A method and system for tracking information of static therapy based on hospital static therapy

Abstract

The present invention discloses an information-based tracking method and system for intravenous therapy based on hospital intravenous therapy, which relates to the technical field of intravenous therapy information tracking, and includes the following steps: the medical staff first sets a starting intravenous infusion flow rate for the patient according to the patient's condition, weight, age, characteristics of the drug and treatment goals; during the patient's intravenous infusion, the patient's heart rate information is obtained in real time using a monitoring device; the key features are extracted after pre-processing the obtained original heart rate data, and the extracted heart rate features are analyzed for abnormalities under a short-term monitoring window. The present invention dynamically adjusts the intravenous infusion flow rate by obtaining the patient's heart rate information in real time and combining it with a machine learning model to achieve personalized management. When the heart rate is abnormal, the system automatically lowers the flow rate to reduce the heart burden and improve safety; at a normal heart rate, the flow rate is increased to speed up the infusion efficiency. The intelligent adjustment mechanism not only improves the treatment effect, but also reduces the workload of medical staff and improves nursing efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of intravenous therapy information tracking, and in particular to an intravenous therapy information tracking method and system based on hospital intravenous therapy. Background Art

[0002] The hospital intravenous therapy information tracking system is a management system for the entire intravenous therapy process achieved through Internet of Things technology and smart devices. The system integrates information such as intravenous therapy equipment, patient data, treatment plans, and drug infusion records. It can monitor the intravenous infusion in real time, track the flow rate, time, and dosage of drug infusion, and dynamically adjust the infusion plan according to the patient's health status. It generates treatment records through data collection and analysis and provides real-time feedback to medical staff to ensure the accuracy and safety of treatment, while improving nursing work efficiency and reducing the occurrence of human errors and infusion complications.

[0003] During the intravenous infusion process, smart sensors will continuously collect data from the infusion system and transmit this data to the central monitoring system via wireless communications (such as Wi-Fi, Bluetooth or a dedicated hospital network) to monitor and track the patient's intravenous infusion process in real time.

[0004] The prior art has the following deficiencies:

[0005] In the prior art, the monitoring and tracking of the intravenous infusion process usually sets a fixed infusion flow rate according to the doctor's order. Medical staff will determine the appropriate flow rate based on the patient's condition and the characteristics of the drug to ensure that the drug enters the patient's body safely and effectively within the specified time. However, if the patient's heart rate undergoes potential abnormal changes due to the excessively fast infusion flow rate during the infusion process, but it is not monitored in real time, continuing to maintain a fixed flow rate may lead to serious consequences. Especially for patients with heart failure, too fast an infusion rate will significantly increase the burden on the heart, resulting in excessive blood volume, and the heart cannot effectively pump blood, thereby inducing acute heart failure. Heart failure will further cause fluid retention in the lungs, leading to acute pulmonary edema, dyspnea and hypoxia, and ultimately endangering the patient's life.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0007] The purpose of the present invention is to provide an information-based tracking method and system for intravenous therapy based on hospital intravenous therapy. By acquiring the patient's heart rate information in real time and combining it with a machine learning model, the heart rate changes are dynamically evaluated, and the intravenous infusion flow rate is flexibly adjusted to achieve personalized management. When an abnormal heart rate is detected, the system will automatically lower the infusion flow rate to reduce the burden on the heart and prevent serious complications. For conventional heart rate fluctuations, the system can increase the flow rate and improve the infusion efficiency. It not only optimizes the infusion process and shortens the treatment time, but also reduces the workload of medical staff through automated monitoring and alarms, improves the overall nursing efficiency, and has a significant safety protection effect on patients with heart failure, in particular, to solve the problems in the above-mentioned background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solution: a method for tracking intravenous therapy information based on hospital intravenous therapy, comprising the following steps:

[0009] Medical staff first set a starting intravenous infusion flow rate for the patient based on the patient's condition, weight, age, characteristics of the medication, and treatment goals;

[0010] During the patient's intravenous infusion, use monitoring equipment to obtain the patient's heart rate information in real time;

[0011] After preprocessing the acquired raw heart rate data, key features are extracted. Under a short-term monitoring window, abnormal analysis is performed on the extracted heart rate features. Subsequently, a pre-trained machine learning model is used to comprehensively evaluate and predict the analyzed heart rate features.

[0012] Based on the prediction results of the machine learning model, the patient's heart rate changes in the short-term monitoring window are divided into potential abnormal changes and regular changes;

[0013] If the heart rate changes are classified as normal changes, the patient's actual infusion rate is increased within the safe intravenous infusion rate range;

[0014] If the heart rate change is identified as a potential abnormal change, the initial intravenous infusion flow rate will be lowered according to the degree of abnormality, and the patient's heart rate changes will continue to be observed to ensure that the patient is safely infused when the heart rate changes normally. If the heart rate still does not return to normal, an alarm will be issued to notify medical staff.

[0015] Preferably, the steps for setting the initial intravenous infusion flow rate are as follows:

[0016] Step 1: Before setting the initial infusion flow rate, medical staff should first conduct a comprehensive assessment of the patient's overall physical condition;

[0017] Step 2: further refine the setting of the initial flow rate according to the characteristics of the drug used and the treatment goal;

[0018] Step 3: Based on the patient's physiological assessment and drug characteristics, the healthcare provider sets the initial intravenous infusion flow rate according to clinical practice guidelines or their clinical experience.

[0019] Preferably, the short-term monitoring window refers to the continuous collection and analysis of heart rate data within a fixed short period of time to capture instantaneous changes and short-term fluctuations in heart rate and identify potential abnormalities in a timely manner.

[0020] Preferably, the acquired raw heart rate data is preprocessed to extract key features, wherein the extracted key features include the rapid change amplitude of the heart rate and the change law of the heartbeat interval time, and after abnormal analysis of the rapid change amplitude of the heart rate and the change law of the heartbeat interval time in a short-term monitoring window, an instantaneous heart rate fluctuation reference value and a heart rate irregularity reference value are generated respectively, the rapid fluctuation amplitude of the heart rate in a short period of time is quantified by the instantaneous heart rate fluctuation reference value, which is used to evaluate whether the instantaneous change of the heart rate exceeds the normal range, and the change law of the heartbeat interval time is quantified by the heart rate irregularity reference value, which is used to evaluate the irregularity of the heartbeat interval;

[0021] After obtaining the instantaneous heart rate fluctuation reference value and heart rate irregularity reference value generated by abnormal analysis during intravenous infusion, the instantaneous heart rate fluctuation reference value and the heart rate irregularity reference value are used to generate a heart rate potential abnormality coefficient through a pre-trained machine learning model, and the heart rate condition of the patient during intravenous infusion is comprehensively evaluated and predicted through the heart rate potential abnormality coefficient.

[0022] Preferably, the heart rate potential abnormality coefficient generated when the patient's heart rate condition during intravenous infusion is comprehensively evaluated and predicted by the machine learning model is compared and analyzed with the pre-set heart rate potential abnormality coefficient reference threshold, and the patient's heart rate changes in the short-term monitoring window are divided. The specific division steps are: if the heart rate potential abnormality coefficient is greater than the heart rate potential abnormality coefficient reference threshold, the patient's heart rate changes in the short-term monitoring window are divided into potential abnormal changes; if the heart rate potential abnormality coefficient is less than or equal to the heart rate potential abnormality coefficient reference threshold, the patient's heart rate changes in the short-term monitoring window are divided into normal changes.

