, including age, weight, gender, and current injection site. Based on the drug characteristic database, injected drug information is collected, including drug name, drug concentration, recommended injection pressure range, and recommended injection speed range. Based on historical data, the number of extravasations in the monitored subjects' historical injection history is extracted. The monitored subjects' age, weight, gender, current injection site, number of extravasations in their historical injection history, drug name, drug concentration, recommended injection pressure range, and recommended injection speed range are stored as individualized initial monitoring tags. The individualized initial monitoring tag-monitoring subject individualized initial monitoring threshold parameter mapping set stored in the database is obtained. Based on the current individualized initial monitoring tag, the corresponding individualized initial monitoring threshold parameters are matched. The individualized initial monitoring threshold parameters specifically include skin temperature threshold, electrical impedance threshold, subcutaneous tissue strain threshold, injection pressure threshold, and injection speed threshold during the injection process.

[0007] As a further step, data from the intelligent monitoring patch is collected from the monitored subjects. Combined with the individualized initial monitoring threshold parameters of each subject, the intelligent monitoring patch indicators are analyzed. The specific analysis process is as follows: an infrared thermal sensor continuously monitors the skin temperature at the injection site; a bioelectrical impedance sensor detects fluctuations in local impedance due to changes in exudate; and a micro-strain sensor reflects changes in subcutaneous tissue tension in real time. The intelligent monitoring patch data collected from the monitored subjects includes real-time skin temperature, real-time impedance, and real-time subcutaneous tissue strain. Based on the intelligent monitoring patch data and the individualized initial monitoring threshold parameters of each subject, the intelligent monitoring patch indicators are analyzed and used as the basis for analyzing the risk of extravasation during injection.

[0008] As a further solution, intelligent monitoring of patch indicators is implemented, and the specific analysis process is as follows:

[0009]

[0010] Wherein, jt is the intelligent monitoring patch index, pt is the real-time skin temperature, dzk is the real-time electrical impedance, zb is the real-time subcutaneous tissue strain, pt0 is the skin temperature threshold, dzk0 is the electrical impedance threshold, zb0 is the subcutaneous tissue strain threshold, a1 is the weighting factor of pt stored in the database, a2 is the weighting factor of dzk stored in the database, and a3 is the weighting factor of zb stored in the database.

[0011] As a further approach, data on the infusion pump linkage of the monitored subjects is collected and analyzed in conjunction with the individualized initial monitoring threshold parameters of the monitored subjects to obtain infusion pump linkage indicators. The specific analysis process is as follows: data on the infusion pump linkage of the monitored subjects is collected, including the injection pressure and real-time injection speed during the injection process; based on the data on the infusion pump linkage of the monitored subjects, combined with the individualized initial monitoring threshold parameters of the monitored subjects, the infusion pump linkage indicators are analyzed to obtain infusion pump linkage indicators, which serve as the basis for analyzing the risk of extravasation during the injection process.

[0012] As a further solution, the analysis process for the infusion pump linkage index is as follows:

[0013]

[0014] Where bl is the infusion pump linkage index, zy is the injection pressure during real-time injection, zv is the real-time injection speed, zy0 is the injection pressure threshold during injection, zv0 is the injection speed threshold, b1 is the weight factor of zy stored in the database, and b2 is the weight factor of zv stored in the database.

