Infusion catheter risk prediction method and infusion catheter
By constructing a risk prediction model based on deep learning algorithms, evaluating the risks of infusion catheters, solving the shortcomings of relying on experience and intuition in the existing technology, and achieving scientific assessment and personalized prediction of risks in infusion catheter use.
Patent Information
- Application Number
- CN202411698088.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-26
- Publication Date
- 2025-03-04
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing infusion catheter risk prediction methods mostly rely on experience and intuition, lack scientific and effective evaluation mechanisms, and insufficient consideration of individual differences between different patients.
By obtaining the patient's historical medical record data and physiological parameter data, the risk impact data of the infusion catheter was obtained, and a risk prediction model was constructed based on deep learning algorithms to evaluate the patient's infusion catheter risk.
A scientific and effective assessment of the risks in the use of infusion catheters was achieved, and individual differences between different patients were taken into account. The increased risk was detected in a timely manner, early intervention was carried out to avoid serious complications and adverse consequences.
Smart Images
Figure CN119560156A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of medical technology, and more specifically, to a risk prediction method for an infusion catheter and an infusion catheter. Background Art
[0002] Infusion catheters are widely used in clinical treatment, especially in the fields of intravenous injection, nutritional support, antibiotic infusion, etc. Although infusion catheters play an important role in treatment, certain risks may still occur during their use, including but not limited to catheter-related infection, thrombosis, catheter blockage, and abnormal catheter position. Existing risk prediction methods mostly rely on experience and intuition, lack a scientific and effective evaluation mechanism, and do not take into account the individual differences of different patients. Therefore, how to predict the risks in the use of infusion catheters through data analysis and intelligent algorithms has become an urgent problem to be solved in the current medical field.
[0003] The above disclosed technical solutions have at least the following technical problems: Existing risk prediction methods mostly rely on experience and intuition, lack a scientific and effective evaluation mechanism, and do not adequately consider individual differences among patients. In view of the above problems, the present invention proposes a solution. Summary of the invention
[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a risk prediction method and an infusion catheter for an infusion catheter. By establishing a risk prediction model, the problems that the existing risk prediction methods mostly rely on experience and intuition, lack a scientific and effective evaluation mechanism, and do not take into account the individual differences of different patients are solved.
[0005] To achieve the above object, the present invention provides the following technical solutions: A risk prediction method for an infusion catheter comprises obtaining historical medical record data and physiological parameter data of a patient, obtaining risk impact data of the infusion catheter, wherein the risk impact data comprises an infection risk coefficient, an infusion catheter abnormality coefficient, and a catheter obstruction risk coefficient; constructing a risk prediction model based on a deep learning algorithm according to the risk impact data; and evaluating the patient's infusion catheter risk according to the risk prediction model.
[0006] In a preferred embodiment, the steps for obtaining the infection risk coefficient are as follows: obtaining various data of the patient, including but not limited to white blood cell count, C-reactive protein, catheter type, clinical symptoms and immune status; processing the various data of the patient, and extracting characteristic variables that affect the infection risk based on a neural network model, wherein the characteristic variables that affect the infection risk include the site of catheter insertion, catheter type, immune status and clinical symptoms; and calculating based on the infection risk coefficient formula according to the characteristic variables to obtain the infection risk coefficient.
