Microbiological test-based double-j tube encrustation infection risk prediction method and system

Abstract

The present application belongs to the technical field of clinical big data processing, and particularly relates to a double J tube film bacteria infection risk prediction method and system based on microbiological examination. Through collecting patient basic data, clinical parameters and physiological indexes, the inhibitory efficiency of the drug on the film bacteria is evaluated through fosfomycin trometamol intervention experiment; microbiological examination and film bacteria quantitative analysis are performed on urine and tube samples; independent risk factors of infection are identified by applying logistic regression analysis; finally, a weighted risk prediction model is constructed, the probability score is calculated, the risk level is divided, and a preventive intervention suggestion report is generated. The present application realizes early identification and quantitative evaluation of infection risk by deeply mining the correlation between microbiological examination data and clinical multi-dimensional indexes, provides evidence-based medical support for preventing urinary tract infection, significantly improves the risk warning accuracy and optimizes the perioperative infection prevention and control system, and has important significance for improving the prognosis quality of urology.

Description

Technical Field

[0001] This invention belongs to the field of clinical big data processing technology, specifically relating to a method and system for predicting the risk of double-J tube capillary bacterial infection based on microbial testing. Background Technology

[0002] With the continuous evolution of endoscopic urological surgery techniques, double-J stents, as ureteral stents, have been widely used in clinical scenarios such as relieving urinary tract obstruction and postoperative drainage, becoming a core means of ensuring the recovery of urinary system function. Biofilm formation and related infections caused by indwelling medical devices remain significant challenges in clinical medicine. The biofilm formed by microorganisms colonizing the catheter surface and secreting extracellular polymers not only protects pathogens from host immune attacks but also significantly enhances bacterial resistance to commonly used antibiotics, greatly increasing the difficulty of treatment and the medical burden on patients.

[0003] Early prevention and risk assessment of double-J duct capsular infections have become key technologies for improving postoperative prognosis. Currently, clinical practice mainly relies on routine urine culture for infection monitoring and attempts to inhibit bacterial colonization through drug intervention. Fosfomycin tromethamine, as a drug that can inhibit bacterial cell wall synthesis and has biofilm penetration capabilities, has extremely high drug concentrations in urine and is considered a potential strategy for the prevention and treatment of urinary tract infections. However, how to combine microbiological test results with multidimensional clinical indicators to construct an accurate risk prediction model and verify the effectiveness of prevention strategies remains a research focus in this field.

[0004] Current technologies still face significant challenges in monitoring and controlling biofilm infections. Traditional clinical testing methods often lag behind the actual infection process, lacking in-depth understanding of the dynamic colonization patterns of microorganisms, resulting in short intervention windows and limited effectiveness. Furthermore, the integrated utilization rate of patients' basic physiological indicators and clinical diagnostic parameters is low, failing to establish a multi-factor coupled risk assessment mechanism, making early identification of high-risk groups for biofilm infection difficult. In addition, the lack of systematic evidence-based evaluation of specific dosing regimens in preventing biofilm formation in clinical applications leads to often untargeted medication strategies, failing to effectively curb the spread of highly drug-resistant bacteria.

[0005] Therefore, a risk prediction scheme for double-J tube membrane infection based on microbial testing is desired. Summary of the Invention

[0006] To achieve the above objectives, this application provides the following technical solution: A method for predicting the risk of bacterial infection on double-J tube membranes based on microbial testing includes the following steps: Step 1: Obtain basic information, clinical diagnostic and treatment parameters, and physiological index data during follow-up of patients undergoing double-J stent implantation to establish an initial patient information database; Step 2: Patients were randomly divided into an intervention group and a control group. The intervention group received fosfomycin tromethamine, while the control group did not receive any drug intervention. The urinary tract status of both groups was continuously monitored during the stent placement period. Step 3: When the predetermined double-J stent removal cycle arrives, collect the patient's midstream urine sample and the removed double-J stent body sample, and transport and preprocess them. Step 4: Perform bacterial culture and quantitative analysis on urine samples using microbial testing techniques to determine the positive rate of urine bacterial culture and colony distribution characteristics; Step 5: Ultrasonic elution combined with microfiltration is used to extract the biofilm on the surface of the double-J tube, and quantitative culture and identification of the biofilm are performed to determine the positive rate of biofilm culture and the biofilm coverage density. Step 6: Using statistical difference analysis, compare the numerical differences between the intervention group and the control group in terms of positive rates of urine bacterial culture and positive rates of biofilm culture, and evaluate the inhibitory efficacy of fosfomycin tromethamine on biofilm formation. Step 7: Integrate the patients' basic and clinical data, extract the characteristic differences between the positive and negative groups of capsular infections, and identify independent risk factors leading to capsular infections through logistic regression analysis; Step 8: Construct a weighted risk prediction model based on the identified independent risk factors, calculate the probability score of patients developing capsular infections, and define different risk level intervals according to the probability score to generate a preventive intervention recommendation report.

[0007] Furthermore, the basic information obtained in step 1 includes the patient's age, gender, body mass index, presence of underlying metabolic diseases, and history of urinary tract infection; The clinical diagnostic and treatment parameters include the type of surgery, duration of double-J stent placement, and type and dosage of postoperative antibiotics. The physiological indicators include urine pH, urine specific gravity, and white blood cell count. The specific fosfomycin tromethamine dosing regimen implemented in step 2 for the intervention group is as follows: At the predetermined time point after double-J stent implantation, patients are instructed to take a preset dose of fosfomycin tromethorphan powder orally, and to take the medication periodically at a preset frequency during the subsequent indwelling period, until the medication is stopped 24 hours before the double-J stent is removed. Fosfomycin tromethamine blocks the production of muramic acid by inhibiting the activity of pyruvyltransferase in bacterial cell wall synthesis. The concentration of fosfomycin tromethamine in urine is maintained above the effective concentration required to inhibit bacterial adhesion after administration. It penetrates the initially formed biofilm matrix through osmosis and disrupts the metabolic environment of the biofilm-covered bacteria.

[0008] Furthermore, the process of collecting the double-J tube body sample in step 3 includes: Under aseptic conditions, the removed double-J tube is placed in a sterile sampling container containing a preset volume of physiological saline. The tube is then cut into multiple segments of equal length by mechanical cutting. These segments include a proximal segment near the kidney, a mid-segment segment, and a proximal segment near the bladder, in order to assess biofilm load at multiple sites. Step 4, the process of culturing bacteria in the urine sample, includes: The collected midstream urine was inoculated onto pre-defined blood agar and MacConkey agar, and placed in a pre-defined constant temperature incubator for 48 hours of static culture. The number of colony-forming units per unit volume was recorded. When the number of colonies exceeded the pre-defined concentration threshold, the urine bacterial culture was considered positive. The identification of urine bacteria in step 4 also includes using a fully automated microbial mass spectrometry detection system to compare with a peptide spectrum library to identify specific pathogenic bacterial species, including Escherichia coli, Enterococcus faecalis, Klebsiella pneumoniae, and Staphylococcus epidermidis, and recording the correlation strength between different bacterial species and the formation of double-J tube membranes.

[0009] Furthermore, the process of extracting the film-coated bacteria in step 5 includes: The cut double-J tube segment was placed in a centrifuge tube containing eluent and subjected to ultrasonic cleaning at a preset frequency to physically break down the biofilm structure attached to the inside and outside of the tube wall and release colonizing bacteria into the eluent. The bacteria are evenly distributed by vortexing. A predetermined volume of elution solution is serially diluted, inoculated onto a solid culture medium, and incubated at a constant temperature. The original bacterial load of the film is calculated based on the colony count. The determination of biofilm coverage density in step 5 also includes observing the microstructure of the surface of the double-J tube using a scanning electron microscope and identifying the area ratio of the biofilm-covered region through image processing algorithms. The image processing algorithm includes grayscale processing of the scanning electron microscope image, calculating the optimal segmentation threshold using the Otsu method, dividing the image into a biofilm region and a pipe matrix region, calculating the ratio of the number of pixels in the biofilm region to the total number of pixels to obtain the biofilm coverage density, and using the biofilm coverage density as an auxiliary indicator for evaluating the severity of infection.