[0015] As a further approach, combining intelligent monitoring patch indicators and infusion pump linkage indicators, the extravasation risk during injection is analyzed, and the extravasation risk assessment result is output. The specific analysis process is as follows: the intelligent monitoring patch indicators and infusion pump linkage indicators are used as inputs to a trained fuzzy comprehensive evaluation model; the fuzzy comprehensive evaluation factor set is determined, specifically including the intelligent monitoring patch indicators and infusion pump linkage indicators; the extravasation risk level is determined, specifically including low risk level, medium risk level, and high risk level; the membership degree of each fuzzy comprehensive evaluation factor to each extravasation risk level is obtained from the database, resulting in a single-factor evaluation matrix; the weight of each fuzzy comprehensive evaluation factor is determined using the analytic hierarchy process (AHP), resulting in a fuzzy comprehensive evaluation factor weight vector; the single-factor evaluation matrix and the fuzzy comprehensive evaluation factor weight vector are input; the comprehensive evaluation vector is calculated through fuzzy transformation; the comprehensive evaluation vector is output, obtaining the comprehensive membership degree of the fuzzy comprehensive evaluation factor to each extravasation risk level; according to the principle of maximum membership degree, the extravasation risk level with the largest membership degree in the comprehensive evaluation vector is selected as the final extravasation risk level, and the output is used as the extravasation risk assessment result.

[0016] As a further solution, based on the extravasation risk assessment results, a multi-level extravasation alarm and linkage intervention response plan is determined. The specific analysis process is as follows: The extravasation risk level-multi-level extravasation alarm and linkage intervention response plan mapping table stored in the database is obtained. Based on the current extravasation risk level, the corresponding multi-level extravasation alarm and linkage intervention response plan is determined. The multi-level extravasation alarm and linkage intervention response plan specifically includes issuing an audible and visual alarm signal, obtaining the flashing indicator light and buzzer notification frequency, and simultaneously pushing specific notification information to the nurse station terminal, including the patient's bed number, injection site, alarm time, and extravasation risk level. Adjustment commands are sent to the infusion pump linkage control module to obtain infusion pump pressure adjustment parameters, infusion pump speed adjustment parameters, and infusion pump stoppage duration.

[0017] As a further solution, data from the intelligent monitoring patch, the infusion pump, and the multi-level extravasation alarm and intervention response plan are archived to achieve closed-loop tracking of the injection process. The specific analysis process is as follows: the subject's name, gender, and date of birth are integrated into a unique identifier; using this unique identifier and timestamp, the intelligent monitoring patch data, infusion pump data, and the multi-level extravasation alarm and intervention response plan are linked to the subject's historical data, forming a complete data chain centered on the subject for data archiving; the data archiving and quality control interface establishes a real-time synchronization interface with the hospital information system and electronic medical record system through a standardized medical data interaction protocol; the system automatically marks key event nodes in the data chain, including the injection start node, the multi-level extravasation alarm and intervention response plan trigger node, the multi-level extravasation alarm and intervention response plan execution node, and the injection end node, and connects the data of each key event node in a timeline format to form a complete injection process event chain.

[0018] The second aspect of this invention provides a sensor- and intelligent identification-based extravasation early warning method during injection, comprising the following steps: collecting basic information of the monitored object and information of the injected drug, and combining the drug characteristic database and historical data of the monitored object to determine individualized initial monitoring threshold parameters for the monitored object; collecting intelligent monitoring patch data of the monitored object, and combining the individualized initial monitoring threshold parameters of the monitored object to analyze and obtain intelligent monitoring patch indicators; collecting infusion pump linkage data of the monitored object, and combining the individualized initial monitoring threshold parameters of the monitored object to analyze and obtain infusion pump linkage indicators; combining the intelligent monitoring patch indicators and the infusion pump linkage indicators to analyze the extravasation risk during injection and output the extravasation risk assessment result; based on the extravasation risk assessment result, determining a multi-level alarm and linkage intervention response scheme for extravasation; and archiving the intelligent monitoring patch data, the infusion pump linkage data, and the multi-level alarm and linkage intervention response scheme for extravasation to achieve closed-loop tracking of the injection process.

[0019] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0020] This invention provides an extravasation warning system and method for injection processes based on sensors and intelligent recognition. Attached Figure Description

[0021] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the system module connections of the present invention.