[0007] In a preferred embodiment, the steps for obtaining the abnormal coefficient of the infusion catheter are as follows: prepare multiple groups of infusion catheters to be tested, perform a sealing test on one group of infusion catheters, if the sealing is normal, proceed to the next step, if it fails, replace the infusion catheter and retest; obtain the infusion catheter data after the air in the tube is removed under normal sealing, the infusion catheter data includes the liquid flow rate and the pressure in the tube; pre-process the infusion catheter data after the air in the tube is removed under normal sealing, remove the abnormal data, and add and average the cleaned data to obtain the characteristics of the liquid flow rate and the pressure in the tube. The invention relates to a method for measuring the influence of a liquid infusion tube on a patient's condition of using a liquid infusion tube, and a method for measuring the influence of a liquid infusion tube on a patient's condition of using a liquid infusion tube. The method comprises the following steps: selecting a group of infusion tubes to be tested, injecting different amounts of gas into the tubes when the infusion tubes are in use, and recording the data of the infusion tubes under different gas content conditions in the tubes; performing linear regression analysis on the infusion tube data after the air in the tubes is discharged under normal sealing conditions and the infusion tube data under different gas content conditions to obtain a gas influence coefficient; obtaining the liquid flow rate and the pressure inside the tube in real time during the use of the infusion tube by the patient; and calculating the infusion tube anomaly coefficient based on a preset infusion tube anomaly coefficient formula according to the liquid flow rate, the pressure inside the tube and the gas influence coefficient to obtain the infusion tube anomaly coefficient.
[0008] In a preferred embodiment, the steps for obtaining the catheter obstruction risk coefficient are as follows: obtaining the patient's medical history, and finding the patient's independent risk factors based on the patient's medical history, wherein the independent risk factors include a history of thrombosis, diabetes, D-dimer, and parenteral nutrition; using a logical scoring method to assign values to the independent risk factors, and using the quartile method to grade the risk level; the quartile method grades the risk level into 0 for low risk, 1 to 3 for medium risk, and 4 to 6 for high risk; the catheter obstruction risk coefficient is calculated based on the catheter obstruction risk coefficient formula according to the quartile method for grading the risk level, and the catheter obstruction risk coefficient is obtained; based on the Hosmer-Lemeshow goodness of fit test and the receiver operating characteristic curve, the established catheter obstruction risk coefficient formula model is verified in the modeling group and the verification group.
[0009] In a preferred embodiment, the patient's infusion catheter risk is evaluated according to the risk prediction model, specifically as follows: the patient's infusion catheter risk is evaluated according to the risk prediction model to obtain a risk prediction coefficient; if the risk prediction coefficient reaches a preset value, an alarm will be sent to the system center to remind the nurse to examine and care for the patient; after the first evaluation, moderate risk patients are evaluated once a week, severe risk patients are evaluated twice a week, and changes in their condition are evaluated at any time until extubation, and ICU patients are evaluated once a day, and changes in their condition are evaluated at any time until extubation.
[0010] An infusion catheter of an infusion catheter risk prediction method includes a data acquisition module, a risk prediction model construction module and an evaluation module, and the modules are connected; the data acquisition module is used to obtain the patient's historical medical record data and physiological parameter data, and obtain the risk impact data of the infusion catheter, and the risk impact data includes an infection risk coefficient, an infusion catheter abnormality coefficient and a catheter obstruction risk coefficient; the risk prediction model construction module is used to construct a risk prediction model based on a deep learning algorithm according to the risk impact data; the evaluation module is used to evaluate the patient's infusion catheter risk according to the risk prediction model.
[0011] Technical effects and advantages of a risk prediction method for an infusion catheter of the present invention: 1. Through regular monitoring and calculation of risk prediction coefficients, the present invention allows the medical team to promptly detect the increased risk of infection or catheter abnormalities in patients, thereby conducting early intervention to prevent the risk from turning into serious complications and adverse consequences. For example, if the prediction coefficient shows that a patient has a higher risk, the doctor can strengthen monitoring, adjust the infusion line, or take other preventive measures to reduce the occurrence of infection and catheter-related complications.
[0012] 2. Infection, catheter obstruction and infusion catheter abnormality are common and important risk factors in clinical practice. Each of these three factors has a significant impact on the patient's health. Considering one factor alone may not fully reflect the risk, but combining them together can provide a more accurate prediction. For example, some patients may already be at risk of infection, but if the risk of infusion catheter abnormality is added, the combined impact may lead to a higher incidence of complications. Through comprehensive evaluation, this method can more accurately foresee the health threats that patients may face.