[0023] Preferably, in a short-term monitoring window, the specific steps of generating an instantaneous heart rate fluctuation reference value after abnormal analysis of the rapid change amplitude of the heart rate are as follows:

[0024] In the short-term monitoring window, continuous heart rate data is collected, set at time point t i At HR(t i ), the heart rate change rate is calculated by acquiring continuous heart rate data, and the calculation expression is: ΔHR(t i )=HR(t i )-HR(t i-1), where HR(t i ) represents the time point t i The heart rate value at t i ) represents the time point t i The difference between the heart rate at the moment and the heart rate at the previous moment indicates the instantaneous rate of change of the heart rate;

[0025] The heart rate change rate is nonlinearly enhanced and converted using a high-order exponential function. The expression of the nonlinear conversion is:

[0026] Φ(t i )=[exp(γ·|ΔHR(t i )| α )-1]

[0027] , where Φ(t i ) is the time point t i is the heart rate change value after nonlinear transformation, γ is the enhancement coefficient and is a positive real number used to adjust the sensitivity of the exponential function, α is the nonlinear order and α>1, which is used to control the sensitivity to large changes, |ΔHR(t i )| α The α power, which represents the absolute value of the heart rate change rate, amplifies the effect of the change rate;

[0028] The heart rate change value after nonlinear conversion is accumulated in the short-term monitoring window to obtain the instantaneous heart rate fluctuation reference value. The calculation expression of the instantaneous heart rate fluctuation reference value is: In the formula, IH rfi is the instantaneous heart rate fluctuation reference value, which quantifies the rapid fluctuation degree of heart rate in a short time, t0 is the starting time point of the short-term monitoring window, and Δt is the duration of the monitoring window.

[0029] Preferably, in a short-term monitoring window, the specific steps of generating a reference value for irregular heart rate after abnormal analysis of the changing pattern of heartbeat intervals are as follows:

[0030] In the short-term monitoring window, the interval time between each heartbeat is extracted from the ECG signal to generate a heartbeat interval time series. The heartbeat interval time series represents the change of the heartbeat interval and is the basis for evaluating the irregularity of heart rate. The expression of the heartbeat interval time series is: RR u =t u+1 -t u , u=1, 2,...,n, where, t u is the time when the uth heartbeat occurs, t u+1 is the time when the u+1th heartbeat occurs, RR u is the interval between the uth heartbeat and the u+1th heartbeat, and n is the number of collected heartbeats;

[0031] The nonlinear change rate is used to describe the dynamic change of the heartbeat interval. The nonlinear change rate is set as the second-order difference of the heartbeat interval time, and the change rate of the heartbeat interval time between adjacent points is quantified. The nonlinear change rate is: ΔRR u =|(RR u+2 -RR u+1 )-(RR u+1 -RR u )| τ , where ΔRR u It represents the nonlinear change rate of the uth interval, τ is the nonlinear adjustment factor, 2≤τ≤4, which reflects the acceleration of the heartbeat interval change;

[0032] The nonlinear change rate of all heartbeat intervals in the entire short-term monitoring window is integrated to generate a heart rate irregularity reference value to quantify the irregularity of the heartbeat interval time in the entire monitoring window. The heart rate irregularity reference value is generated by accumulating the nonlinear change rate and normalizing the overall heartbeat interval time series. The expression is:

[0033]

[0034] , where IR ri represents the reference value of heart rate irregularity, and λ is the sensitivity adjustment factor, which is used to control the response of the exponential function to the sensitivity of heart rate changes.

[0035] Preferably, if the heart rate change is classified as a normal change, the specific steps for increasing the actual infusion flow rate of the patient within a safe intravenous infusion flow rate range are as follows:

[0036] Calculate the coefficient difference between the potential abnormal coefficient of heart rate and the reference threshold of the potential abnormal coefficient of heart rate, and determine the amplitude of adjusting the flow rate according to the coefficient difference. The calculation expression of the coefficient difference is: ΔC = C t -Heart Rate , where ΔC is the coefficient difference, C t Heart rate potential abnormal coefficient reference threshold, Heart Rate is the potential abnormal coefficient of heart rate, indicating the change of heart rate in the current short-term monitoring window;

[0037] The flow rate adjustment ratio is calculated based on the coefficient difference. The flow rate adjustment ratio is proportional to the coefficient difference. The adjusted flow rate will be increased based on the starting intravenous infusion flow rate. The calculation expression of the flow rate adjustment ratio is: In the formula, μ is the flow rate adjustment ratio, which indicates the multiple relationship between the new infusion flow rate and the initial infusion flow rate, and k is the adjustment coefficient proportional constant, which is used to limit the maximum change of flow rate. The ratio of the heart rate stability to the reference threshold of the potential abnormal coefficient of the heart rate controls the ratio of the flow rate increase;

[0038] According to the flow rate adjustment ratio, the initial intravenous infusion flow rate is adjusted to the new infusion flow rate under the normal changes of the patient's heart rate. The calculation expression for the adjustment is: F new =μ·F0, where F0 is the initial intravenous infusion flow rate, i.e. the initially set infusion flow rate, and F new It is the new intravenous infusion flow rate under routine changes, that is, the actual infusion rate after adjustment.

[0039] Preferably, if the heart rate change is identified as a potential abnormal change, the initial intravenous infusion flow rate is lowered according to the degree of abnormality, and the patient's heart rate change is continued to be observed. The specific steps are as follows:

[0040] Step 1: When the heart rate change is identified as a potential abnormal change, the degree of abnormality is first quantified to determine the magnitude of the downward adjustment of the infusion flow rate. To this end, the abnormality coefficient is calculated to assess the severity of the heart rate abnormality. The calculation expression is: Where ω is the abnormality adjustment factor and ω>0, which is used to adjust the sensitivity of the abnormality coefficient. ln(·) is the natural logarithm function to ensure the smooth growth of the abnormality coefficient. ∈ is the abnormality coefficient, the larger the value, the higher the abnormality of the heart rate;

[0041] Step 2: When there is a potential abnormal change in the patient's heart rate, a new infusion flow rate is calculated based on the initial intravenous infusion flow rate and the abnormality coefficient to reduce the heart burden. The calculation expression is:

[0042]

[0043] , where η is the flow rate adjustment coefficient and η>0, which controls the influence of abnormality on the intravenous infusion flow rate, θ is the intravenous infusion flow rate adjustment index and θ>0, which is used to adjust the sensitivity of intravenous infusion flow rate decrease, F new 'Set the new IV flow rate for patients with potential abnormal changes in heart rate, ensuring that the IV flow rate decreases in a nonlinear manner as the degree of abnormality increases, is an adjustment factor less than 1;

[0044] Step 3: adjust the flow rate of the infusion device to the new intravenous infusion flow rate, and continuously monitor the patient's heart rate changes within the new short-term monitoring window. At the end of each new monitoring window, reacquire the heart rate data, update the potential abnormal coefficient of the heart rate, repeat steps 1 and 2, and dynamically adjust the actual intravenous infusion flow rate according to the latest intravenous infusion flow rate and abnormal degree coefficient;

[0045] Step 4: If, after multiple adjustments, the heart rate potential abnormality coefficient continues to be higher than the heart rate potential abnormality coefficient reference threshold, and the abnormality degree coefficient exceeds the preset alarm threshold, the alarm mechanism is triggered to notify medical personnel to intervene.