[0023] Figure 2 This is a schematic diagram of the method steps of the present invention. Detailed Implementation

[0024] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0025] Reference Figure 1 As shown, the first aspect of the present invention provides an extravasation early warning system for injection based on sensors and intelligent identification, including: an individualized initial monitoring threshold determination module, an intelligent monitoring patch index analysis module, an infusion pump linkage index analysis module, an extravasation risk analysis module during injection, a multi-level alarm and linkage intervention response module, and an injection process closed-loop tracking module.

[0026] The individualized initial monitoring threshold determination module is used to collect basic information and injected drug information of the monitored object, and combine it with the drug characteristic database and historical data of the monitored object to determine the individualized initial monitoring threshold parameters of the monitored object.

[0027] The specific analysis process is as follows: Basic information of the monitored subjects is collected, including their age, weight, gender, and current injection site. Based on a drug characteristic database, information on the injected drugs is collected, including drug name, drug concentration, recommended injection pressure range, and recommended injection speed range. Based on historical data, the number of extravasations in the monitored subjects' historical injection history is extracted. The monitored subjects' age, weight, gender, current injection site, number of extravasations in their historical injection history, drug name, drug concentration, recommended injection pressure range, and recommended injection speed range are stored as individualized initial monitoring tags. The individualized initial monitoring tag-monitoring subject individualized initial monitoring threshold parameter mapping set stored in the database is obtained. Based on the current individualized initial monitoring tag, the corresponding individualized initial monitoring threshold parameters for the monitored subjects are matched. The individualized initial monitoring threshold parameters for the monitored subjects specifically include skin temperature threshold, electrical impedance threshold, subcutaneous tissue strain threshold, injection pressure threshold, and injection speed threshold during the injection process.

[0028] By comprehensively collecting basic information such as age, weight, gender, and injection site of the monitored subjects, and combining this with drug name, concentration, recommended injection pressure, and rate range from the drug characteristic database, and incorporating personalized data such as the number of historical extravasation injections, a unique individualized initial monitoring label is constructed. Using the label-threshold parameter mapping set stored in the database, individualized initial monitoring threshold parameters, including skin temperature, electrical impedance, subcutaneous tissue stress, injection pressure, and rate, are accurately matched. This breaks through the limitations of traditional uniform thresholds, fully considering the physiological differences of different monitored subjects, the characteristics of injected drugs, and individual differences in historical injection risks. This makes the monitoring thresholds more closely aligned with each subject's actual situation, effectively improving the targeting and accuracy of extravasation early warning, reducing false alarms or missed alarms due to individual differences, and providing a scientific and personalized judgment benchmark for subsequent extravasation monitoring based on sensors and intelligent identification. This allows for more accurate detection of early signs of extravasation, buying valuable time for timely intervention.

[0029] The intelligent monitoring patch index analysis module is used to collect intelligent monitoring patch data of the monitored object, and combine it with the individualized initial monitoring threshold parameters of the monitored object to analyze and obtain the intelligent monitoring patch index.

[0030] The specific analysis process is as follows: an infrared thermal sensor continuously monitors the skin temperature at the injection site; a bioelectrical impedance sensor detects fluctuations in local electrical impedance as exudate changes; and a micro-strain sensor reflects changes in subcutaneous tissue tension in real time. Data from the intelligent monitoring patch is collected, specifically including real-time skin temperature, real-time electrical impedance, and real-time subcutaneous tissue strain. Based on the intelligent monitoring patch data and combined with the individualized initial monitoring threshold parameters of the monitored subject, intelligent monitoring patch indicators are analyzed and used as the basis for analyzing the risk of extravasation during the injection process.

[0031] The intelligent monitoring of patch indicators and the specific analysis process are as follows:

[0032]

[0033] Wherein, jt is the intelligent monitoring patch index, pt is the real-time skin temperature, dzk is the real-time electrical impedance, zb is the real-time subcutaneous tissue strain, pt0 is the skin temperature threshold, dzk0 is the electrical impedance threshold, zb0 is the subcutaneous tissue strain threshold, a1 is the weighting factor of pt stored in the database, a2 is the weighting factor of dzk stored in the database, and a3 is the weighting factor of zb stored in the database.