[0013] 3. By combining the patient's historical medical data, physiological parameter data and infusion catheter risk impact data, the system can comprehensively consider various influencing factors, including infection risk, catheter abnormality risk and catheter blockage risk, and predict risks from multiple dimensions. Using deep learning algorithms to build a risk prediction model can achieve accurate modeling of complex and nonlinear relationships, ensuring high accuracy of the evaluation results.
[0014] 4. The present invention can perform personalized assessments based on the unique situation of each patient by acquiring the patient's physiological data and relevant clinical information in real time. The disease background and physical differences of different patients can be reflected through algorithm modeling, avoiding a "one-size-fits-all" risk assessment and improving the pertinence and practicality of risk prediction. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 It is a schematic structural diagram of a risk prediction method for an infusion catheter of the present invention.
[0016] Figure 2 A schematic diagram of the structure of an infusion catheter according to a risk prediction method for an infusion catheter of the present invention. DETAILED DESCRIPTION
[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0018] Embodiment 1, Figure 1 A risk prediction method for an infusion catheter is provided in the present invention, which is as follows: S1, obtaining the patient's historical medical record data and physiological parameter data, and obtaining the risk impact data of the infusion catheter, wherein the risk impact data includes an infection risk coefficient, an infusion catheter abnormality coefficient, and a catheter obstruction risk coefficient.
[0019] In this embodiment, the patient's historical medical record data and physiological parameter data are obtained to obtain the risk impact data of the infusion catheter, and the risk impact data includes an infection risk coefficient, an infusion catheter abnormality coefficient, and a catheter obstruction risk coefficient, as follows: Infection risk coefficient refers to the probability of infection after contact with the source of infection. In addition, the infection risk coefficient can also be measured by relative risk, that is, the ratio of the incidence rate of the exposed group to the incidence rate of the non-exposed group. This coefficient is used to assess the impact of specific risk factors on the possibility of infection, thereby helping to formulate preventive measures and response strategies.
[0020] The steps for obtaining the infection risk coefficient are as follows: Acquiring various data of the patient, including but not limited to white blood cell count, C-reactive protein, catheter type, clinical symptoms and immune status; Processing various types of patient data, and extracting characteristic variables that affect infection risk based on a neural network model, wherein the characteristic variables that affect infection risk include catheter insertion site, catheter type, immune status, and clinical symptoms; The infection risk coefficient is calculated based on the infection risk coefficient formula according to the characteristic variables to obtain the infection risk coefficient.
[0021] The calculation formula of the infection risk coefficient is as follows:
[0022] The calculation formula of the activation function is as follows:
[0023] In the formula, is the infection risk factor, is the activation function, is the number of feature variables, is the weight of the ith feature, is the ith feature variable, is the intercept.
[0024] It should be noted that the role of the activation function is to convert the linear result into a probability value. The value range of the infection risk coefficient is [0,1]. The smaller the infection risk coefficient, the smaller the probability of infection risk.
[0025] The infusion catheter abnormality coefficient refers to the frequency of infusion failure or interruption during the infusion process due to catheter malfunction, blockage, detachment and infection. In addition, catheter abnormalities also include catheter blockage, infection and other situations. These abnormal events require medical staff to detect them in time and take corresponding measures to deal with them.
[0026] The steps for obtaining the abnormal coefficient of the infusion catheter are specifically as follows: Prepare multiple groups of infusion catheters to be tested, and perform a sealing test on one group of infusion catheters. If the sealing is normal, proceed to the next step. If it fails, replace the infusion catheter and retest; Acquire the infusion catheter data after the air in the tube is removed under normal sealing, wherein the infusion catheter data includes the liquid flow rate and the pressure in the tube; Preprocessing the infusion catheter data after the air in the normal sealing condition is removed, removing abnormal data, and adding and averaging the cleaned data to obtain characteristic data of liquid flow rate and pressure in the tube; Select a group of infusion catheters to be tested, inject different amounts of gas into the tubes when the infusion catheters are in use, and record the infusion catheter data under different gas content conditions in the tubes; The gas influence coefficient was obtained by performing linear regression analysis based on the data of the infusion catheter after the air inside was removed under normal sealing and the data of the infusion catheter under different gas content conditions; Obtain the liquid flow rate and pressure inside the tube in real time when the patient is using the infusion tube; The infusion catheter abnormality coefficient is calculated based on a preset infusion catheter abnormality coefficient formula according to the liquid flow rate, the pressure in the tube and the gas influence coefficient to obtain the infusion catheter abnormality coefficient.