[0046] An information-based tracking system for intravenous therapy in hospitals, comprising a starting flow rate setting module, a heart rate monitoring module, a heart rate characteristic analysis module, a heart rate change classification module, a flow rate enhancement module, and a flow rate adjustment and alarm module;

[0047] In the initial flow rate setting module, medical staff first set a starting intravenous infusion flow rate for the patient based on the patient's condition, weight, age, drug characteristics, and treatment goals;

[0048] Heart rate monitoring module, which uses monitoring equipment to obtain the patient's heart rate information in real time during intravenous infusion;

[0049] The heart rate feature analysis module extracts key features after preprocessing the acquired raw heart rate data. In a short-term monitoring window, it performs abnormal analysis on the extracted heart rate features. Subsequently, it uses a pre-trained machine learning model to comprehensively evaluate and predict the analyzed heart rate features.

[0050] The heart rate change classification module divides the patient's heart rate changes in a short-term monitoring window into potential abnormal changes and regular changes based on the prediction results of the machine learning model;

[0051] The flow rate increase module increases the patient's actual infusion flow rate within a safe intravenous infusion flow rate range if the heart rate change is classified as a normal change;

[0052] The flow rate adjustment and alarm module, if the heart rate change is identified as a potential abnormal change, the starting intravenous infusion flow rate will be lowered according to the degree of abnormality, and the patient's heart rate changes will continue to be observed to ensure that the patient is safely infused when the heart rate changes normally. If the heart rate still does not return to normal, an alarm will be issued to notify medical staff.

[0053] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0054] The present invention obtains the patient's heart rate information in real time, combines the preprocessing and analysis of the heart rate characteristics, and uses a pre-trained machine learning model to dynamically evaluate the patient's heart rate changes. Unlike the traditional fixed-flow intravenous infusion method, the present invention can flexibly adjust the infusion flow rate according to the patient's heart rate fluctuations within a short-term monitoring window, thereby achieving personalized infusion process management. In particular, when potential abnormal changes in heart rate are detected, the system will automatically lower the starting infusion flow rate to reduce the patient's heart burden and prevent serious complications such as acute heart failure or pulmonary edema caused by excessively fast infusion speeds. This dynamic control mechanism significantly improves the safety of the infusion process, and has an important protective effect on patients with heart failure in particular.

[0055] The present invention not only accurately identifies the type of heart rate changes through machine learning models, but also automatically increases the patient's actual infusion flow rate when the heart rate is assessed as a regular fluctuation, ensuring that the effective infusion of drugs is accelerated within a safe range. This intelligent flow rate adjustment mechanism overcomes the overly conservative settings in the traditional infusion process, not only improving the infusion efficiency and shortening the treatment time, but also reducing the workload of medical staff in heart rate monitoring and manual adjustment through automated monitoring and alarm functions. This systematic management allows medical staff to focus more on other complex nursing tasks, significantly improving the efficiency of overall nursing work. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0057] Figure 1 The present invention is a method flow chart of an intravenous therapy information tracking method based on hospital intravenous therapy.

[0058] Figure 2 The present invention is a module schematic diagram of an intravenous therapy information tracking system based on hospital intravenous therapy. DETAILED DESCRIPTION

[0059] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0060] The present invention provides Figure 1 The illustrated method for tracking intravenous therapy information based on hospital intravenous therapy includes the following steps:

[0061] Medical staff first set a starting intravenous infusion flow rate for the patient based on the patient's condition, weight, age, characteristics of the medication, and treatment goals;

[0062] The steps to set the initial intravenous infusion flow rate are as follows:

[0063] Step 1: Before setting the initial infusion flow rate, medical staff should first conduct a comprehensive assessment of the patient's overall physical condition;

[0064] This step includes collecting and analyzing the patient's medical history, existing medical conditions (such as heart disease, renal insufficiency, diabetes, etc.), vital signs (such as blood pressure, heart rate, blood oxygen level, etc.), and physiological data (such as age, weight, height, etc.). This information is crucial for judging the patient's tolerance and formulating a personalized treatment plan. For example, patients with weaker heart function may not be able to tolerate a higher fluid load during infusion, so the initial flow rate needs to be set lower; patients with severe dehydration or acute blood loss may require a higher initial flow rate to quickly replenish body fluids.

[0065] The purpose of this step is to determine the patient's tolerance to the infusion rate through the patient's comprehensive physiological and pathological data. This is the basis for ensuring the safety of infusion and avoiding adverse effects on patients due to excessively high or low flow rates. In the prior art, medical staff usually rely on these data to preliminarily assess the patient's tolerance to infusion.

[0066] Step 2: further refine the setting of the initial flow rate according to the characteristics of the drug used and the treatment goal;

[0067] The type of drug, concentration, and purpose of infusion (such as fluid replacement, antibiotic treatment, nutritional support, etc.) will directly affect the flow rate setting. For example, certain irritant drugs (such as chemotherapy drugs or certain high-concentration antibiotics) need to be infused slowly to avoid damage to blood vessels; while rapid fluid resuscitation for critically ill patients requires a faster infusion rate. In addition, pharmacokinetic parameters such as the half-life and onset time of the drug will also affect the flow rate setting. When the infusion goal is clear, such as relieving dehydration or correcting electrolyte imbalance, the initial flow rate setting will be accurately calculated based on the expected infusion volume and time.

[0068] The characteristics of the drug and the treatment goal directly determine the flow rate setting. For example, some drugs need to be released slowly, and rapid infusion may cause serious adverse reactions. Therefore, when setting the flow rate, the characteristics of the drug itself and the treatment goal must be considered to ensure safety and efficacy.

[0069] Step 3: Based on the patient's physiological assessment and drug characteristics, the medical staff sets the initial intravenous infusion flow rate according to clinical practice guidelines or their clinical experience;

[0070] Clinical guidelines usually provide a recommended flow rate range for specific conditions or drugs, within which medical staff can make personalized adjustments. For example, for basic fluid rehydration in general adults, the commonly used flow rate is 20-40 ml / hour; while for rapid fluid rehydration in critically ill patients, the flow rate may be as high as 100 ml / hour or even higher. In the absence of a special emergency, the initial flow rate is set relatively conservatively so that it can be dynamically adjusted based on the patient's response. This step also includes the setting of the equipment, such as entering the appropriate flow rate parameters on the infusion pump and starting the infusion.

[0071] The purpose of this step is to set an appropriate flow rate based on reference specifications and actual conditions to ensure the safety and effectiveness of treatment. In the prior art, clinical guidelines provide basic reference standards for medical staff to combine their personal experience and judgment to ultimately determine the most appropriate starting flow rate.