[0034] Utilizing multiple sensors such as infrared thermometry, bioelectrical impedance analysis, and micro-strain analysis, real-time skin temperature, electrical impedance, and subcutaneous tissue strain at the injection site are precisely collected. Combined with pre-determined individualized initial monitoring threshold parameters, a formula incorporating weighting factors for each parameter is used to calculate the intelligent monitoring patch index. By integrating multi-dimensional data on skin temperature, electrical impedance, and subcutaneous tissue strain, it comprehensively and dynamically reflects changes in the physiological state of the injection site. As a basis for extravasation risk analysis, it can accurately capture abnormalities in skin temperature, electrical impedance, and tissue strain caused by extravasation, enabling scientific and quantitative monitoring of injection extravasation risk, improving the accuracy and timeliness of extravasation warnings, and providing more reliable technical support for injection safety.

[0035] The infusion pump linkage index analysis module is used to collect infusion pump linkage data of the monitored objects and analyze the infusion pump linkage index by combining the individualized initial monitoring threshold parameters of the monitored objects.

[0036] The specific analysis process is as follows: Collect the infusion pump linkage data of the monitored object, including the injection pressure and real-time injection speed during the injection process; Based on the infusion pump linkage data of the monitored object, combined with the individualized initial monitoring threshold parameters of the monitored object, analyze and obtain the infusion pump linkage index, which serves as the basis for analyzing the risk of extravasation during the injection process.

[0037] The specific analysis process for the infusion pump linkage index is as follows:

[0038]

[0039] Where bl is the infusion pump linkage index, zy is the injection pressure during real-time injection, zv is the real-time injection speed, zy0 is the injection pressure threshold during injection, zv0 is the injection speed threshold, b1 is the weight factor of zy stored in the database, and b2 is the weight factor of zv stored in the database.

[0040] By collecting real-time pump-linked data such as injection pressure and rate, and combining this data with individualized initial monitoring threshold parameters, the pump linkage index is calculated using a formula. This comprehensively considers the relationship between real-time pressure and rate values ​​and thresholds, and assigns appropriate weights to different parameters using weighting factors. This allows for dynamic and accurate reflection of abnormal pressure and rate changes that may occur during injection due to extravasation, from the perspective of pump operation. It provides pump-side data support for extravasation risk analysis, and works in conjunction with intelligent monitoring patch indicators to cover injection scenarios from multiple dimensions, improving the comprehensiveness and accuracy of extravasation risk identification and helping to detect potential extravasation risks earlier.

[0041] The extravasation risk analysis module during injection combines intelligent monitoring patch indicators and infusion pump linkage indicators to analyze the risk of extravasation during injection and output the extravasation risk assessment results.

[0042] The specific analysis process is as follows: The intelligent monitoring patch indicators and the injection pump linkage indicators are used as inputs to the trained fuzzy comprehensive evaluation model; the fuzzy comprehensive evaluation factor set is determined, specifically including the intelligent monitoring patch indicators and the injection pump linkage indicators; the extravasation risk levels are determined, specifically including low-risk, medium-risk, and high-risk levels; the membership degrees of each fuzzy comprehensive evaluation factor to each extravasation risk level are obtained from the database, resulting in a single-factor evaluation matrix; the weights of each fuzzy comprehensive evaluation factor are determined using the analytic hierarchy process (AHP), resulting in a fuzzy comprehensive evaluation factor weight vector; the single-factor evaluation matrix and the fuzzy comprehensive evaluation factor weight vector are input; the comprehensive evaluation vector is calculated through fuzzy transformation; the comprehensive evaluation vector is output, obtaining the comprehensive membership degree of the fuzzy comprehensive evaluation factor to each extravasation risk level; based on the principle of maximum membership degree, the extravasation risk level with the highest membership degree in the comprehensive evaluation vector is selected as the final extravasation risk level, and the output is used as the extravasation risk determination result.