[0027] The calculation formula of the infusion catheter abnormality coefficient is as follows:
[0028]
[0029]
[0030] In the formula, is the infusion catheter abnormality coefficient, is the abnormality of the flow velocity, is the abnormal degree of pressure, is the gas influence coefficient, and is the weight coefficient, is the real-time liquid flow rate, is the liquid flow rate under normal conditions, is the real-time pressure inside the tube, is the pressure inside the tube under normal conditions.
[0031] Catheter occlusion risk factor refers to the risk assessment index of catheter obstruction due to thrombosis, mechanical blockage or other factors during the use of central venous catheter (CVC). Through multivariate logistic regression analysis, history of thrombosis, diabetes, D-dimer and intravenous nutrition were identified as independent risk factors for CVC obstruction.
[0032] The steps for obtaining the catheter obstruction risk coefficient are specifically as follows: Obtaining the patient's medical history, and finding the patient's independent risk factors according to the patient's medical history, wherein the independent risk factors include a history of thrombosis, diabetes, D-dimer, and parenteral nutrition; The independent risk factors were assigned values using the logical scoring method, and the risk levels were graded using the quartile method; The quartile method classifies the risk level into 0 for low risk, 1 to 3 for medium risk, and 4 to 6 for high risk; The risk level is graded according to the quartile method and the catheter obstruction risk coefficient is calculated based on the catheter obstruction risk coefficient formula; The established catheter obstruction risk coefficient formula model was verified in the modeling group and the verification group based on the Hosmer-Lemeshow goodness-of-fit test and the receiver operating characteristic curve.
[0033] The calculation formula of the catheter obstruction risk coefficient is as follows:
[0034] In the formula, is the catheter occlusion risk factor, is the weight coefficient, is the risk level classification of independent risk factors, and n is the number of independent risk factors S2, constructing a risk prediction model based on the risk impact data and a deep learning algorithm.
[0035] In this embodiment, the risk prediction model is constructed based on the risk impact data based on the deep learning algorithm, as follows: The calculation formula of the risk prediction model is as follows:
[0036] In the formula, is the risk prediction coefficient, , , , and is the weight coefficient, is the infection risk factor, is the infusion catheter abnormality coefficient, is the risk factor for catheter occlusion.
[0037] It should be noted that the index It means that the risk of infection may increase exponentially with the increase of certain factors, while the logarithmic term It plays a smoothing role when the risk is low, avoiding too sharp an increase. The combination of the exponential and square root takes into account the nonlinear effect of catheter abnormalities, and the square root term acts as a smoothing factor when the risk is low. An inverse format was used to indicate that the increase in total risk would be more gradual when the risk of catheter occlusion was low and would be steeper when the risk of catheter occlusion was high.
[0038] Further, the advantages of obtaining the risk prediction coefficient through the infection risk coefficient and the infusion catheter abnormality coefficient are analyzed: The risk prediction coefficient provides an effective tool for early identification of potential health problems. By regularly monitoring and calculating this coefficient, the medical team can promptly detect the patient's increased risk of infection or catheter abnormalities, thereby conducting early intervention to prevent the risk from turning into serious complications and adverse consequences. For example, if the prediction coefficient shows that a patient is at a higher risk, the doctor can strengthen monitoring, adjust the infusion line, or take other preventive measures to reduce the occurrence of infection and catheter-related complications; Infection, catheter obstruction, and infusion catheter abnormalities are common and important risk factors in clinical practice. Each of these three factors has a significant impact on the patient's health. Considering one factor alone may not fully reflect the risk, but combining them together can provide a more accurate prediction. For example, some patients may already be at risk of infection, but if the risk of infusion catheter abnormalities is added, the combined impact may lead to a higher incidence of complications. Through comprehensive evaluation, this method can more accurately foresee the health threats that patients may face; By combining the patient's historical medical data, physiological parameter data and infusion catheter risk impact data, the system can comprehensively consider various influencing factors, including infection risk, catheter abnormality risk and catheter blockage risk, and make risk predictions from multiple dimensions. By using deep learning algorithms to build a risk prediction model, it can achieve accurate modeling of complex and nonlinear relationships, ensuring the high accuracy of the evaluation results.