[0010] Furthermore, the process of evaluating inhibitory efficacy in step 6 includes: The decrease in the positive rate in the intervention group and the control group was calculated, and the significance of the decrease was determined by a significance test. The decrease is equal to the difference between the positive rate of the control group and the positive rate of the intervention group, divided by the positive rate of the control group; When the significance level variable is less than the preset probability threshold, fosfomycin tromethamine is determined to have significant efficacy in preventing urinary tract infections and inhibiting the colonization of tunica albuginea. Data analysis software was used to examine the balance between the two groups of patients to ensure that the intervention group and the control group were comparable in terms of age, disease distribution, and confounding factors. The process of identifying independent risk factors in step 7 includes: Univariate analysis was performed to screen out candidate variables that showed significant differences between the positive and negative groups of capsular infection; Candidate variables are substituted into the multivariate logistic regression equation, and confounding factors are eliminated through stepwise regression. Variables with regression coefficients greater than the preset threshold and statistical significance are identified as independent risk factors. The independent risk factors include double-J stent placement for more than a preset number of days, uncontrolled diabetes, urine pH consistently outside the preset range, and lack of postoperative prophylactic medication.

[0011] Furthermore, the logic for constructing the weighted risk prediction model in step 8 includes: Each independent risk factor is assigned a weight coefficient proportional to its regression coefficient. The quantitative values ​​of each patient's indicators are multiplied by the corresponding weight coefficients and summed to obtain the original risk score. The original risk score is mapped to a probability range between zero and one by using a logical function to obtain the probability of infection with capsular bacteria. During the calculation process, different baseline correction coefficients were used for patients of different genders to eliminate the systematic bias caused by physiological differences in infection risk prediction. The risk level range is divided into low-risk, medium-risk and high-risk areas; When the risk probability is less than the first preset threshold, it is defined as low risk and routine clinical observation is carried out. When the risk probability is between the first preset threshold and the second preset threshold, it is defined as medium risk, and the frequency of urine testing is increased and prophylactic medication is administered. When the risk probability is greater than the second preset threshold, it is defined as high risk, and an enhanced intervention program with fosfomycin tromethamine as the core is implemented.

[0012] Furthermore, the preventive intervention recommendation report includes personalized extubation timing recommendations for specific patients, recommendations for optimized dosing cycles of fosfomycin tromethamine, and postoperative lifestyle guidance. The method also includes establishing a feedback database containing clinical records, periodically transmitting actual infection case data back to the risk prediction model, and dynamically adjusting the weight coefficients using an iterative learning algorithm. Whenever a preset number of new clinical cases are accumulated, the model automatically triggers retraining logic to update the weights based on the gradient bias of the new samples, so as to improve the prediction accuracy of the model under different population distributions. The reports generated by the risk prediction model are automatically synchronized to the hospital information system via an electronic interface, providing real-time decision support for clinicians. For high-risk patients, intervention recommendations also include adjusting the perfusion pressure during the surgical procedure to keep it below a preset pressure threshold, and increasing the postoperative fluid resuscitation rate as suggested by the model to reduce the probability of bacterial colonization through physical flushing.

[0013] Furthermore, the method also involves monitoring the resistance to fosfomycin tromethamine, determining the minimum inhibitory concentration of fosfomycin on the extracted cloacal bacteria through drug susceptibility testing, and integrating the resistance evolution trend into a risk prediction model to correct the expected long-term intervention effect. Step 3 requires that the number of settled bacteria and airborne bacteria in the environment be lower than the preset cleanliness standard throughout the entire process of collection, transportation and laboratory processing, to ensure that all detected bacteria originate from the patient sample itself. The process of extracting risk factors in step 7 eliminates irregular fluctuations in data caused by sampling errors or random laboratory errors, and improves the signal-to-noise ratio of the input data through a smoothing algorithm. The method also studies the relationship between the double-J tube material and the adhesion of the biofilm, introduces the scaffold material type variable into the risk prediction model, assigns different initial basic risk values ​​to double-J tubes of different materials, and refines the calculation logic of risk scoring.

[0014] Furthermore, the method also includes quantitative monitoring of the effective concentration of fosfomycin in the patient's urine, measuring the urine drug concentration curve at different time points after administration using high performance liquid chromatography, and analyzing the positive correlation between the peak urine drug concentration and the inhibition rate of the biofilm. When the monitored urinary drug concentration is lower than the preset minimum effective inhibitory concentration, the patient's infection risk score is increased in the risk prediction model; The calculation logic of the risk probability score introduces a time decay factor to consider the nonlinear increasing effect of material aging and decreased biocompatibility of the double-J tube on the risk of infection as the retention time increases. The time decay factor is set as a base greater than one and an exponential growth term of the number of days of detention. As the number of days of detention increases, the original risk score is multiplied by the time decay factor to achieve a non-linear correction of the risk probability. The method for predicting the risk of double-J tube membrane infection based on microbial testing integrates clinical indicators, drug intervention variables, and etiological evidence to construct a closed-loop infection prevention and control system covering the entire diagnosis and treatment cycle. The model is constructed using cross-validation to evaluate its performance. The area under the receiver operating characteristic curve (ROC) of the model is calculated. When the area is greater than a preset performance threshold, the model is deemed to have the reliability for clinical application. By establishing a multi-dimensional risk assessment matrix, laboratory data from microbiology testing is deeply integrated with clinical epidemiological data to achieve quantitative early warning of potential infection risks in the early stages of double-J tube placement. The preventive interventions include advising doctors to adjust the perfusion pressure during surgery and the postoperative fluid resuscitation rate based on the risk level, reducing the initial adhesion of bacteria to the surface of the double-J tube through continuous hydrodynamic flushing, and synergistically reducing the risk of biofilm formation by combining the inhibitory effect of fosfomycin on cell wall synthesis.

[0015] According to a second aspect of the present invention, the present invention claims protection for a double-J tube membrane bacterial infection risk prediction system based on microbial testing, comprising: One or more processors; A memory having stored one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method for predicting the risk of double-J tube membrane infection based on microbial testing.