[0043] By leveraging the intelligent monitoring patch indicators and the infusion pump linkage indicators, and employing a trained fuzzy comprehensive evaluation model, the factor set encompassing both types of indicators and the low, medium, and high extravasation risk levels are first identified. A single-factor evaluation matrix is ​​constructed by obtaining membership degrees, and the indicator weights are determined using the analytic hierarchy process (AHP). A comprehensive evaluation vector is calculated via fuzzy transformation to derive the comprehensive membership degree for each risk level. Finally, the final risk level is determined based on the principle of maximum membership degree. By integrating multi-dimensional monitoring data and utilizing the advantages of fuzzy comprehensive evaluation in handling uncertainty and complexity, this approach enables comprehensive and accurate analysis of extravasation risk, scientifically classifying risk levels, and providing a reliable basis for medical personnel to promptly and accurately assess extravasation and take corresponding intervention measures. This effectively improves the scientific rigor and effectiveness of injection extravasation risk identification and management.

[0044] The multi-level alarm and linkage intervention response module is used to determine the multi-level alarm and linkage intervention response plan for leakage based on the leakage risk assessment results.

[0045] The specific analysis process is as follows: Obtain the mapping table of extravasation risk level and multi-level extravasation alarm and linkage intervention response scheme stored in the database; based on the current extravasation risk level, determine the corresponding multi-level extravasation alarm and linkage intervention response scheme; the multi-level extravasation alarm and linkage intervention response scheme specifically includes issuing an audible and visual alarm signal, obtaining the flashing indicator light and buzzer prompt frequency, simultaneously pushing specific prompt information to the nurse station terminal, including the patient's bed number, injection site, alarm time, and extravasation risk level, and sending adjustment instructions to the infusion pump linkage control module to obtain the infusion pump pressure adjustment parameters, infusion pump speed adjustment parameters, and infusion pump stop running duration.

[0046] Based on the extravasation risk assessment results, and utilizing a mapping table between risk levels and response plans, corresponding multi-level alarms and coordinated intervention measures are accurately matched. By issuing differentiated audible and visual alarms (such as different flashing and beeping frequencies), prompts containing detailed information such as the patient's bed number and injection site are pushed to the nurse station terminal. It can also coordinate with the infusion pump to adjust pressure, speed, or pause operation, achieving efficient connection from risk identification to alarm and intervention. This provides clear guidance for medical staff to quickly locate and handle emergencies, and enables timely adjustment of infusion pump operating parameters, reducing extravasation hazards from multiple dimensions and improving the timeliness and accuracy of injection safety and emergency response.

[0047] The injection process closed-loop tracking module is used to archive data from the intelligent monitoring patch of the monitored object, the linkage data of the infusion pump of the monitored object, and the multi-level alarm and linkage intervention response plan for extravasation, so as to realize closed-loop tracking of the injection process.

[0048] The specific analysis process is as follows: The name, gender, and date of birth of the monitored subject are integrated into a unique identifier for the monitored subject; through the unique identifier and timestamp, the intelligent monitoring patch data, the infusion pump linkage data, the extravasation multi-level alarm and linkage intervention response plan, and the historical data of the monitored subject are associated to form a complete data chain centered on the monitored subject for data archiving; the data archiving and quality control interface establishes a real-time synchronization interface with the hospital information system and electronic medical record system through a standardized medical data interaction protocol; the system automatically marks key event nodes in the data chain, including the injection start node, the extravasation multi-level alarm and linkage intervention response plan trigger node, the extravasation multi-level alarm and linkage intervention response plan execution node, and the injection end node, and the data of each key event node are linked together in the form of a timeline to form a complete injection process event chain.