[0072] During the patient's intravenous infusion, use monitoring equipment to obtain the patient's heart rate information in real time;

[0073] These devices can be bedside monitors, portable electrocardiographs, or wearable devices such as smart bracelets, chest straps, etc. Heart rate data is continuously transmitted to a central system under the monitoring window via a wireless network or wired connection.

[0074] After preprocessing the acquired raw heart rate data, key features are extracted. Under a short-term monitoring window, abnormal analysis is performed on the extracted heart rate features. Subsequently, a pre-trained machine learning model is used to comprehensively evaluate and predict the analyzed heart rate features.

[0075] Preprocessing the acquired raw heart rate data usually includes the following steps:

[0076] Denoising, removing interfering signals or environmental noise from data through filtering or wavelet transform;

[0077] Data smoothing, using moving average or exponential smoothing to reduce short-term fluctuations in heart rate data;

[0078] Missing data processing, solving the problem of missing values ​​in the data through interpolation or filling methods;

[0079] Standardization or normalization is the process of normalizing the data so that features of different ranges have the same scale to improve the stability of the model and the computational efficiency.

[0080] The role of preprocessing is to ensure the quality and consistency of data, remove errors and outliers, and provide reliable input for subsequent feature extraction and model analysis.

[0081] The short-term monitoring window refers to the continuous collection and analysis of heart rate data within a fixed short period of time (such as a few minutes or seconds) to capture instantaneous changes and short-term fluctuations in heart rate and identify potential abnormalities in a timely manner.

[0082] The key features are extracted after preprocessing the acquired raw heart rate data, wherein the extracted key features include the rapid change amplitude of the heart rate and the changing law of the heart beat interval time. Under the short-term monitoring window, after abnormal analysis of the rapid change amplitude of the heart rate and the changing law of the heart beat interval time, an instantaneous heart rate fluctuation reference value and a heart rate irregularity reference value are generated respectively. The rapid fluctuation amplitude of the heart rate in a short period of time is quantified by the instantaneous heart rate fluctuation reference value, which is used to evaluate whether the instantaneous change of the heart rate exceeds the normal range. The changing law of the heart beat interval time is quantified by the heart rate irregularity reference value, which is used to evaluate the irregularity of the heart beat interval.

[0083] After obtaining the instantaneous heart rate fluctuation reference value and heart rate irregularity reference value generated by abnormal analysis during intravenous infusion, the instantaneous heart rate fluctuation reference value and the heart rate irregularity reference value are used to generate a heart rate potential abnormality coefficient through a pre-trained machine learning model, and the heart rate condition of the patient during intravenous infusion is comprehensively evaluated and predicted through the heart rate potential abnormality coefficient.

[0084] A pre-trained machine learning model refers to an algorithmic model that has been trained and optimized with a large amount of historical data to identify and predict specific patterns. In this solution, the model uses relevant characteristic data of heart rate, such as instantaneous heart rate fluctuation reference values ​​and heart rate irregularity reference values, to predict whether the patient's heart rate is potentially abnormal during intravenous infusion. These models are trained through heart rate changes in historical data, patient conditions, and physiological signals when cardiac events occur, and learn the complex relationship between heart rate characteristics and heart rate abnormalities (such as arrhythmias, myocardial ischemia, etc.). The training process involves model parameter optimization, weight adjustment, fitting control, etc., so that the model can efficiently predict and evaluate new and unseen heart rate data.

[0085] When training the model, commonly used algorithms include support vector machines, random forests, neural networks, etc., which learn the relationship between features and labels through supervised learning. Labels are usually derived from real heart rate abnormalities in clinical data, such as recorded abnormal heart events or doctors' diagnoses. Through the feature input of a large amount of heart rate data, the model can identify key heart rate fluctuation patterns, and combine the advanced features of instantaneous heart rate fluctuation reference values ​​and irregular heart rate reference values ​​to achieve accurate prediction of potential heart rate abnormality coefficients. This prediction not only helps medical staff assess the patient's heart condition in real time, but also provides early warning before abnormalities occur, ensuring the safety of the infusion process and the stability of heart function.

[0086] If a patient's heart rate changes rapidly and with a large amplitude during intravenous infusion, it usually indicates that the patient's heart rate may have potential abnormalities. This rapid change in amplitude may reflect that the patient's heart is not responding normally to the infusion rate or infused medication, which may be related to increased cardiac workload, drug response, or abnormal regulation of the autonomic nervous system. Rapid fluctuations in heart rate beyond the normal range may indicate potential arrhythmias, cardiac insufficiency, or even serious conditions such as acute heart failure, so timely monitoring and evaluation are required to ensure safety during the infusion process.

[0087] In the short-term monitoring window, the specific steps for generating the instantaneous heart rate fluctuation reference value after abnormal analysis of the rapid change amplitude of the heart rate are as follows:

[0088] In the short-term monitoring window, continuous heart rate data is collected, set at time point t i At HR(t i ), the heart rate change rate is calculated by acquiring continuous heart rate data, and the calculation expression is: ΔHR(t i )=HR(t i )-HR(t i-1 ), where HR(t i ) represents the time point t i The heart rate value at t i ) represents the time point t i The difference between the heart rate at the moment and the heart rate at the previous moment indicates the instantaneous rate of change of the heart rate;

[0089] In order to emphasize the rapid change of heart rate, the heart rate change rate is nonlinearly enhanced and converted using a high-order exponential function. The expression of the nonlinear conversion is:

[0090] Φ(t i )=[exp(γ·|ΔHR(t i )| α )-1]

[0091] , where Φ(t i ) is the time point t i is the heart rate change value after nonlinear transformation, γ is the enhancement coefficient and is a positive real number used to adjust the sensitivity of the exponential function, α is the nonlinear order and α>1, which is used to control the sensitivity to large changes, |ΔHR(t i )| α The α power, which represents the absolute value of the rate of change of heart rate, amplifies the effect of larger rates of change;

[0092] The heart rate change value after nonlinear conversion is accumulated in the short-term monitoring window to obtain the instantaneous heart rate fluctuation reference value. The calculation expression of the instantaneous heart rate fluctuation reference value is: In the formula, IH rfiis the instantaneous heart rate fluctuation reference value, which quantifies the rapid fluctuation degree of heart rate in a short time, t0 is the starting time point of the short-term monitoring window, and Δt is the duration of the monitoring window.

[0093] It can be seen from the instantaneous heart rate fluctuation reference value that, in a short-term monitoring window, the larger the performance value of the instantaneous heart rate fluctuation reference value generated after abnormal analysis of the rapid change amplitude of the heart rate, the larger the rapid fluctuation amplitude of the patient's heart rate, which in turn indicates that the patient has a higher risk of potential heart rate abnormalities. This is because large and frequent heart rate fluctuations are usually associated with cardiac dysfunction, sympathetic overexcitation, or arrhythmia. Conversely, a smaller instantaneous heart rate fluctuation reference value indicates that the heart rate fluctuations are stable, the patient's heart state is relatively normal, and the potential abnormality risk is low. Therefore, the instantaneous heart rate fluctuation reference value can be used as an important indicator to quantify the risk of abnormal heart rate.

