System for predicting urinary tract infection risk after unilateral pyeloplasty of children

Abstract

The invention provides a system for predicting urinary tract infection risk after a child unilateral pyeloplasty operation, and relates to the field of bioinformatics. The system determines significant risk prediction factors by statistically analyzing variable data in case data related to urinary tract infection (UTI) after pyeloplasty by applying Lasso regression and logistic regression modeling, designs a prediction system according to the prediction factors, and determines the risk of the UTI through a series of data processing processes such as sample detection, data calculation, data judgment and data output. The prediction of the urinary tract infection risk after the unilateral pyeloplasty of children is realized, and the treatment process is accelerated.

Description

Technical Field

[0001] The present invention relates to the field of bioinformatics, and in particular to a system for predicting the risk of urinary tract infection after unilateral pyeloplasty in children. Background Art

[0002] Ureteropelvic junction obstruction (UPJO) is defined as a blockage at the junction of the kidney and ureter, resulting in reduced urine flow from the pelvis to the ureter. UPJO is one of the leading causes of hydronephrosis in infants. If left untreated, it may lead to hydronephrosis, chronic infection or urolithiasis, and often progressive renal insufficiency.

[0003] There are multiple treatment options for UPJO, and surgery remains the gold standard, especially when urine flow has not improved after 18 months. However, despite the high success rate of UPJO, surgical treatment is not without complications. The main complications after surgical treatment of UPJO are urinary tract infection (UTI), urinary extravasation and leakage, pyelonephritis, bleeding, and recurrent UPJO. Among them, the incidence of UTI remains high and challenging for pediatric urologists. UTI, mainly caused by Escherichia coli, is one of the most common types of infection in children and is more common in boys (3.7%) than girls (2%) in the first year of life and more common in girls after infancy. Previous studies have described risk factors for UTI in children, including male sex, weight, urea nitrogen level, recurrence of UTI within 3 months, long-term catheter retention, double-J catheter retention, and bilateral double-J catheter retention. However, studies on risk factors associated with UTI after unilateral pyeloplasty are still limited, especially those with large sample sizes. Summary of the invention

[0004] In view of the shortcomings and deficiencies of the above-mentioned prior art, the present invention aims to introduce a large-scale risk factor prediction model for UTI after unilateral pyeloplasty in children, in order to further reduce the incidence of postoperative UTI complications and provide more references for the treatment of UTI after pyeloplasty in children.

[0005] A first aspect of the present invention provides a system for predicting the risk of urinary tract infection in children after unilateral pyeloplasty, characterized in that the system comprises:

[0006] A detection module, at least used to detect the content or quantity data of a marker, wherein the marker includes variable data in the case data;

[0007] The calculation module is at least used to calculate the content or quantity data obtained by the detection module according to the judgment formula to obtain a score; the judgment module is at least used to compare the score with the threshold to judge the risk of urinary tract infection after unilateral pyeloplasty in children. If the score is higher than the threshold, it is judged that the risk of urinary tract infection after unilateral pyeloplasty in children is high, and if the score is lower than the threshold, it is judged that the risk of urinary tract infection after unilateral pyeloplasty in children is low;

[0008] The output module is at least used to output the result of the judgment module.

[0009] In a specific embodiment of the present invention, the variable data in the case data include: gender, operation time, postoperative drainage method, history of urinary tract infection before operation, history of nephrostomy before operation, age, serum urea nitrogen value, and renal cortical thickness value. Among them, postoperative drainage method: DJ tube drainage or pyelostomy tube external drainage; history of urinary tract infection before operation: that is, clinical diagnosis of urinary tract infection before operation; history of nephrostomy before operation: that is, renal puncture and ostomy were first performed due to abdominal pain, severe hydrops, etc. before operation, and radical surgery was performed in the second stage; age (in months); renal parenchymal thickness (measured by ultrasound before operation), in mm.

[0010] The judgment formula of the system of the present invention is as follows:

[0011] Score

[0012] Among them, logit (P) = 1.857* (postoperative drainage method) + 0.767* (sex) + 1.369* (preoperative urinary tract infection history) + 1.450* (preoperative nephrostomy history) + 0.010* (operative time) 0.017* (age) - 0.263* (renal cortical thickness) - 0.159* (serum urea nitrogen value) - 2.8748. The threshold is 0.5.