[0049] By integrating monitoring object identifiers, associating and archiving multi-dimensional data, and synchronizing with hospital systems in real time using standardized protocols, key event nodes are also marked to form an event chain. This not only builds a complete and traceable injection data chain for each monitored object, facilitating medical staff to review and analyze the entire injection process and assist in optimizing treatment plans; it also enables efficient interaction with the hospital's existing information systems, helping the hospital integrate and manage data; at the same time, the clear presentation of the event chain makes key nodes in the injection process readily apparent, facilitating timely review and quality control, ensuring injection safety and quality, and promoting the formation of a complete closed loop of "monitoring-intervention-traceability" in injection management.

[0050] Reference Figure 2 As shown, the second aspect of the present invention provides a method for early warning of extravasation during injection based on sensors and intelligent identification, including the following steps: collecting basic information of the monitored object and information of the injected drug, and combining the drug characteristic database and historical data of the monitored object to determine the individualized initial monitoring threshold parameter of the monitored object.

[0051] Data from the intelligent monitoring patch on the monitored object is collected and analyzed to obtain the intelligent monitoring patch index, combined with the individualized initial monitoring threshold parameters of the monitored object.

[0052] Data on the linkage of the infusion pumps of the monitored subjects were collected and analyzed to obtain the linkage index of the infusion pumps, combined with the individualized initial monitoring threshold parameters of the monitored subjects.

[0053] By combining intelligent monitoring patch indicators and infusion pump linkage indicators, the risk of extravasation during injection is analyzed, and the extravasation risk assessment results are output.

[0054] Based on the risk assessment results of leakage, a multi-level alarm and linkage intervention response plan for leakage was determined.

[0055] Data archives are created for the intelligent monitoring patch data of the monitored object, the linkage data of the infusion pump of the monitored object, and the multi-level alarm and linkage intervention response plan for extravasation, so as to realize closed-loop tracking of the injection process.

[0056] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

Claims

1. An extravasation early warning system during injection based on sensors and intelligent recognition, characterized in that, include: The individualized initial monitoring threshold determination module is used to collect basic information and injected drug information of the monitored object, and combine them with the drug characteristic database and historical data of the monitored object to determine the individualized initial monitoring threshold parameters of the monitored object. The intelligent monitoring patch index analysis module is used to collect intelligent monitoring patch data of the monitored object, and analyze the intelligent monitoring patch index by combining the individualized initial monitoring threshold parameters of the monitored object. The infusion pump linkage index analysis module is used to collect infusion pump linkage data of the monitored objects and analyze the infusion pump linkage index by combining the individualized initial monitoring threshold parameters of the monitored objects. The extravasation risk analysis module during injection is used to analyze the risk of extravasation during injection by combining intelligent monitoring patch indicators and injection pump linkage indicators, and output the extravasation risk assessment results. The multi-level alarm and linkage intervention response module is used to determine the multi-level alarm and linkage intervention response plan for leakage based on the leakage risk assessment results. The injection process closed-loop tracking module is used to archive data from the intelligent monitoring patch of the monitored object, the linkage data of the infusion pump of the monitored object, and the multi-level alarm and linkage intervention response plan for extravasation, so as to realize closed-loop tracking of the injection process.

2. The extravasation early warning system during injection based on sensors and intelligent recognition according to claim 1, characterized in that: The basic information of the monitored subjects and the information of the injected drugs are collected, and combined with the drug characteristic database and the historical data of the monitored subjects, the individualized initial monitoring threshold parameters of the monitored subjects are determined. The specific analysis process is as follows: Collect basic information about the monitored subjects, including their age, weight, gender, and injection site. Based on a drug characteristic database, information on the injectable drugs of the monitored subjects is collected, including drug name, drug concentration, recommended injection pressure range, and recommended injection speed range. Based on the historical data of the monitored subjects, extract the number of extravasations during the historical injection history of the monitored subjects; The monitoring subject's age, weight, gender, current injection site, number of extravasations in the subject's previous injection history, drug name, drug concentration, recommended injection pressure range, and recommended injection rate range are stored as individualized initial monitoring labels. Retrieve the individualized initial monitoring tag-monitoring object individualized initial monitoring threshold parameter mapping set stored in the database, and match the corresponding individualized initial monitoring threshold parameter of the monitoring object based on the current individualized initial monitoring tag; The individualized initial monitoring threshold parameters for the monitored subjects specifically include skin temperature threshold, electrical impedance threshold, subcutaneous tissue strain threshold, injection pressure threshold, and injection speed threshold during the injection process.