[0094] Irregular changes in the patient's heartbeat intervals usually indicate a potential abnormality in the patient's heart rate. The stability of the heartbeat intervals reflects the normal rhythm of the heart's electrical activity. When the intervals change irregularly, it may indicate problems such as arrhythmia, atrial fibrillation, or other abnormal heart function. This change is especially noteworthy during intravenous infusion, because too fast an infusion rate or drug stimulation may increase the burden on the heart and cause arrhythmia. Therefore, timely identification of abnormal fluctuations in the heartbeat intervals is important for preventing potential heart complications.

[0095] In the short-term monitoring window, the specific steps for generating the reference value of irregular heart rate after abnormal analysis of the changing rules of heartbeat intervals are as follows:

[0096] In the short-term monitoring window, the interval time between each heartbeat is extracted from the ECG signal to generate a heartbeat interval time series. The heartbeat interval time series represents the change of the heartbeat interval and is the basis for evaluating the irregularity of heart rate. The expression of the heartbeat interval time series is: RR u =t u+1 -t u , u=1, 2,...,n, where, t u is the time when the uth heartbeat occurs, t u+1 is the time when the u+1th heartbeat occurs, RR u is the interval between the uth heartbeat and the u+1th heartbeat, and n is the number of collected heartbeats;

[0097] In order to avoid simple linear characteristics, the nonlinear change rate is used to describe the dynamic change of the heartbeat interval. The nonlinear change rate is set as the second-order difference of the heartbeat interval time, and the change rate of the heartbeat interval time between adjacent points is quantified. The nonlinear change rate is: ΔRR u =|(RR u+2 -RRu+1 )-(RR u+1 -RR u )| τ , where ΔRR u It represents the nonlinear change rate of the uth interval, τ is the nonlinear adjustment factor, 2≤τ≤4 (usually 2 or 3 to amplify the influence of irregular fluctuations), reflecting the acceleration of the heartbeat interval change;

[0098] The nonlinear change rate of all heartbeat intervals in the entire short-term monitoring window is integrated to generate a heart rate irregularity reference value to quantify the irregularity of the heartbeat interval time in the entire monitoring window. The heart rate irregularity reference value is generated by accumulating the nonlinear change rate and normalizing the overall heartbeat interval time series. The expression is:

[0099]

[0100] , where IR ri It represents the reference value of heart rate irregularity, and λ is the sensitivity adjustment factor, which is used to control the response of the exponential function to the sensitivity of heart rate changes to ensure that excessive or small changes can be reasonably captured.

[0101] It can be seen from the reference value of irregular heart rate that in the short-term monitoring window, the larger the performance value of the reference value of irregular heart rate generated after abnormal analysis of the changing pattern of the heartbeat interval time, it usually means that the fluctuation amplitude and irregularity of the heartbeat interval are more significant, indicating that the patient is at greater risk of potential heart rate abnormalities. This may indicate abnormal heart function problems such as arrhythmia and atrial fibrillation. On the contrary, if the performance value of the reference value of irregular heart rate is smaller, it means that the heartbeat interval is relatively stable and the patient is less likely to have potential heart rhythm abnormalities. This quantitative method can more intuitively assess the patient's heart health status.

[0102] The machine learning model is not specifically limited here, and can realize the instantaneous heart rate fluctuation reference value IH rfi and heart rate irregularity reference value IR ri Comprehensive analysis generates potential abnormal heart rate coefficient Heart Rate In order to realize the technical solution of the present invention, the present invention provides a specific implementation method; Heart rate potential abnormality coefficient Heart Rate The resulting calculation formula is: In the formula, They are respectively the instantaneous heart rate fluctuation reference value IH rfi and heart rate irregularity reference value IR ri The preset scaling factor of Both are greater than 0.

[0103] It can be seen from the potential abnormality coefficient of heart rate that, under the short-term monitoring window, the greater the performance value of the instantaneous heart rate fluctuation reference value generated after the abnormal analysis of the rapid change amplitude of the heart rate, the greater the performance value of the heart rate irregularity reference value generated after the abnormal analysis of the changing law of the heartbeat interval time, that is, the greater the performance value of the potential abnormality coefficient of heart rate generated when the patient's heart rate during intravenous infusion is comprehensively evaluated and predicted by the machine learning model, the greater the risk of potential heart rate abnormality in the patient, and vice versa.

[0104] Based on the prediction results of the machine learning model, the patient's heart rate changes in the short-term monitoring window are divided into potential abnormal changes and regular changes;

[0105] The potential abnormal heart rate coefficient generated when the machine learning model is used to comprehensively evaluate and predict the patient's heart rate during intravenous infusion is compared and analyzed with the pre-set reference threshold of the potential abnormal heart rate coefficient, and the patient's heart rate changes in the short-term monitoring window are divided. The specific division steps are: if the potential abnormal heart rate coefficient is greater than the reference threshold of the potential abnormal heart rate coefficient, the patient's heart rate changes in the short-term monitoring window are divided into potential abnormal changes, which may indicate serious problems such as increased cardiac burden and arrhythmia, and require immediate treatment; if the potential abnormal heart rate coefficient is less than or equal to the reference threshold of the potential abnormal heart rate coefficient, the patient's heart rate changes in the short-term monitoring window are divided into regular changes, which may be caused by normal physiological fluctuations or slight external stimulation and do not require special intervention.

[0106] Through this classification, appropriate infusion strategies can be adopted according to different heart rate changes.

[0107] If the heart rate changes are classified as normal changes, the patient's actual infusion rate is increased within the safe intravenous infusion rate range;

[0108] If the heart rate changes are classified as normal, the specific steps to increase the patient's actual infusion rate within the safe intravenous infusion rate range are as follows:

[0109] Calculate the coefficient difference between the potential abnormal coefficient of heart rate and the reference threshold of the potential abnormal coefficient of heart rate, and determine the amplitude of adjusting the flow rate according to the coefficient difference. The calculation expression of the coefficient difference is: ΔC = C t -Heart Rate , where ΔC is the coefficient difference. The larger the coefficient difference, the more stable the heart rate and the greater the space for flow rate improvement. t Heart rate potential abnormal coefficient reference threshold, Heart Rate is the potential abnormal coefficient of heart rate, indicating the change of heart rate in the current short-term monitoring window;

[0110] The flow rate adjustment ratio is calculated based on the coefficient difference. The flow rate adjustment ratio is proportional to the coefficient difference. The adjusted flow rate will be increased accordingly based on the starting intravenous infusion flow rate. The calculation expression of the flow rate adjustment ratio is: In the formula, μ is the flow rate adjustment ratio, which indicates the multiple relationship between the new infusion flow rate and the initial infusion flow rate, and k is the adjustment coefficient proportional constant, which is used to limit the maximum change of the flow rate to prevent excessive adjustment. The ratio of the heart rate stability to the reference threshold of the potential abnormal coefficient of the heart rate controls the ratio of the flow rate increase;

[0111] The setting of the adjustment coefficient proportional constant k needs to comprehensively consider the individual differences of patients, clinical safety, and the specific needs of infusion. Generally, the setting of k should be based on clinical trial data and the patient's physiological condition to ensure that the flow rate adjustment does not exceed the safe range. Generally speaking, the value of k should be small to control the magnitude of the flow rate increase and avoid excessive burden on the heart due to too fast infusion. For ordinary patients, k can be set between 0.1 and 0.3 to ensure a moderate increase in flow rate; for patients with weak heart function or potential risks, the value of k should be lower, possibly set to 0.05, to prevent the infusion rate from being too fast and causing adverse effects on the heart and circulatory system.