[0016] This invention belongs to the field of clinical big data processing technology, specifically involving a method and system for predicting the risk of double-J urinary tract endoscopic bacterial infection based on microbial testing. The method involves collecting patient baseline data, clinical parameters, and physiological indicators; evaluating the inhibitory efficacy of fosfomycin tromethamine on endoscopic bacteria through an intervention experiment; performing microbial testing and quantitative analysis of endoscopic bacteria on urine and urinary tract samples; applying logistic regression analysis to identify independent risk factors for infection; and finally constructing a weighted risk prediction model, calculating probability scores, assigning risk levels, and generating a preventive intervention recommendation report. This invention, by deeply exploring the correlation between microbial testing data and multidimensional clinical indicators, achieves early identification and quantitative assessment of infection risk, providing evidence-based medical support for the prevention of urinary tract infections, significantly improving the accuracy of risk warning, and optimizing the perioperative infection control system, which is of great significance for improving the quality of urological prognosis. Attached Figure Description

[0017] Figure 1 The flowchart illustrates the workflow of the double-J tube membrane infection risk prediction method based on microbial testing, as claimed in the embodiments of the present invention. Figure 2This is a second flowchart of the method for predicting the risk of infection of double-J tube membrane bacteria based on microbial testing, as claimed in the embodiments of the present invention. Figure 3 The third flowchart is a method for predicting the risk of infection of double-J tube membrane bacteria based on microbial testing, as claimed in the embodiments of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0019] The terms "first," "second," and "third" in this application are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of those features. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications in the embodiments of this application, such as up, down, left, right, front, back, etc., are only used to explain the relative positional relationships and movements between components in a specific orientation as shown in the accompanying drawings. If the specific orientation changes, the directional indications will change accordingly. Furthermore, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0020] References to embodiments herein mean that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0022] According to a first embodiment of this aspect, the present invention claims protection for a method for predicting the risk of bacterial infection of double-J tube membranes based on microbial testing, with reference to... Figure 1 This includes the following steps: Step 1: Obtain basic information, clinical diagnostic and treatment parameters, and physiological index data during follow-up of patients undergoing double-J stent implantation to establish an initial patient information database; Step 2: Patients were randomly divided into an intervention group and a control group. The intervention group received fosfomycin tromethamine, while the control group did not receive any drug intervention. The urinary tract status of both groups was continuously monitored during the stent placement period. Step 3: When the predetermined double-J stent removal cycle arrives, collect the patient's midstream urine sample and the removed double-J stent body sample, and transport and preprocess them. Step 4: Perform bacterial culture and quantitative analysis on urine samples using microbial testing techniques to determine the positive rate of urine bacterial culture and colony distribution characteristics; Step 5: Ultrasonic elution combined with microfiltration is used to extract the biofilm on the surface of the double-J tube, and quantitative culture and identification of the biofilm are performed to determine the positive rate of biofilm culture and the biofilm coverage density. Step 6: Using statistical difference analysis, compare the numerical differences between the intervention group and the control group in terms of positive rates of urine bacterial culture and positive rates of biofilm culture, and evaluate the inhibitory efficacy of fosfomycin tromethamine on biofilm formation. Step 7: Integrate the patients' basic and clinical data, extract the characteristic differences between the positive and negative groups of capsular infections, and identify independent risk factors leading to capsular infections through logistic regression analysis; Step 8: Construct a weighted risk prediction model based on the identified independent risk factors, calculate the probability score of patients developing capsular infections, and define different risk level intervals according to the probability score to generate a preventive intervention recommendation report.

[0023] In this embodiment, step 1 involves acquiring basic information, clinical parameters, and physiological data during follow-up for patients undergoing double-J stent implantation to establish an initial patient information database. Specifically, the acquired basic information includes the patient's age, gender, body mass index (BMI), presence of underlying metabolic diseases such as diabetes and hypertension, and a detailed record of any history of urinary tract infections. The clinical parameters include the specific type of surgery (e.g., percutaneous nephrolithotomy, ureteroscopic lithotripsy), the total duration of double-J stent placement, and the specific types and dosages of postoperative antibiotics (e.g., cephalosporins, quinolones). The physiological data includes dynamically monitored urine pH, urine specific gravity, and white blood cell count. During the establishment of the initial patient information database, the system performs structuring processing on the acquired unstructured clinical records, extracts keywords using natural language processing technology, and aligns the various physiological indicators with the time nodes of the surgery and follow-up based on time series data. For missing data items, imputation methods based on mean or regression interpolation were used to ensure the integrity of the database and the accuracy of subsequent analysis. The database was constructed using a relational database management system, patient privacy information was anonymized, and a multi-level indexing mechanism was established to improve the efficiency of data retrieval and access.

[0024] In step 2, patients were randomly divided into an intervention group and a control group. The intervention group received a fosfomycin tromethamine administration regimen, while the control group did not receive this drug intervention. The urinary tract status of both groups was continuously monitored during stent placement. Specifically, the fosfomycin tromethamine administration regimen for the intervention group was as follows: at the first predetermined time point after double-J stent implantation, such as within 24 hours post-procedure, patients were instructed to orally administer 3 grams of fosfomycin tromethamine powder. This was continued periodically throughout the placement period at a predetermined frequency of once every 7 days until 24 hours before stent removal. Fosfomycin tromethamine achieves a broad-spectrum bactericidal effect by inhibiting the activity of pyruvyltransferase in bacterial cell wall synthesis, thus blocking the production of muramic acid. Simultaneously, the concentration of this drug in urine remained above the effective concentration required to inhibit bacterial adhesion for a prolonged period after administration, penetrating the initially formed biofilm matrix through osmosis and disrupting the metabolic environment of the biofilm-covered bacteria. During the monitoring process, detailed records were kept of whether the patient experienced urinary tract irritation symptoms such as urinary frequency, urgency, and pain. The color, transparency, and presence of visible flocculent matter in the urine were observed, and this dynamic information was entered into the initial patient information database in real time.

[0025] In step 3, upon arrival of the predetermined double-J stent removal cycle, a midstream urine sample and a sample of the removed double-J stent are collected from the patient, and transported and pretreated strictly according to aseptic procedures. These aseptic procedures require that the number of settling bacteria and airborne bacteria in the environment be below a preset cleanliness standard throughout the entire process of collection, transport, and laboratory processing, ensuring that all detected bacteria originate from the patient's own sample. The specific collection process includes: on the day of double-J stent removal, at least 20 ml of midstream urine is collected from the patient upon waking and placed in a sterile sampling tube. Under aseptic conditions, the removed double-J stent is placed in a sterile sampling container containing 10 ml of physiological saline. Using a sterile scalpel, the stent is mechanically cut into three equal segments: a proximal segment near the kidney, a midstream segment, and a proximal segment near the bladder. All samples are transported to the laboratory via cold chain within 30 minutes of collection, with the transport temperature strictly controlled between 2°C and 8°C to prevent uncontrolled bacterial proliferation or death after removal from the body.

[0026] In step 4, urine samples are cultured and quantitatively analyzed using microbial testing techniques to determine the positive rate and colony distribution characteristics. The specific logic includes: after thorough shaking of the collected midstream urine, 10 μL is inoculated onto blood agar and MacConkey agar, and streaked in four zones using an inoculation loop. The culture medium is then placed in a 35°C incubator for static culture at 5% carbon dioxide for up to 48 hours. After culture, the number of colony-forming units (CFU) per unit volume is recorded using an automated colony counter. A positive urine bacterial culture is defined as a colony culture when the colony count exceeds a concentration threshold of 1000 CFU per milliliter. Furthermore, an automated microbial mass spectrometry system is used to compare with a peptide library to identify specific pathogenic bacterial species, including *Escherichia coli*, *Enterococcus faecalis*, *Klebsiella pneumoniae*, and *Staphylococcus epidermidis*, and the distribution ratio of different species in the intervention and control groups is recorded.

[0027] In step 5, ultrasonic elution combined with microfiltration is used to extract biofilm from the surface of the double-J tubes. Quantitative culture and identification of the biofilm are then performed, and the positive rate and biofilm coverage density are determined. The extraction process specifically includes: placing three cut double-J tube segments into centrifuge tubes containing 5 ml of sterile eluent, and oscillating them for 5 minutes using an ultrasonic cleaner at a frequency of 40 kHz. This process utilizes the cavitation effect of ultrasound to physically break down the biofilm structure attached to the inside and outside of the tube wall, releasing colonizing bacteria into the eluent. Subsequently, a vortex mixer is used for 30 seconds of vigorous oscillation to ensure uniform bacterial distribution. 100 μL of eluent is serially diluted from 10 to 1000 times and inoculated onto solid culture medium for isothermal incubation. The original bacterial load is calculated by multiplying the colony count by the dilution factor. Simultaneously, the microstructure of the double-J tube surface is observed using a scanning electron microscope, and the area ratio of the biofilm-covered region is identified using image processing algorithms. The algorithm first performs grayscale processing on the scanning electron microscope image, calculates the optimal segmentation threshold using the Otsu method, divides the image into a biofilm region and a pipe matrix region, and calculates the ratio of the number of pixels in the biofilm region to the total number of pixels to obtain the biofilm coverage density.