3. The extravasation early warning system during injection based on sensors and intelligent recognition according to claim 1, characterized in that: The data from the intelligent monitoring patch of the monitored object is collected and combined with the individualized initial monitoring threshold parameters of the monitored object to analyze and obtain the intelligent monitoring patch index. The specific analysis process is as follows: Infrared thermal sensors continuously monitor skin temperature at the injection site, bioelectrical impedance sensors detect fluctuations in local impedance as exudate changes, and micro-strain sensors reflect changes in subcutaneous tissue tension in real time. Collects data from the intelligent monitoring patch on the monitored object, specifically including real-time skin temperature, real-time electrical impedance, and real-time subcutaneous tissue strain. Based on the intelligent monitoring patch data of the monitored objects, combined with the individualized initial monitoring threshold parameters of the monitored objects, the intelligent monitoring patch index is obtained. The intelligent monitoring patch index serves as the basis for analyzing the risk of extravasation during the injection process.

4. The extravasation early warning system during injection based on sensors and intelligent recognition according to claim 3, characterized in that: The specific analysis process for the intelligent monitoring patch indicators is as follows: Wherein, jt is the intelligent monitoring patch index, pt is the real-time skin temperature, dzk is the real-time electrical impedance, zb is the real-time subcutaneous tissue strain, pt0 is the skin temperature threshold, dzk0 is the electrical impedance threshold, zb0 is the subcutaneous tissue strain threshold, a1 is the weighting factor of pt stored in the database, a2 is the weighting factor of dzk stored in the database, and a3 is the weighting factor of zb stored in the database.

5. The extravasation early warning system during injection based on sensors and intelligent recognition according to claim 1, characterized in that: The collected data on the infusion pump linkage of the monitored objects, combined with the individualized initial monitoring threshold parameters of the monitored objects, are used to analyze the infusion pump linkage index. The specific analysis process is as follows: Collect and monitor data on the infusion pump linkage of the monitored object, specifically including the injection pressure and real-time injection speed during the injection process; Based on the infusion pump linkage data of the monitored subjects, and combined with the individualized initial monitoring threshold parameters of the monitored subjects, the infusion pump linkage index is obtained. The infusion pump linkage index serves as the basis for analyzing the risk of extravasation during the injection process.

6. The extravasation early warning system during injection based on sensors and intelligent recognition according to claim 5, characterized in that: The specific analysis process for the infusion pump linkage index is as follows: Where bl is the infusion pump linkage index, zy is the injection pressure during real-time injection, zv is the real-time injection speed, zy0 is the injection pressure threshold during injection, zv0 is the injection speed threshold, b1 is the weight factor of zy stored in the database, and b2 is the weight factor of zv stored in the database.

7. The extravasation early warning system during injection based on sensors and intelligent recognition according to claim 1, characterized in that: The method combines intelligent monitoring patch indicators and infusion pump linkage indicators to analyze the risk of extravasation during injection and outputs the extravasation risk assessment result. The specific analysis process is as follows: The intelligent monitoring patch index and the injection pump linkage index are used as inputs to the trained fuzzy comprehensive evaluation model; The set of fuzzy comprehensive evaluation factors was determined, specifically including intelligent monitoring patch indicators and injection pump linkage indicators; Determine the risk level of infiltration, specifically including low risk level, medium risk level, and high risk level; Obtain the membership degree of each fuzzy comprehensive evaluation factor stored in the database to each extravasation risk level to obtain the single-factor evaluation matrix; The weights of each fuzzy comprehensive evaluation factor are determined using the analytic hierarchy process (AHP), resulting in a weight vector for the fuzzy comprehensive evaluation factors. Input the single-factor evaluation matrix and the fuzzy comprehensive evaluation factor weight vector; The comprehensive evaluation vector is calculated using fuzzy transformation. Output the comprehensive evaluation vector to obtain the comprehensive membership degree of the fuzzy comprehensive evaluation factors to each extravasation risk level; Based on the principle of maximum membership, the infiltration risk level with the highest membership in the comprehensive evaluation vector is selected as the final infiltration risk level, and the output is used as the infiltration risk determination result.