[0039] S3, assessing the patient's infusion catheter risk based on the risk prediction model.
[0040] In this embodiment, the patient's infusion catheter risk is assessed according to the risk prediction model, as follows: The patient's infusion catheter risk is assessed according to the risk prediction model to obtain a risk prediction coefficient; If the risk prediction coefficient reaches the preset value, an alarm will be sent to the system center to remind the nurse to examine and care for the patient; After the first assessment, patients with moderate risk will be assessed once a week, and those with severe risk will be assessed twice a week. Changes in their condition will be assessed at any time until extubation. ICU patients will be assessed once a day, and changes in their condition will be assessed at any time until extubation.
[0041] Embodiment 2, Figure 2 The invention provides an infusion catheter of a risk prediction method for an infusion catheter, comprising a data acquisition module, a risk prediction model building module and an evaluation module, wherein the modules are connected; A data acquisition module, used to obtain the patient's historical medical data and physiological parameter data, and obtain the risk impact data of the infusion catheter, wherein the risk impact data includes an infection risk coefficient, an infusion catheter abnormality coefficient, and a catheter obstruction risk coefficient; A risk prediction model building module, used to build a risk prediction model based on the risk impact data and a deep learning algorithm; An assessment module is used to assess the patient's infusion catheter risk based on the risk prediction model.
[0042] 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.
[0043] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.
[0044] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0045] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.
[0046] 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.
[0047] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A risk prediction method for an infusion catheter, characterized in that: include: Acquire the patient's historical medical record data and physiological parameter data, and acquire risk impact data of the infusion catheter, wherein the risk impact data includes an infection risk coefficient, an infusion catheter abnormality coefficient, and a catheter obstruction risk coefficient; Constructing a risk prediction model based on the risk impact data and a deep learning algorithm; The patient's infusion catheter risk is assessed based on the risk prediction model.
2. The risk prediction method for an infusion catheter according to claim 1, characterized in that: The steps for obtaining the infection risk coefficient are as follows: Acquiring various data of the patient, including but not limited to white blood cell count, C-reactive protein, catheter type, clinical symptoms and immune status; Processing various types of patient data, and extracting characteristic variables that affect infection risk based on a neural network model, wherein the characteristic variables that affect infection risk include catheter insertion site, catheter type, immune status, and clinical symptoms; The infection risk coefficient is calculated based on the infection risk coefficient formula according to the characteristic variables to obtain the infection risk coefficient.
3. The risk prediction method for an infusion catheter according to claim 2, characterized in that: The steps for obtaining the abnormal coefficient of the infusion catheter are specifically as follows: Prepare multiple groups of infusion catheters to be tested, and perform a sealing test on one group of infusion catheters. If the sealing is normal, proceed to the next step. If it fails, replace the infusion catheter and retest; Acquire the infusion catheter data after the air in the tube is removed under normal sealing, wherein the infusion catheter data includes the liquid flow rate and the pressure in the tube; Preprocessing the infusion catheter data after the air in the normal sealing condition is removed, removing abnormal data, and adding and averaging the cleaned data to obtain characteristic data of liquid flow rate and pressure in the tube; Select a group of infusion catheters to be tested, inject different amounts of gas into the tubes when the infusion catheters are in use, and record the infusion catheter data under different gas content conditions in the tubes; The gas influence coefficient was obtained by performing linear regression analysis based on the data of the infusion catheter after the air was removed under normal sealing and the data of the infusion catheter under different gas content conditions; Obtain the liquid flow rate and pressure inside the tube in real time when the patient is using the infusion tube; The infusion catheter abnormality coefficient is calculated based on a preset infusion catheter abnormality coefficient formula according to the liquid flow rate, the pressure in the tube and the gas influence coefficient to obtain the infusion catheter abnormality coefficient.