[0028] In step 6, statistical difference analysis is used to compare the numerical differences in the positive rates of urine bacterial culture and urotrophic bacteria culture between the intervention group and the control group, evaluating the inhibitory efficacy of fosfomycin tromethamine on urotrophic bacteria formation. The specific evaluation logic includes calculating the decrease in the positive rate between the intervention group and the control group. For example, the decrease is equal to the difference between the positive rate of the control group and the positive rate of the intervention group, divided by the positive rate of the control group. Data analysis software is used to test the balance between the two groups to ensure comparability in terms of confounding factors such as age and disease distribution. The significance level variable P-value is calculated using the chi-square test or Fisher's exact test. When the significance level variable is less than the preset probability threshold of 0.05, the difference between the two groups is considered statistically significant, thus determining that fosfomycin tromethamine has significant efficacy in preventing urinary tract infections and inhibiting urotrophic bacteria colonization. Simultaneously, the clinical benefits of the drug intervention are quantitatively described by calculating the relative risk (RR) and specific risk (AR).

[0029] In step 7, basic and clinical data of patients are integrated, and characteristic differences between the positive and negative groups of capsular infections are extracted. Logistic regression analysis is then used to identify independent risk factors leading to capsular infection. The identification process includes: first, univariate analysis is performed, using t-tests or rank-sum tests to screen for candidate variables showing significant differences between the positive and negative groups. These variables include double-J catheter indwelling time, glycated hemoglobin levels, and urine pH fluctuation range. During extraction, a smoothing algorithm is used to exclude irregular fluctuations caused by sampling errors or laboratory random errors, improving the signal-to-noise ratio of the input data. Subsequently, the candidate variables are substituted into the multivariate logistic regression equation, and confounding factors are eliminated using stepwise regression methods such as backward elimination. Variables with regression coefficients greater than 0 and statistical significance levels less than 0.05 are identified as independent risk factors. In this embodiment, the independent risk factors identified include: double-J stent placement time exceeding 28 days, uncontrolled diabetes with glycated hemoglobin greater than 7%, urine pH consistently greater than 7.0, and failure to receive effective prophylactic intervention with fosfomycin postoperatively.

[0030] In step 8, a weighted risk prediction model is constructed based on the identified independent risk factors. This model calculates the probability score of the patient developing capsular infection and delineates different risk level intervals based on the probability score, generating a preventative intervention recommendation report. The model construction logic includes: assigning a weight coefficient proportional to the regression coefficient of each independent risk factor. For example, if the regression coefficient of the double-J tube indwelling time is 2.5, then a corresponding weight is assigned. The quantitative values ​​of various patient indicators, such as indwelling days and blood glucose levels, are multiplied by the corresponding weight coefficients and summed to obtain the original risk score. The original risk score is then mapped to a probability interval between 0 and 1 using the logistic function to obtain the probability of capsular infection risk. Specifically, the risk probability is equal to 1 divided by the sum of 1 and the negative power of the original risk score (the natural constant). Different baseline correction coefficients are used for patients of different genders during the calculation process. The risk levels are divided into the following ranges: a risk probability less than 0.3 is defined as low risk, and routine clinical observation is recommended; a risk probability between 0.3 and 0.7 is defined as medium risk, and it is recommended to increase the frequency of urine testing, performing routine urine tests every 3 days, and considering prophylactic administration of fosfomycin every 5 days; a risk probability greater than 0.7 is defined as high risk, and it is recommended to immediately implement an intensive intervention program centered on fosfomycin tromethamine, and to remove the double-J stent early. The prophylactic intervention recommendation report includes personalized extubation timing recommendations for specific patients, optimized dosing cycles for fosfomycin tromethamine, and postoperative lifestyle guidance such as increasing fluid intake and avoiding prolonged sitting.

[0031] The method also involves monitoring fosfomycin-tromethorphan resistance by determining the minimum inhibitory concentration (MIC) of fosfomycin against the extracted captive bacteria through drug susceptibility testing. The trend of resistance evolution is integrated into the risk prediction model, and the coefficients of the prevention intervention variable in the model are adjusted to modify the expected long-term intervention effect. Furthermore, a feedback database containing a large number of clinical records is established, and actual infection case data are periodically fed back into the risk prediction model, using an iterative learning algorithm to dynamically adjust the weight coefficients. Specifically, whenever 100 new clinical cases are accumulated, the model automatically triggers a retraining logic, updating the weights based on the gradient bias of the new samples to improve the model's prediction accuracy under different population distributions.

[0032] The method for predicting the risk of double-J tube capsular infection based on microbial testing also includes quantitative monitoring of the effective concentration of fosfomycin in the patient's urine. Using high-performance liquid chromatography (HPLC), with the mobile phase ratio and detection wavelength set, the urinary drug concentration in the intervention group was measured at 12, 24, 48, and 72 hours after administration. By constructing a curve showing the change in urinary drug concentration over time, the positive correlation between the peak urinary drug concentration, the area under the curve, and the capsular inhibition rate was analyzed. When the monitored urinary drug concentration is lower than a preset minimum effective inhibition concentration, such as 128 μg / mL, the system automatically increases the patient's infection risk score in the risk prediction model.

[0033] The risk prediction model incorporates a time decay factor in its calculation logic. Because double-J stents gradually undergo surface crusting, material degradation, and decreased biocompatibility in the urinary environment, the difficulty of bacterial colonization increases non-linearly over time. The time decay factor is set as an exponentially growing term with a base greater than 1, where the exponent is the number of days of placement. The original risk score is non-linearly corrected by multiplying it by this time decay factor as the number of placement days increases. Furthermore, the model introduces a stent material type variable, assigning different initial risk values ​​to double-J stents with different coefficients of friction and hydrophilicity, such as silicone and polyurethane, further refining the risk scoring calculation logic.

[0034] The preventative intervention recommendation report generated by the method is automatically synchronized to the Hospital Information System (HIS) via electronic interfaces such as the HL7 protocol or the FHIR standard. When clinicians access a patient's electronic medical record, the system automatically pops up a high-risk warning window, displaying the current infection probability and corresponding evidence-based medical recommendations. For high-risk patients, it is recommended that physicians adjust the perfusion pressure during surgical procedures, maintaining it below 20 mmHg, and, based on model recommendations, increase the postoperative fluid resuscitation rate to 150 ml per hour to reduce the probability of bacterial colonization on the tube surface through continuous physical flushing.

[0035] In practical applications, the method of this embodiment can be further refined into the following operation process: Sub-step A: Data Collection Phase. A medical record was created for a newly admitted male patient undergoing ureteral stone surgery. His basic information was collected: age 45, BMI 26, no history of diabetes, and one previous urinary tract infection. The surgical record showed the use of a polyurethane double-J stent, planned for 35 days. The system automatically marked him as a data collection subject.

[0036] Sub-step B: Experimental grouping and protocol execution. The patient was assigned to the intervention group. On postoperative day 1, the patient was instructed to take 3 grams of fosfomycin tromethamine. Subsequently, the patient took the medication once every 7 days during the follow-up period. Monitoring records showed that the patient's urine pH remained between 6.2 and 6.8 during the medication period, and the white blood cell count fluctuated within the normal range.