[0112] According to the flow rate adjustment ratio, the initial intravenous infusion flow rate is adjusted to the new infusion flow rate under the normal changes of the patient's heart rate. The calculation expression for the adjustment is: F new =μ·F0, where F0 is the initial intravenous infusion flow rate, i.e. the initially set infusion flow rate, and F new It is the new intravenous infusion flow rate under routine changes, that is, the actual infusion rate after adjustment.

[0113] When the heart rate changes are classified as regular changes, increasing the patient's actual infusion flow rate can improve infusion efficiency and speed up the infusion of drugs or fluids, thereby achieving treatment effects faster. This adjustment can not only shorten the treatment time and reduce the patient's discomfort, but also optimize the treatment process while ensuring the stability of the heart burden and physiological state. At the same time, it can reduce the occupancy rate of medical resources and improve nursing efficiency, especially in the case of acute treatment or high fluid replacement needs, which helps to improve the overall treatment response speed.

[0114] Increasing the actual infusion flow rate of the patient within the safe intravenous infusion flow rate range helps speed up the infusion of drugs or fluids and improves treatment efficiency. For example, if the patient tolerates it, the flow rate is increased from the original 50mL / h to 60mL / h. In the prior art, the infusion flow rate is usually adjusted manually by medical staff, and frequent adjustments increase the workload. This solution improves efficiency through automated adjustment.

[0115] If the heart rate change is identified as a potential abnormal change, the initial intravenous infusion flow rate will be lowered according to the degree of abnormality, and the patient's heart rate changes will continue to be observed to ensure that the patient can receive a safe infusion when the heart rate changes normally. If the heart rate still does not return to normal, an alarm will be issued to inform the medical staff;

[0116] If the heart rate change is identified as a potential abnormal change, the initial intravenous infusion flow rate is adjusted down according to the degree of abnormality, and the patient's heart rate changes are continued to be observed. The specific steps are as follows:

[0117] Step 1: When the heart rate change is identified as a potential abnormal change, the degree of abnormality is first quantified to determine the magnitude of the downward adjustment of the infusion flow rate. To this end, the abnormality coefficient is calculated to assess the severity of the heart rate abnormality. The calculation expression is: Where ω is the abnormality adjustment factor and ω>0, which is used to adjust the sensitivity of the abnormality coefficient. ln(·) is the natural logarithm function to ensure the smooth growth of the abnormality coefficient and avoid excessive amplification. ∈ is the abnormality coefficient, the larger the value, the higher the abnormality of the heart rate;

[0118] Step 2: When there is a potential abnormal change in the patient's heart rate, a new infusion flow rate is calculated based on the initial intravenous infusion flow rate and the abnormality coefficient to reduce the heart burden. The calculation expression is:

[0119]

[0120] , where η is the flow rate adjustment coefficient and η>0, which controls the influence of abnormality on the intravenous infusion flow rate, θ is the intravenous infusion flow rate adjustment index and θ>0, which is used to adjust the sensitivity of intravenous infusion flow rate decrease, F new 'Set the new IV flow rate for patients with potential abnormal changes in heart rate, ensuring that the IV flow rate decreases in a nonlinear manner as the degree of abnormality increases, It is an adjustment factor less than 1. The higher the abnormality, the smaller the value, resulting in a greater decrease in the infusion flow rate.

[0121] Step 3: adjust the flow rate of the infusion device to the new intravenous infusion flow rate, and continuously monitor the patient's heart rate changes within the new short-term monitoring window. At the end of each new monitoring window, reacquire the heart rate data, update the potential abnormal coefficient of the heart rate, repeat steps 1 and 2, and dynamically adjust the actual intravenous infusion flow rate according to the latest intravenous infusion flow rate and abnormal degree coefficient;

[0122] Step 4: If after multiple adjustments (the number of adjustments can be set according to actual conditions and is not specifically limited here), the potential abnormal coefficient of heart rate continues to be higher than the reference threshold of the potential abnormal coefficient of heart rate, and the abnormal degree coefficient exceeds the preset alarm threshold, the alarm mechanism is triggered to notify medical personnel to intervene;

[0123] The alarm threshold is a preset value, exceeding which indicates that the heart rate is seriously abnormal and requires immediate human intervention.

[0124] For example, the flow rate is reduced from 50mL / h to 40mL / h. At the same time, the system will continue to monitor the patient's heart rate in real time to observe whether there is any improvement. If the heart rate still does not return to normal, the system may sound an alarm to remind medical staff to conduct further inspections and interventions. In the prior art, the lack of such a real-time response mechanism may result in abnormal heart rate not being discovered in time, increasing the risk to patients.

[0125] The present invention obtains the patient's heart rate information in real time, combines the preprocessing and analysis of the heart rate characteristics, and uses a pre-trained machine learning model to dynamically evaluate the patient's heart rate changes. Unlike the traditional fixed-flow intravenous infusion method, the present invention can flexibly adjust the infusion flow rate according to the patient's heart rate fluctuations within a short-term monitoring window, thereby achieving personalized infusion process management. In particular, when potential abnormal changes in heart rate are detected, the system will automatically lower the starting infusion flow rate to reduce the patient's heart burden and prevent serious complications such as acute heart failure or pulmonary edema caused by excessively fast infusion speeds. This dynamic control mechanism significantly improves the safety of the infusion process, and has an important protective effect on patients with heart failure in particular.

[0126] The present invention not only accurately identifies the type of heart rate changes through machine learning models, but also automatically increases the patient's actual infusion flow rate when the heart rate is assessed as a regular fluctuation, ensuring that the effective infusion of drugs is accelerated within a safe range. This intelligent flow rate adjustment mechanism overcomes the overly conservative settings in the traditional infusion process, not only improving the infusion efficiency and shortening the treatment time, but also reducing the workload of medical staff in heart rate monitoring and manual adjustment through automated monitoring and alarm functions. This systematic management allows medical staff to focus more on other complex nursing tasks, significantly improving the efficiency of overall nursing work.

[0127] The present invention provides Figure 2 The illustrated intravenous therapy information tracking system based on hospital intravenous therapy includes a starting flow rate setting module, a heart rate monitoring module, a heart rate characteristic analysis module, a heart rate change classification module, a flow rate enhancement module, and a flow rate adjustment and alarm module;

[0128] In the initial flow rate setting module, medical staff first set a starting intravenous infusion flow rate for the patient based on the patient's condition, weight, age, drug characteristics, and treatment goals;

[0129] Heart rate monitoring module, which uses monitoring equipment to obtain the patient's heart rate information in real time during intravenous infusion;

[0130] The heart rate feature analysis module extracts key features after preprocessing the acquired raw heart rate data. In a short-term monitoring window, it performs abnormal analysis on the extracted heart rate features. Subsequently, it uses a pre-trained machine learning model to comprehensively evaluate and predict the analyzed heart rate features.