[0013] The gender, age, postoperative drainage method, history of urinary tract infection before surgery, and history of nephrostomy before surgery are all calculated in an assignment manner. In an embodiment of the present invention, the assignment rules are as follows: Gender: male = 1, female = 0; Operation time: continuous variable, unit is minute (min); Postoperative drainage method: DJ tube = 1, nephrostomy drainage = 0; History of urinary tract infection before surgery: yes = 1, no = 0; History of nephrostomy before surgery: yes = 1, no = 0; Age is a continuous variable, unit is month (month); Serum urea nitrogen value is a continuous variable, unit is mmol / L; Renal cortex thickness value is a continuous variable, unit is millimeter (mm). In a specific embodiment of the present invention, the system also includes a data acquisition module, which is used for sample extraction to obtain a sample to be tested. The sample is a biological fluid sample, preferably, the sample is selected from at least one of venous blood, peripheral blood, serum, plasma, and urine.

[0014] The second aspect of the present invention provides that the system is used to prepare one or more of an instrument, a device or a kit for predicting the risk of urinary tract infection after unilateral pyeloplasty in children.

[0015] The third aspect of the present invention provides an information data processing terminal, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the processor executes the computer program to calculate the system.

[0016] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and the program is executed by a processor to calculate the system.

[0017] The beneficial effects of the present invention include but are not limited to:

[0018] 1. The system established by the present invention based on the markers establishes a practical and feasible quantitative method for measuring the risk of urinary tract infection in children after unilateral pyeloplasty through a formula;

[0019] 2. The accuracy of the system for predicting the risk of urinary tract infection after unilateral pyeloplasty in children of the present invention is 78.1%, which can accurately predict the risk of urinary tract infection after unilateral pyeloplasty in children and accelerate the treatment process. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0021] Figure 1 is a schematic diagram of the result of the lasso regression process of filtering variables in an embodiment of the present invention, Figure 1 (A) The optimal parameter (lambda) was determined by ten-fold cross validation, where the dashed line represents the optimal point determined using the minimum criterion and one standard deviation boundary; Figure 1 (B) Lasso coefficient profile of latent features;

[0022] Figure 2 It is a schematic diagram of the nomogram of UTI after unilateral UPJO in an embodiment of the present invention;

[0023] Figure 3 A schematic diagram of a research nomogram ROC curve generated in an embodiment of the present invention;

[0024] Figure 4 A schematic diagram of a nomogram calibration curve in an embodiment of the present invention;

[0025] Figure 5 This is a schematic diagram of the results of the internal validation set in an embodiment of the present invention;

[0026] Figure 6 It is a schematic diagram of a DCA (Decision Curve Analysis) curve for studying a nomogram in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to more clearly explain the overall concept of the present invention, the following is described in detail in conjunction with the accompanying drawings of the specification by way of embodiment. In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features known in the art are not described.

[0028] If no specific conditions are specified in the examples, the experiments were carried out according to conventional conditions or conditions recommended by the manufacturer.

[0029] Unless otherwise specified, in the following embodiments, the reagents or instruments used without indicating the manufacturer are all conventional products that can be purchased commercially.

[0030] Unless otherwise stated, the experimental methods, detection methods, and preparation methods disclosed in the present invention all adopt conventional techniques in microbiology, biochemistry, analytical chemistry, cell culture, and related fields conventional in the technical field.

[0031] Postoperative urinary tract infection (UTI): The definition of postoperative UTI was based on the following two guidelines: (1) the 2015 European Association of Urology (EAU) guidelines for urinary tract infection and (2) the 2016 evidence-based guidelines for the diagnosis and treatment of urinary tract infection from the Nephrology Group of the Chinese Society of Pediatrics.

[0032] Diagnosis of a UTI involves several scenarios:

[0033] (1) Clinical symptoms: patients with urinary tract infection who have symptoms such as frequent urination, urgency, pain during urination, interrupted urination, turbid urine, hematuria, urinary retention, urinary incontinence, and lower abdominal pain;

[0034] (2) The number of white blood cells in the urine sample collected for routine testing is ≥5 / HP for two consecutive times.

[0035] In addition to the above criteria, the diagnosis of upper urinary tract infection also requires the following three characteristics: (1) systemic symptoms such as fever, anorexia, and drowsiness; nausea and vomiting; back pain and renal percussion pain. (2) The presence of white blood cell casts in three urine tests. (3) Gram-stained smears obtained by clean catheterization are positive. And one of the following: (1) Color Doppler ultrasound, abdominal spiral CT examination, or radionuclide renal static imaging (Dimercaptosuccin ic Acid, DMSA) indicates focal renal infectious lesions. (2) Urinary pathogens cultured with >103 colonies / mL using a catheter; (3) Urine pathogen colony counts >105 colonies / mL regardless of whether a catheter is used.