4. The risk prediction method for an infusion catheter according to claim 3, characterized in that: The steps for obtaining the catheter obstruction risk coefficient are specifically as follows: Obtaining the patient's medical history, and finding the patient's independent risk factors according to the patient's medical history, wherein the independent risk factors include a history of thrombosis, diabetes, D-dimer, and parenteral nutrition; The independent risk factors were assigned values using the logical scoring method, and the risk levels were graded using the quartile method; The quartile method classifies the risk level into 0 for low risk, 1 to 3 for medium risk, and 4 to 6 for high risk; The risk level is graded according to the quartile method and the catheter obstruction risk coefficient is calculated based on the catheter obstruction risk coefficient formula; The established catheter obstruction risk coefficient formula model was verified in the modeling group and the verification group based on the Hosmer-Lemeshow goodness-of-fit test and the receiver operating characteristic curve.
5. The risk prediction method for an infusion catheter according to claim 4, characterized in that: The patient's infusion catheter risk is assessed according to the risk prediction model, specifically as follows: The patient's infusion catheter risk is assessed according to the risk prediction model to obtain a risk prediction coefficient; If the risk prediction coefficient reaches the preset value, an alarm will be sent to the system center to remind the nurse to examine and care for the patient; After the first assessment, patients with moderate risk will be assessed once a week, and those with severe risk will be assessed twice a week. Changes in their condition will be assessed at any time until extubation. ICU patients will be assessed once a day, and changes in their condition will be assessed at any time until extubation.
6. The risk prediction method for an infusion catheter according to claim 5, characterized in that: The calculation formula of the infection risk coefficient is as follows: The calculation formula of the activation function is as follows: In the formula, is the infection risk factor, is the activation function, is the number of feature variables, is the weight of the ith feature, is the ith feature variable, is the intercept.
7. A risk prediction method for an infusion catheter according to claim 6, characterized in that: The calculation formula of the infusion catheter abnormality coefficient is as follows: In the formula, is the infusion catheter abnormality coefficient, is the abnormality of the flow velocity, is the abnormal degree of pressure, is the gas influence coefficient, and is the weight coefficient, is the real-time liquid flow rate, is the liquid flow rate under normal conditions, is the real-time pressure inside the tube, is the pressure inside the tube under normal conditions.
8. The risk prediction method for an infusion catheter according to claim 7, characterized in that: The calculation formula of the catheter obstruction risk coefficient is as follows: In the formula, is the catheter occlusion risk factor, is the weight coefficient, is the risk level classification of independent risk factors, and n is the number of independent risk factors.
9. The risk prediction method for an infusion catheter according to claim 8, characterized in that: The calculation formula of the risk prediction model is as follows: In the formula, is the risk prediction coefficient, , and is the weight coefficient, is the infection risk factor, is the infusion catheter abnormality coefficient, is the risk factor for catheter occlusion.
10. An infusion catheter using the risk prediction method for an infusion catheter according to any one of claims 1 to 9, comprising a data acquisition module, a risk prediction model building module and an evaluation module, wherein the modules are connected; A data acquisition module, used to obtain the patient's historical medical data and physiological parameter data, and obtain the risk impact data of the infusion catheter, wherein the risk impact data includes an infection risk coefficient, an infusion catheter abnormality coefficient, and a catheter obstruction risk coefficient; A risk prediction model building module, used to build a risk prediction model based on the risk impact data and a deep learning algorithm; An assessment module is used to assess the patient's infusion catheter risk based on the risk prediction model.