[0131] The heart rate change classification module divides the patient's heart rate changes in a short-term monitoring window into potential abnormal changes and regular changes based on the prediction results of the machine learning model;

[0132] The flow rate increase module increases the patient's actual infusion flow rate within a safe intravenous infusion flow rate range if the heart rate change is classified as a normal change;

[0133] The flow rate adjustment and alarm module, if the heart rate change is identified as a potential abnormal change, will reduce the initial intravenous infusion flow rate according to the degree of abnormality, and continue to observe the patient's heart rate changes to ensure that the patient is safely infused when the heart rate changes normally. If the heart rate still does not return to normal, an alarm will be issued to inform the medical staff;

[0134] An information-based tracking method for intravenous therapy based on hospital intravenous therapy provided in an embodiment of the present invention is implemented by the above-mentioned information-based tracking system for intravenous therapy based on hospital intravenous therapy. The specific method and process of the information-based tracking system for intravenous therapy based on hospital intravenous therapy are detailed in the embodiment of the above-mentioned information-based tracking method for intravenous therapy based on hospital intravenous therapy, which will not be repeated here.

[0135] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0136] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0137] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for tracking intravenous therapy based on hospital intravenous therapy information, characterized in that: The following steps are involved: Medical staff first set a starting intravenous infusion flow rate for the patient based on the patient's condition, weight, age, characteristics of the medication, and treatment goals; During the patient's intravenous infusion, use monitoring equipment to obtain the patient's heart rate information in real time; After preprocessing the acquired raw heart rate data, key features are extracted. Under a short-term monitoring window, abnormal analysis is performed on the extracted heart rate features. Subsequently, a pre-trained machine learning model is used to comprehensively evaluate and predict the analyzed heart rate features. Based on the prediction results of the machine learning model, the patient's heart rate changes in the short-term monitoring window are divided into potential abnormal changes and regular changes; If the heart rate changes are classified as normal changes, the patient's actual infusion rate is increased within the safe intravenous infusion rate range; If the heart rate change is identified as a potential abnormal change, the initial intravenous infusion flow rate will be lowered according to the degree of abnormality, and the patient's heart rate changes will continue to be observed to ensure that the patient can receive a safe infusion when the heart rate changes normally. If the heart rate still does not return to normal, an alarm will be issued to inform the medical staff; The key features are extracted after preprocessing the acquired raw heart rate data, wherein the extracted key features include the rapid change amplitude of the heart rate and the change law of the heartbeat interval time. Under the short-term monitoring window, after abnormal analysis of the rapid change amplitude of the heart rate and the change law of the heartbeat interval time, an instantaneous heart rate fluctuation reference value and a heart rate irregularity reference value are generated respectively. The rapid fluctuation amplitude of the heart rate in a short period of time is quantified by the instantaneous heart rate fluctuation reference value, which is used to evaluate whether the instantaneous change of the heart rate exceeds the normal range. The change law of the heartbeat interval time is quantified by the heart rate irregularity reference value, which is used to evaluate the irregularity of the heartbeat interval; After obtaining the instantaneous heart rate fluctuation reference value and the heart rate irregularity reference value generated by abnormal analysis during the intravenous infusion of the patient, the instantaneous heart rate fluctuation reference value and the heart rate irregularity reference value are used to generate a heart rate potential abnormality coefficient through a pre-trained machine learning model, and the heart rate condition of the patient during the intravenous infusion is comprehensively evaluated and predicted through the heart rate potential abnormality coefficient; In the short-term monitoring window, the specific steps for generating the instantaneous heart rate fluctuation reference value after abnormal analysis of the rapid change amplitude of the heart rate are as follows: In the short-term monitoring window, continuous heart rate data is collected, set at time point t i At HR(t i ), the heart rate change rate is calculated by acquiring continuous heart rate data, and the calculation expression is: ΔHR(t i )=HR(t i )-HR(t i-1 ), where HR(t i ) represents the time point t i The heart rate value at t i ) represents the time point t i The difference between the heart rate at the moment and the heart rate at the previous moment indicates the instantaneous rate of change of the heart rate; The heart rate change rate is nonlinearly enhanced and converted using a high-order exponential function. The expression of the nonlinear conversion is: Φ(t i )=[exp(γ·|ΔHR(t i )| α )-1], In the formula, Φ(t i ) is the time point t i is the heart rate change value after nonlinear transformation, γ is the enhancement coefficient and is a positive real number used to adjust the sensitivity of the exponential function, α is the nonlinear order and α>1, which is used to control the sensitivity to large changes, |ΔHR(t i )| α The α power, which represents the absolute value of the heart rate change rate, amplifies the effect of the change rate; The heart rate change value after nonlinear conversion is accumulated in the short-term monitoring window to obtain the instantaneous heart rate fluctuation reference value. The calculation expression of the instantaneous heart rate fluctuation reference value is: In the formula, IH rfi is the instantaneous heart rate fluctuation reference value, which quantifies the rapid fluctuation degree of heart rate in a short period of time, t0 is the starting time point of the short-term monitoring window, and Δt is the duration of the monitoring window; In the short-term monitoring window, the specific steps for generating the reference value of irregular heart rate after abnormal analysis of the changing rules of heartbeat intervals are as follows: In the short-term monitoring window, the interval time between each heartbeat is extracted from the ECG signal to generate a heartbeat interval time series. The heartbeat interval time series represents the change of the heartbeat interval and is the basis for evaluating the irregularity of heart rate. The expression of the heartbeat interval time series is: RR u =t u+1 -t u , u=1, 2,...,n, where, t u is the time when the uth heartbeat occurs, t u+1 is the time when the u+1th heartbeat occurs, RR u is the interval between the uth heartbeat and the u+1th heartbeat, and n is the number of collected heartbeats; The nonlinear change rate is used to describe the dynamic change of the heartbeat interval. The nonlinear change rate is set as the second-order difference of the heartbeat interval time, and the change rate of the heartbeat interval time between adjacent points is quantified. The nonlinear change rate is: ΔRR u =|(RR u+2 -RR u+1 )-(RR u+1 -RR u )|τ, where ΔRR u It represents the nonlinear change rate of the uth interval, τ is the nonlinear adjustment factor, 2≤τ≤4, which reflects the acceleration of the heartbeat interval change; The nonlinear change rate of all heartbeat intervals in the entire short-term monitoring window is integrated to generate a heart rate irregularity reference value to quantify the irregularity of the heartbeat interval time in the entire monitoring window. The heart rate irregularity reference value is generated by accumulating the nonlinear change rate and normalizing the overall heartbeat interval time series. The expression is: ， In the formula, IR ri represents the reference value of heart rate irregularity, and λ is the sensitivity adjustment factor, which is used to control the response of the exponential function to the sensitivity of heart rate changes.