[0037] Sub-step C: Sample processing and microbiological analysis. The double-J stent was removed on postoperative day 35. Samples were collected according to standard procedures. Urine culture showed a colony count of 200 CFU / mL, which was considered negative. Ultrasonic elution was performed on the three segments of the double-J stent. Culture results for the eluent near the kidney showed 50 CFU / mL, the mid-segment 30 CFU / mL, and the near-bladder segment 80 CFU / mL. Mass spectrometry identification revealed a small amount of Staphylococcus epidermidis colonization, but no mature biofilm had formed.

[0038] Sub-step D: Risk model assessment. Enter the above data into the predictive model. Since the patient is taking medication as scheduled and has a good urinary tract condition, although the indwelling time exceeds 28 days, the model calculates a risk probability score of 0.25. The system output report indicates low risk, recommending continued observation for 72 hours after catheter removal, without the need for additional intensive medication.

[0039] Sub-step E: Model Iteration. The patient's clinical data, medication records, and final microbiological test results are fed back to the feedback database. The model identified that under fosfomycin intervention, even with long-term indwelling, the colonization of the membrane remained at an extremely low level, thus fine-tuning the interaction weight coefficients between indwelling time and preventive intervention in subsequent training.

[0040] Furthermore, the basic information obtained in step 1 includes the patient's age, gender, body mass index, presence of underlying metabolic diseases, and history of urinary tract infection; The clinical diagnostic and treatment parameters include the type of surgery, duration of double-J stent placement, and type and dosage of postoperative antibiotics. The physiological indicators include urine pH, urine specific gravity, and white blood cell count. The specific fosfomycin tromethamine dosing regimen implemented in step 2 for the intervention group is as follows: At the predetermined time point after double-J stent implantation, patients are instructed to take a preset dose of fosfomycin tromethorphan powder orally, and to take the medication periodically at a preset frequency during the subsequent indwelling period, until the medication is stopped 24 hours before the double-J stent is removed. Fosfomycin tromethamine blocks the production of muramic acid by inhibiting the activity of pyruvyltransferase in bacterial cell wall synthesis. The concentration of fosfomycin tromethamine in urine is maintained above the effective concentration required to inhibit bacterial adhesion after administration. It penetrates the initially formed biofilm matrix through osmosis and disrupts the metabolic environment of the biofilm-covered bacteria.

[0041] Furthermore, the process of collecting the double-J tube body sample in step 3 includes: Under aseptic conditions, the removed double-J tube is placed in a sterile sampling container containing a preset volume of physiological saline. The tube is then cut into multiple segments of equal length by mechanical cutting. These segments include a proximal segment near the kidney, a mid-segment segment, and a proximal segment near the bladder, in order to assess biofilm load at multiple sites. Step 4, the process of culturing bacteria in the urine sample, includes: The collected midstream urine was inoculated onto pre-defined blood agar and MacConkey agar, and placed in a pre-defined constant temperature incubator for 48 hours of static culture. The number of colony-forming units per unit volume was recorded. When the number of colonies exceeded the pre-defined concentration threshold, the urine bacterial culture was considered positive. The identification of urine bacteria in step 4 also includes using a fully automated microbial mass spectrometry detection system to compare with a peptide spectrum library to identify specific pathogenic bacterial species, including Escherichia coli, Enterococcus faecalis, Klebsiella pneumoniae, and Staphylococcus epidermidis, and recording the correlation strength between different bacterial species and the formation of double-J tube membranes.

[0042] Furthermore, the process of extracting the film-coated bacteria in step 5 includes: The cut double-J tube segment was placed in a centrifuge tube containing eluent and subjected to ultrasonic cleaning at a preset frequency to physically break down the biofilm structure attached to the inside and outside of the tube wall and release colonizing bacteria into the eluent. The bacteria are evenly distributed by vortexing. A predetermined volume of elution solution is serially diluted, inoculated onto a solid culture medium, and incubated at a constant temperature. The original bacterial load of the film is calculated based on the colony count. The determination of biofilm coverage density in step 5 also includes observing the microstructure of the surface of the double-J tube using a scanning electron microscope and identifying the area ratio of the biofilm-covered region through image processing algorithms. The image processing algorithm includes grayscale processing of the scanning electron microscope image, calculating the optimal segmentation threshold using the Otsu method, dividing the image into a biofilm region and a pipe matrix region, calculating the ratio of the number of pixels in the biofilm region to the total number of pixels to obtain the biofilm coverage density, and using the biofilm coverage density as an auxiliary indicator for evaluating the severity of infection.

[0043] Furthermore, referring to Figure 2 The process of evaluating inhibitory efficacy in step 6 includes: The decrease in the positive rate of the intervention group and the control group was calculated, and the significance of the decrease was determined by a significance test. The decrease was equal to the difference between the positive rate of the control group and the positive rate of the intervention group divided by the positive rate of the control group. When the significance level variable is less than the preset probability threshold, fosfomycin tromethamine is determined to have significant efficacy in preventing urinary tract infections and inhibiting the colonization of tunica albuginea. Data analysis software was used to examine the balance between the two groups of patients to ensure that the intervention group and the control group were comparable in terms of age, disease distribution, and confounding factors. The process of identifying independent risk factors in step 7 includes: Univariate analysis was performed to screen out candidate variables that showed significant differences between the positive and negative groups of capsular infection; Candidate variables are substituted into the multivariate logistic regression equation, and confounding factors are eliminated through stepwise regression. Variables with regression coefficients greater than the preset threshold and statistical significance are identified as independent risk factors. The independent risk factors include double-J stent placement for more than a preset number of days, uncontrolled diabetes, urine pH consistently outside the preset range, and lack of postoperative prophylactic medication.

[0044] Furthermore, referring to Figure 3 The logic for constructing the weighted risk prediction model in step 8 includes: Each independent risk factor is assigned a weight coefficient proportional to its regression coefficient. The quantitative values ​​of each patient's indicators are multiplied by the corresponding weight coefficients and summed to obtain the original risk score. The original risk score is mapped to a probability range between zero and one by using a logical function to obtain the probability of infection with capsular bacteria. During the calculation process, different baseline correction coefficients were used for patients of different genders to eliminate the systematic bias caused by physiological differences in infection risk prediction.

[0045] The risk level range is divided into low-risk, medium-risk and high-risk areas; When the risk probability is less than the first preset threshold, it is defined as low risk and routine clinical observation is carried out. When the risk probability is between the first preset threshold and the second preset threshold, it is defined as medium risk, and the frequency of urine testing is increased and prophylactic medication is administered. When the risk probability is greater than the second preset threshold, it is defined as high risk, and an enhanced intervention program with fosfomycin tromethamine as the core is implemented.

[0046] The second embodiment, based on the first embodiment, further refines the processing logic of the risk prediction model for scenarios of co-infection by multiple pathogens, and introduces a dynamic quantitative assessment of biofilm thickness.

[0047] In step 5, the determination of biofilm coverage density also includes fluorescent staining of live and dead bacteria using a laser confocal scanning microscope (CLSM). Specifically, live bacteria are labeled with green fluorescent dye, and damaged or dead bacteria are labeled with red fluorescent dye. A three-dimensional structural model of the biofilm is constructed by acquiring optical section images at different depths. The average thickness of the biofilm and the proportion of live bacteria are calculated using image analysis software. The average thickness, as a key auxiliary indicator for evaluating the severity of infection, is incorporated into the weighted risk prediction model in step 8.