8. The extravasation early warning system during injection based on sensors and intelligent recognition according to claim 1, characterized in that: Based on the risk assessment results of extravasation, a multi-level alarm and linkage intervention response scheme for extravasation was determined. The specific analysis process is as follows: Obtain the mapping table of extravasation risk level - extravasation multi-level alarm and linkage intervention response scheme stored in the database, and determine the corresponding extravasation multi-level alarm and linkage intervention response scheme based on the current extravasation risk level; The extravasation multi-level alarm and linkage intervention response scheme specifically includes issuing audible and visual alarm signals, obtaining the frequency of indicator light flashing and buzzer prompts, and simultaneously pushing specific prompt information to the nurse station terminal, including the patient's bed number, injection site, alarm time, and extravasation risk level. It also sends adjustment commands to the infusion pump linkage control module to obtain infusion pump pressure adjustment parameters, infusion pump speed adjustment parameters, and infusion pump stop running duration.

9. The extravasation early warning system during injection based on sensors and intelligent recognition according to claim 1, characterized in that: The data archives are performed on the intelligent monitoring patch data of the monitored object, the linkage data of the monitoring object's infusion pump, and the multi-level alarm and linkage intervention response plan for extravasation, to achieve closed-loop tracking of the injection process. The specific analysis process is as follows: The name, gender, and date of birth of the monitored subject will be integrated into a unique identifier for the monitored subject. By monitoring unique identifiers and timestamps, the data of intelligent monitoring patches, the data of injection pumps, the multi-level alarm and linkage intervention response plan for extravasation, and the historical data of the monitored objects are linked to form a complete data chain centered on the monitored objects for data archiving. The data archiving and quality control interface establishes a real-time synchronization interface with the hospital information system and electronic medical record system through a standardized medical data interaction protocol. The system automatically marks key event nodes in the data chain, including the injection start node, the trigger node of the extravasation multi-level alarm and linkage intervention response plan, the execution node of the extravasation multi-level alarm and linkage intervention response plan, and the injection end node. The data of each key event node are linked together in the form of a time axis to form a complete injection process event chain.

10. A sensor- and intelligent recognition-based extravasation warning method for injection, applied to the sensor- and intelligent recognition-based extravasation warning system for injection as described in any one of claims 1-9, characterized in that, Includes the following steps: Collect basic information on the monitored subjects and information on injected drugs, and combine this information with a drug characteristic database and historical data of the monitored subjects to determine individualized initial monitoring threshold parameters for the monitored subjects; Collect data from the intelligent monitoring patch of the monitored object, and combine it with the individualized initial monitoring threshold parameters of the monitored object to analyze and obtain the intelligent monitoring patch index; Collect data on the linkage of the infusion pumps of the monitored subjects, and combine this data with the individualized initial monitoring threshold parameters of the monitored subjects to obtain the linkage index of the infusion pumps. By combining intelligent monitoring patch indicators and infusion pump linkage indicators, the risk of extravasation during injection is analyzed, and the extravasation risk assessment results are output. Based on the results of the extravasation risk assessment, a multi-level alarm and linkage intervention response plan for extravasation was determined. Data archives are created for the intelligent monitoring patch data of the monitored object, the linkage data of the infusion pump of the monitored object, and the multi-level alarm and linkage intervention response plan for extravasation, so as to realize closed-loop tracking of the injection process.

Images