2. The method for tracking intravenous therapy information based on hospital intravenous therapy according to claim 1, characterized in that: The steps to set the initial intravenous infusion flow rate are as follows: Step 1: Before setting the initial infusion flow rate, medical staff should first conduct a comprehensive assessment of the patient's overall physical condition; Step 2: further refine the setting of the initial flow rate according to the characteristics of the drug used and the treatment goal; Step 3: Based on the patient's physiological assessment and drug characteristics, the healthcare provider sets the initial intravenous infusion flow rate according to clinical practice guidelines or their clinical experience.

3. The method for tracking intravenous therapy information based on hospital intravenous therapy according to claim 1, characterized in that: The short-term monitoring window refers to the continuous collection and analysis of heart rate data within a fixed short period of time to capture instantaneous changes and short-term fluctuations in heart rate and identify potential abnormalities in a timely manner.

4. The method for tracking intravenous therapy based on hospital intravenous therapy information according to claim 1, characterized in that: The potential abnormal heart rate coefficient generated by the comprehensive evaluation and prediction of the patient's heart rate during intravenous infusion through the machine learning model is compared and analyzed with the pre-set reference threshold of the potential abnormal heart rate coefficient, and the patient's heart rate changes in the short-term monitoring window are divided. The specific division steps are: if the potential abnormal heart rate coefficient is greater than the reference threshold of the potential abnormal heart rate coefficient, the patient's heart rate changes in the short-term monitoring window are divided into potential abnormal changes; if the potential abnormal heart rate coefficient is less than or equal to the reference threshold of the potential abnormal heart rate coefficient, the patient's heart rate changes in the short-term monitoring window are divided into normal changes.

5. The method for tracking intravenous therapy information based on hospital intravenous therapy according to claim 4, characterized in that: If the heart rate changes are classified as normal, the specific steps to increase the patient's actual infusion rate within the safe intravenous infusion rate range are as follows: Calculate the coefficient difference between the potential abnormal coefficient of heart rate and the reference threshold of the potential abnormal coefficient of heart rate, and determine the amplitude of adjusting the flow rate according to the coefficient difference. The calculation expression of the coefficient difference is: ΔC = C t -Heart Rate , where ΔC is the coefficient difference, C t Heart rate potential abnormal coefficient reference threshold, Heart Rate is the potential abnormal coefficient of heart rate, indicating the change of heart rate in the current short-term monitoring window; The flow rate adjustment ratio is calculated based on the coefficient difference. The flow rate adjustment ratio is proportional to the coefficient difference. The adjusted flow rate will be increased based on the starting intravenous infusion flow rate. The calculation expression of the flow rate adjustment ratio is: In the formula, μ is the flow rate adjustment ratio, which indicates the multiple relationship between the new infusion flow rate and the initial infusion flow rate, and k is the adjustment coefficient proportional constant, which is used to limit the maximum change of flow rate. The ratio of the heart rate stability to the reference threshold of the potential abnormal coefficient of the heart rate controls the ratio of the flow rate increase; According to the flow rate adjustment ratio, the initial intravenous infusion flow rate is adjusted to the new infusion flow rate under the normal changes of the patient's heart rate. The calculation expression for the adjustment is: F new =μ·F0, where F0 is the initial intravenous infusion flow rate, i.e. the initially set infusion flow rate, and F new It is the new intravenous infusion flow rate under routine changes, that is, the actual infusion rate after adjustment.

6. The method for tracking intravenous therapy information based on hospital intravenous therapy according to claim 5, characterized in that: If the heart rate change is identified as a potential abnormal change, the initial intravenous infusion flow rate is adjusted down according to the degree of abnormality, and the patient's heart rate changes are continued to be observed. The specific steps are as follows: Step 1: When the heart rate change is identified as a potential abnormal change, the degree of abnormality is first quantified to determine the magnitude of the downward adjustment of the infusion flow rate. To this end, the abnormality coefficient is calculated to assess the severity of the heart rate abnormality. The calculation expression is: Where ω is the abnormality adjustment factor and ω>0, which is used to adjust the sensitivity of the abnormality coefficient. ln(·) is the natural logarithm function to ensure the smooth growth of the abnormality coefficient. ∈ is the abnormality coefficient, the larger the value, the higher the abnormality of the heart rate; Step 2: When there is a potential abnormal change in the patient's heart rate, a new infusion flow rate is calculated based on the initial intravenous infusion flow rate and the abnormality coefficient to reduce the heart burden. The calculation expression is: ， Where η is the flow rate adjustment proportional coefficient and η>0, which controls the influence of abnormality on the intravenous infusion flow rate, θ is the intravenous infusion flow rate adjustment index and θ>0, which is used to adjust the sensitivity of intravenous infusion flow rate decrease, and F new 'Set the new IV flow rate for patients with potential abnormal changes in heart rate, ensuring that the IV flow rate decreases in a nonlinear manner as the degree of abnormality increases, is an adjustment factor less than 1; Step 3: adjust the flow rate of the infusion device to the new intravenous infusion flow rate, and continuously monitor the patient's heart rate changes within the new short-term monitoring window. At the end of each new monitoring window, reacquire the heart rate data, update the potential abnormal coefficient of the heart rate, repeat steps 1 and 2, and dynamically adjust the actual intravenous infusion flow rate according to the latest intravenous infusion flow rate and abnormal degree coefficient; Step 4: If, after multiple adjustments, the heart rate potential abnormality coefficient continues to be higher than the heart rate potential abnormality coefficient reference threshold, and the abnormality degree coefficient exceeds the preset alarm threshold, the alarm mechanism is triggered to notify medical personnel to intervene.

7. An information-based tracking system for intravenous therapy based on hospital intravenous therapy, used to implement the information-based tracking method for intravenous therapy based on hospital intravenous therapy as described in any one of claims 1 to 6, characterized in that: It includes a starting flow rate setting module, a heart rate monitoring module, a heart rate characteristic analysis module, a heart rate change classification module, a flow rate enhancement module, and a flow rate adjustment and alarm module; In the initial flow rate setting module, medical staff first set a starting intravenous infusion flow rate for the patient based on the patient's condition, weight, age, drug characteristics, and treatment goals; Heart rate monitoring module, which uses monitoring equipment to obtain the patient's heart rate information in real time during intravenous infusion; The heart rate feature analysis module extracts key features after preprocessing the acquired raw heart rate data. In a short-term monitoring window, it performs abnormal analysis on the extracted heart rate features. Subsequently, it uses a pre-trained machine learning model to comprehensively evaluate and predict the analyzed heart rate features. The heart rate change classification module divides the patient's heart rate changes in a short-term monitoring window into potential abnormal changes and regular changes based on the prediction results of the machine learning model; The flow rate increase module increases the patient's actual infusion flow rate within a safe intravenous infusion flow rate range if the heart rate change is classified as a normal change; The flow rate adjustment and alarm module, if the heart rate change is identified as a potential abnormal change, the starting intravenous infusion flow rate will be lowered according to the degree of abnormality, and the patient's heart rate changes will continue to be observed to ensure that the patient is safely infused when the heart rate changes normally. If the heart rate still does not return to normal, an alarm will be issued to notify medical staff.

Images