[0048] In step 8, the construction logic of the weighted risk prediction model further includes modifications for the pathogenicity of different bacterial species. Since *Escherichia coli* and *Klebsiella pneumoniae* have stronger biofilm formation capabilities and higher viral virulence, when the mass spectrometry identification results in step 4 show the presence of these high-risk bacterial species, the model will add an additional bacterial species weight gain term to the original risk score. The value of this gain term is preset based on the correlation strength between different bacterial species and double-J tube membrane formation. For example, the gain term for *Escherichia coli* is set to 1.5, while the gain term for *Staphylococcus epidermidis* is set to 1.1.

[0049] Furthermore, the method in this embodiment also includes an imaging assessment of the crust formation on the surface of the double-J tube in the patient. High-density shadows on the surface of the double-J tube wall are identified using high-resolution CT images, and the volume of the crust layer is measured using a computer-aided diagnostic system. The presence of the crust layer provides a more stable attachment matrix for capsular bacteria. Therefore, the crust volume is introduced as an independent continuous variable into the logistic regression equation. Analysis revealed that for every 1 cubic millimeter increase in crust volume, the risk of capsular bacterial infection increases by 1.2 times (OR).

[0050] In assessing the efficacy of fosfomycin intervention, this embodiment introduces a parallel control experiment of in vitro biofilm inhibition. Pathogenic bacteria isolated from patient urine were used to establish an in vitro biofilm model under laboratory conditions. The inhibition curves of biofilm formation were observed by adding different concentrations of fosfomycin. The half-maximal inhibitory concentration (IC50) was calculated. The experimentally measured IC50 value was compared with the actual drug concentration in the patient's urine. If the actual concentration was significantly higher than the IC50, the negative correlation weight of the preventive intervention variable was increased in the risk prediction model.

[0051] The risk prediction model in this embodiment also considers fluctuations in postoperative fluid intake. Daily urine output was collected using a smart urine monitor. The study found that patients with an average daily urine output of less than 1500 ml showed a significantly faster rate of solute deposition on the surface of the double-J catheter. The model introduced a variable of average daily urine flow, which is inversely proportional to the probability of infection risk. Through a multi-dimensional risk assessment matrix, deep integration of microbiological laboratory data, imaging crusting data, pharmacokinetic data, and lifestyle data achieved extremely high-precision early warning of potential infection risks in the early stages of double-J catheter placement.

[0052] Furthermore, the preventive intervention recommendation report includes personalized extubation timing recommendations for specific patients, recommendations for optimized dosing cycles of fosfomycin tromethamine, and postoperative lifestyle guidance. The method also includes establishing a feedback database containing clinical records, periodically transmitting actual infection case data back to the risk prediction model, and dynamically adjusting the weight coefficients using an iterative learning algorithm. Whenever a preset number of new clinical cases are accumulated, the model automatically triggers retraining logic to update the weights based on the gradient bias of the new samples, so as to improve the prediction accuracy of the model under different population distributions. The reports generated by the risk prediction model are automatically synchronized to the hospital information system via an electronic interface, providing real-time decision support for clinicians. For high-risk patients, intervention recommendations also include adjusting the perfusion pressure during the surgical procedure to keep it below a preset pressure threshold, and increasing the postoperative fluid resuscitation rate as suggested by the model to reduce the probability of bacterial colonization through physical flushing.

[0053] Furthermore, the method also involves monitoring the resistance to fosfomycin tromethamine, determining the minimum inhibitory concentration of fosfomycin on the extracted cloacal bacteria through drug susceptibility testing, and integrating the resistance evolution trend into a risk prediction model to correct the expected long-term intervention effect. Step 3 requires that the number of settled bacteria and airborne bacteria in the environment be lower than the preset cleanliness standard throughout the entire process of collection, transportation and laboratory processing, to ensure that all detected bacteria originate from the patient sample itself. The process of extracting risk factors in step 7 eliminates irregular fluctuations in data caused by sampling errors or random laboratory errors, and improves the signal-to-noise ratio of the input data through a smoothing algorithm. The method also studies the relationship between the double-J tube material and the adhesion of the biofilm, introduces the scaffold material type variable into the risk prediction model, assigns different initial basic risk values ​​to double-J tubes of different materials, and refines the calculation logic of risk scoring.

[0054] Furthermore, the method also includes quantitative monitoring of the effective concentration of fosfomycin in the patient's urine, measuring the urine drug concentration curve at different time points after administration using high performance liquid chromatography, and analyzing the positive correlation between the peak urine drug concentration and the inhibition rate of the biofilm. When the monitored urinary drug concentration is lower than the preset minimum effective inhibitory concentration, the patient's infection risk score is increased in the risk prediction model; The calculation logic of the risk probability score introduces a time decay factor to consider the nonlinear increasing effect of material aging and decreased biocompatibility of the double-J tube on the risk of infection as the retention time increases. The time decay factor is set as a base greater than one and an exponential growth term of the number of days of detention. As the number of days of detention increases, the original risk score is multiplied by the time decay factor to achieve a non-linear correction of the risk probability. The method for predicting the risk of double-J tube membrane infection based on microbial testing integrates clinical indicators, drug intervention variables, and etiological evidence to construct a closed-loop infection prevention and control system covering the entire diagnosis and treatment cycle. The model is constructed using cross-validation to evaluate its performance. The area under the receiver operating characteristic curve (ROC) of the model is calculated. When the area is greater than a preset performance threshold, the model is deemed to have the reliability for clinical application. By establishing a multi-dimensional risk assessment matrix, laboratory data from microbiology testing is deeply integrated with clinical epidemiological data to achieve quantitative early warning of potential infection risks in the early stages of double-J tube placement. The preventive interventions include advising doctors to adjust the perfusion pressure during surgery and the postoperative fluid resuscitation rate based on the risk level, reducing the initial adhesion of bacteria to the surface of the double-J tube through continuous hydrodynamic flushing, and synergistically reducing the risk of biofilm formation by combining the inhibitory effect of fosfomycin on cell wall synthesis.

[0055] According to a second embodiment of the present invention, the present invention claims protection for a double-J tube membrane bacterial infection risk prediction system based on microbial testing, comprising: One or more processors; A memory having stored one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method for predicting the risk of double-J tube membrane infection based on microbial testing.

[0056] According to a third embodiment of the present invention, a method for predicting the risk of infection of double-J tube capsular bacteria based on microbial testing using a cloud computing architecture is provided, aiming to achieve multi-center data collaboration and continuous model evolution.

[0057] In step 1, the initial patient information database is deployed on a cloud server. Medical institutions at all levels upload anonymized patient clinical parameters in real time via a standardized application programming interface (API). The cloud system utilizes a distributed computing framework to perform parallel preprocessing of massive amounts of data, including automated unit conversion, outlier detection, and multi-source data fusion.

[0058] In step 7, the logistic regression analysis employs a federated learning mechanism. Each medical institution trains a local prediction model on its local server using its own data, uploading only the updated gradient parameters to the cloud center. The cloud center aggregates the gradients from all institutions using a weighted average algorithm, updates the global risk prediction model, and then distributes it back to each institution. This approach, while protecting patient privacy, fully utilizes the sample diversity of multiple centers and solves the model overfitting problem caused by the limited sample size of a single center.

[0059] In step 8, the generated preventative intervention recommendation report has dynamic adjustment characteristics. The model not only provides a single risk score but also generates a risk trend curve based on real-time fluctuations in the patient's physiological indicators. For example, if the patient's urine pH level shows an upward trend for three consecutive days, the model will promptly update the report, indicating a change in risk level from low to medium, and automatically send intervention adjustment recommendations to the attending physician's mobile terminal.

[0060] For the dosing regimen of fosfomycin tromethamine, the method in this embodiment also includes a machine learning-based dose optimization module. This module uses a reinforcement learning algorithm to find the optimal balance between dosing frequency and antibacterial effect based on the patient's weight, renal function indicators such as creatinine clearance, and previous drug sensitivity results. The generated personalized dosing recommendations are pushed to the patient via a health management mini-program, reminding the patient to take the medication on time and record their post-dosing reactions.

[0061] The method also integrates a simulation algorithm for the aging and degradation of double-J tube materials. Based on the known degradation rate curves of materials immersed in urine, combined with the biochemical components of the patient's urine, such as calcium oxalate saturation and uric acid concentration, the method predicts the change in the micro-roughness of the double-J tube surface over time. This simulation prediction result based on physicochemical principles is injected as prior knowledge into the weighted risk prediction model, enabling the model to provide a preliminary risk assessment with reference value based on the retention time even in the absence of real-time microbiological test data.

[0062] This embodiment also includes closed-loop feedback logic for postoperative complications. If a patient develops significant urinary tract infection symptoms after catheter removal, the relevant clinical diagnostic results, blood culture, and urine culture data will be automatically annotated into the previous prediction cases as the gold standard. The model uses a backpropagation algorithm to adjust the weights of risk factors with large prediction biases. Through this continuous closed-loop learning, the risk prediction model can adapt to the ever-changing bacterial resistance spectrum and the development of clinical diagnosis and treatment technologies.

[0063] The proposed method for predicting the risk of double-J tube capsular infection based on microbial testing integrates clinical indicators, drug intervention variables, and etiological evidence to construct a closed-loop infection control system covering the entire diagnosis and treatment cycle. Cross-validation was used to evaluate the model's performance during construction, calculating the area under the receiver operating characteristic (ROC) curve (AUC). When this AUC value exceeds a preset performance threshold of 0.85, the model is deemed to possess the reliability for clinical application.

[0064] Through retrospective validation using a large amount of clinical data, the method of this invention can improve the predictive accuracy of capsular bacterial infection to over 90%. Compared to traditional methods that rely solely on urine culture, this invention, through direct extraction and quantitative analysis of capsular bacteria, can identify occult infections—those with negative urine cultures but with a large number of colonizing bacteria on the tubing surface—at an earlier stage. This has extremely important clinical value for guiding the precise prophylactic use of fosfomycin, optimizing the timing of double-J stent removal, and reducing the incidence of postoperative urosepsis.

[0065] In the specific implementation process, fosfomycin tromethamine, as the core intervention, demonstrated its unique ability to penetrate biofilms in a fully quantifiable manner within the model. By comparing the changes in biofilm structure before and after the intervention, it was found that fosfomycin significantly reduced the content of extracellular polysaccharide polymers (EPS) in the biofilm matrix, making the originally dense biofilm structure looser, thereby increasing the sensitivity of bacteria to conventional antibiotics and the body's immune system. This quantitative description of pharmacological effects provides a scientific basis for the model to predict complex drug-resistant bacterial infections.

[0066] Finally, the method of this invention deeply integrates laboratory data from microbiological testing with clinical epidemiological data by establishing a multi-dimensional risk assessment matrix, enabling quantitative early warning of potential infection risks in the early stages of double-J tube placement. This system not only focuses on the results of a single test but also emphasizes the evolution of data over time, providing clinicians with a dynamic, accurate, and operable decision support tool.

[0067] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0068] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0069] The specific embodiments of the invention have been described in detail above, but they are only examples, and this application is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications or substitutions to the invention are also within the scope of this application. Therefore, all equivalent changes, modifications, and improvements made without departing from the spirit and principles of this application should be covered within the scope of this application.

Claims

1. A method for predicting the risk of bacterial infection on double-J tube membranes based on microbial testing, characterized in that, Includes the following steps: Step 1: Obtain basic information, clinical diagnostic and treatment parameters, and physiological index data during the follow-up period of patients undergoing double-J stent implantation to establish an initial patient information database; Step 2: Patients were randomly divided into an intervention group and a control group. The intervention group received fosfomycin tromethamine, while the control group did not receive any drug intervention. The urinary tract status of both groups was continuously monitored during the stent placement period. Step 3: When the predetermined double-J stent removal cycle arrives, collect the patient's midstream urine sample and the removed double-J stent body sample, and transport and preprocess them. Step 4: Perform bacterial culture and quantitative analysis on urine samples using microbial testing techniques to determine the positive rate of urine bacterial culture and colony distribution characteristics; Step 5: Ultrasonic elution combined with microfiltration is used to extract the biofilm on the surface of the double-J tube, and quantitative culture and identification of the biofilm are performed to determine the positive rate of biofilm culture and the biofilm coverage density. Step 6: Using statistical difference analysis, compare the numerical differences between the intervention group and the control group in terms of positive rates of urine bacterial culture and positive rates of biofilm culture, and evaluate the inhibitory efficacy of fosfomycin tromethamine on biofilm formation. Step 7: Integrate the patients' basic and clinical data, extract the characteristic differences between the positive and negative groups of capsular infections, and identify independent risk factors leading to capsular infections through logistic regression analysis; Step 8: Construct a weighted risk prediction model based on the identified independent risk factors, calculate the probability score of patients developing capsular infections, and define different risk level intervals according to the probability score to generate a preventive intervention recommendation report.

2. The method for predicting the risk of double-J tube membrane infection based on microbial testing according to claim 1, characterized in that, The basic information obtained in step 1 includes the patient's age, gender, body mass index, presence of underlying metabolic diseases, and history of urinary tract infection; The clinical diagnostic and treatment parameters include the type of surgery, duration of double-J stent placement, and type and dosage of postoperative antibiotics. The physiological indicators include urine pH, urine specific gravity, and white blood cell count. The specific fosfomycin tromethamine dosing regimen implemented in step 2 for the intervention group is as follows: At the predetermined time point after double-J stent implantation, patients are instructed to take a preset dose of fosfomycin tromethorphan powder orally, and to take the medication periodically at a preset frequency during the subsequent indwelling period, until the medication is stopped 24 hours before the double-J stent is removed. Fosfomycin tromethamine blocks the production of muramic acid by inhibiting the activity of pyruvyltransferase in bacterial cell wall synthesis. The concentration of fosfomycin tromethamine in urine is maintained above the effective concentration required to inhibit bacterial adhesion after administration. It penetrates the initially formed biofilm matrix through osmosis and disrupts the metabolic environment of the biofilm-covered bacteria.

3. The method for predicting the risk of double-J tube membrane infection based on microbial testing according to claim 1, characterized in that, The process of collecting double-J tube samples in step 3 includes: Under aseptic conditions, the removed double-J tube is placed in a sterile sampling container containing a preset volume of physiological saline. The tube is then cut into multiple segments of equal length by mechanical cutting. These segments include a proximal segment near the kidney, a mid-segment segment, and a proximal segment near the bladder, in order to assess biofilm load at multiple sites. Step 4, the process of culturing bacteria in the urine sample, includes: The collected midstream urine was inoculated onto pre-defined blood agar and MacConkey agar, and placed in a pre-defined constant temperature incubator for 48 hours of static culture. The number of colony-forming units per unit volume was recorded. When the number of colonies exceeded the pre-defined concentration threshold, the urine bacterial culture was considered positive. The identification of urine bacteria in step 4 also includes using a fully automated microbial mass spectrometry detection system to compare with a peptide spectrum library to identify specific pathogenic bacterial species, including Escherichia coli, Enterococcus faecalis, Klebsiella pneumoniae, and Staphylococcus epidermidis, and recording the correlation strength between different bacterial species and the formation of double-J tube membranes.

4. The method for predicting the risk of bacterial infection of double-J tube membranes based on microbial testing according to claim 1, characterized in that, The process of extracting the biofilm-coated bacteria in step 5 includes: The cut double-J tube segment was placed in a centrifuge tube containing eluent and subjected to ultrasonic cleaning at a preset frequency to physically break down the biofilm structure attached to the inside and outside of the tube wall and release colonizing bacteria into the eluent. The bacteria are evenly distributed by vortexing. A predetermined volume of elution solution is serially diluted, inoculated onto a solid culture medium, and incubated at a constant temperature. The original bacterial load of the film is calculated based on the colony count. The determination of biofilm coverage density in step 5 also includes observing the microstructure of the surface of the double-J tube using a scanning electron microscope and identifying the area ratio of the biofilm-covered region through image processing algorithms. The image processing algorithm includes grayscale processing of the scanning electron microscope image, calculating the optimal segmentation threshold using the Otsu method, dividing the image into a biofilm region and a pipe matrix region, calculating the ratio of the number of pixels in the biofilm region to the total number of pixels to obtain the biofilm coverage density, and using the biofilm coverage density as an auxiliary indicator for evaluating the severity of infection.

5. The method for predicting the risk of bacterial infection on double-J tube membranes based on microbial testing according to claim 1, characterized in that, The process of evaluating inhibitory efficacy in step 6 includes: The decrease in the positive rate in the intervention group and the control group was calculated, and the significance of the decrease was determined by a significance test. The decrease is equal to the difference between the positive rate of the control group and the positive rate of the intervention group, divided by the positive rate of the control group; When the significance level variable is less than the preset probability threshold, fosfomycin tromethamine is determined to have significant efficacy in preventing urinary tract infections and inhibiting the colonization of tunica albuginea. Data analysis software was used to examine the balance between the two groups of patients to ensure that the intervention group and the control group were comparable in terms of age, disease distribution, and confounding factors. The process of identifying independent risk factors in step 7 includes: Univariate analysis was performed to screen out candidate variables that showed significant differences between the positive and negative groups of capsular infection; Candidate variables are substituted into the multivariate logistic regression equation, and confounding factors are eliminated through stepwise regression. Variables with regression coefficients greater than the preset threshold and statistical significance are identified as independent risk factors. The independent risk factors include double-J stent placement for more than a preset number of days, uncontrolled diabetes, urine pH consistently outside the preset range, and lack of postoperative prophylactic medication.

6. The method for predicting the risk of bacterial infection of double-J tube membranes based on microbial testing according to claim 1, characterized in that, The logic for constructing the weighted risk prediction model in step 8 includes: Each independent risk factor is assigned a weight coefficient proportional to its regression coefficient. The quantitative values ​​of each patient's indicators are multiplied by the corresponding weight coefficients and summed to obtain the original risk score. The original risk score is mapped to a probability range between zero and one by using a logical function to obtain the probability of infection with capsular bacteria. During the calculation process, different baseline correction coefficients were used for patients of different genders to eliminate the systematic bias caused by physiological differences in infection risk prediction. The risk level range is divided into low-risk, medium-risk and high-risk areas; When the risk probability is less than the first preset threshold, it is defined as low risk and routine clinical observation is carried out. When the risk probability is between the first preset threshold and the second preset threshold, it is defined as medium risk, and the frequency of urine testing is increased and prophylactic medication is administered. When the risk probability is greater than the second preset threshold, it is defined as high risk, and an enhanced intervention program with fosfomycin tromethamine as the core is implemented.

7. The method for predicting the risk of bacterial infection of double-J tube membranes based on microbial testing according to claim 1, characterized in that, The preventive intervention recommendation report includes personalized extubation timing recommendations for specific patients, recommendations for optimized dosing cycles of fosfomycin tromethamine, and postoperative lifestyle guidance. The method also includes establishing a feedback database containing clinical records, periodically transmitting actual infection case data back to the risk prediction model, and dynamically adjusting the weight coefficients using an iterative learning algorithm. Whenever a preset number of new clinical cases are accumulated, the model automatically triggers retraining logic to update the weights based on the gradient bias of the new samples, so as to improve the prediction accuracy of the model under different population distributions. The reports generated by the risk prediction model are automatically synchronized to the hospital information system via an electronic interface, providing real-time decision support for clinicians. For high-risk patients, intervention recommendations also include adjusting the perfusion pressure during the surgical procedure to keep it below a preset pressure threshold, and increasing the postoperative fluid resuscitation rate as suggested by the model to reduce the probability of bacterial colonization through physical flushing.

8. The method for predicting the risk of bacterial infection of double-J tube membranes based on microbial testing according to claim 1, characterized in that, The method also involves monitoring resistance to fosfomycin tromethamine, determining the minimum inhibitory concentration of fosfomycin on the extracted cloacal bacteria through drug susceptibility testing, and integrating the resistance evolution trend into a risk prediction model to correct the expected long-term intervention effect. Step 3 requires that the number of settled bacteria and airborne bacteria in the environment be lower than the preset cleanliness standard throughout the entire process of collection, transportation and laboratory processing, to ensure that all detected bacteria originate from the patient sample itself. The process of extracting risk factors in step 7 eliminates irregular fluctuations in data caused by sampling errors or random laboratory errors, and improves the signal-to-noise ratio of the input data through a smoothing algorithm. The method also studies the relationship between the double-J tube material and the adhesion of the biofilm, introduces the scaffold material type variable into the risk prediction model, assigns different initial basic risk values ​​to double-J tubes of different materials, and refines the calculation logic of risk scoring.

9. The method for predicting the risk of bacterial infection of double-J tube membranes based on microbial testing according to claim 1, characterized in that, The method also includes quantitative monitoring of the effective concentration of fosfomycin in the patient's urine, measuring the urine drug concentration curve at different time points after administration using high performance liquid chromatography, and analyzing the positive correlation between the peak urine drug concentration and the inhibition rate of the biofilm. When the monitored urinary drug concentration is lower than the preset minimum effective inhibitory concentration, the patient's infection risk score is increased in the risk prediction model; The calculation logic of the risk probability score introduces a time decay factor to consider the nonlinear increasing effect of material aging and decreased biocompatibility of the double-J tube on the risk of infection as the retention time increases. The time decay factor is set as a base greater than one and an exponential growth term of the number of days of detention. As the number of days of detention increases, the original risk score is multiplied by the time decay factor to achieve a non-linear correction of the risk probability. The method for predicting the risk of double-J tube membrane infection based on microbial testing integrates clinical indicators, drug intervention variables, and etiological evidence to construct a closed-loop infection prevention and control system covering the entire diagnosis and treatment cycle. The model is constructed using cross-validation to evaluate its performance. The area under the receiver operating characteristic curve (ROC) of the model is calculated. When the area is greater than a preset performance threshold, the model is deemed to have the reliability for clinical application. By establishing a multi-dimensional risk assessment matrix, laboratory data from microbiology testing is deeply integrated with clinical epidemiological data to achieve quantitative early warning of potential infection risks in the early stages of double-J tube placement. The preventive interventions include advising doctors to adjust the perfusion pressure during surgery and the postoperative fluid resuscitation rate based on the risk level, reducing the initial adhesion of bacteria to the surface of the double-J tube through continuous hydrodynamic flushing, and synergistically reducing the risk of biofilm formation by combining the inhibitory effect of fosfomycin on cell wall synthesis.

10. A double-J tube membrane bacterial infection risk prediction system based on microbial testing, characterized in that, include: One or more processors; A memory having stored one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method for predicting the risk of double-J tube membrane infection based on microbial testing according to any one of claims 1 to 9.

Images