[0036] The present invention retrospectively analyzed the clinical data of children who underwent unilateral pyeloplasty at the Capital Institute of Pediatrics, Children's Hospital between January 2012 and January 2022. The present invention used variables such as gender, age (age in months), BMI, surgical method, type of drainage tube, and various blood and urine tests. Statistical analysis, including Lasso regression, logistic regression, and random forest modeling, was performed to identify significant predictive factors. A nomogram was constructed by selecting very important variables, and its performance was evaluated by decision curve analysis (DCA).

[0037] Results: Among 764 patients, 265 (35%) developed UTI after surgery. The main risk factors included surgical method, UPJO side of the patient, type of drainage tube, blood urea nitrogen (BUN) level, and patient height. Lasso regression identified 14 predictive variables, while logistic regression analysis identified independent risk and protective factors. Finally, eight variables, including gender, surgical time, postoperative drainage method, history of urinary tract infection before surgery, history of nephrostomy before surgery, age, serum urea nitrogen value, and renal cortical thickness, were selected to construct a nomogram for disease prediction.

[0038] The present invention shows that UTI after unilateral pyeloplasty is affected by multiple factors. Despite the use of prophylactic antibiotics, the incidence of UTI is significant. Identifying risk factors such as surgical approach, UPJO side, and drainage tube type is the key to better clinical management.

[0039] The present invention successfully identified eight risk factors for UTI after unilateral pyeloplasty in children. The developed prediction model is an effective tool to assist clinicians in identifying high-risk patients, which facilitates better preoperative planning and postoperative care.

[0040] Pre-experimental preparation: screening of subjects and preventive measures

[0041] We retrospectively analyzed the clinical data of children who underwent unilateral pyeloplasty at the Capital Institute of Pediatrics, Children's Hospital from January 2012 to January 2022. The subjects were selected according to the following criteria:

[0042] 1) Inclusion criteria: (1) Children who underwent unilateral pyeloplasty for the first time; (2) All patients who met the diagnostic criteria of the EAU UPJO guidelines; (3) All patients with complete follow-up records;

[0043] 2) Exclusion criteria: (1) patients with bilateral hydronephrosis; (2) patients receiving both double J stent (internal drainage) and temporary pyelostomy fistula (external drainage); (3) patients with more than one congenital urinary tract stenosis; (4) patients with renal duplication or double ureter;

[0044] (5) Guardians who refused to participate in the study and refused to sign the informed consent form; (6) Patients with incomplete follow-up data.

[0045] The protocol was approved by the Institutional Review Board before the study began, and all guardians of enrolled patients provided written informed consent. All parents provided written informed consent to participate in this study. Written informed consent from parents was obtained for the publication of any potentially identifiable images or data included in this article.

[0046] All subjects took cefuroxime sodium (30 mg / kg / day, divided into 3-4 times) and ceftriaxone sodium (20-80 mg / kg / day) 2 days before surgery. In addition, all patients received intravenous anti-inflammatory drugs after surgery. The Foley catheter was retained during surgery and removed 3-5 days after surgery. A double J-stent was placed during surgery and removed 8 weeks after surgery. In this study, the postoperative urinary tract infection period was defined as the occurrence of urinary tract infection within 8 weeks after surgery.

[0047] Statistical analysis method of the present invention:

[0048] All statistical analyses were performed using R4.1.3 software. The t-test was used for univariate analysis of continuous variables, and the chi-square test was used for grouped variables. The included participants were then randomly divided into training data and test data in a ratio of 7:3. Next, the least absolute selection and shrinkage operator (LASSO) regression method was used to implement the penalty term. Specifically, lasso regression was used to reduce the number of independent variables first. Lasso regression sets a penalty term λ after a linear model. As the penalty coefficient decreases, the parameters of the model gradually decrease until the minimum value of λmin is reached, which is the optimal number of independent variables for the model. However, it may be too strict in practical applications, so the number of independent variables corresponding to λ1se., which is 1 standard error away from λmin, is often used to include in the multifactor model. This makes the results more reliable and avoids over-ranking of independent variables. For example, some independent variables are not simply linearly related, but may be quadratic or cubic. When the logarithm (λ)min is reached, the optimal variable is determined by setting the random number seed, and the 10-fold cross-validation method is used. The compressed independent variables are included in the multivariate logistic regression, and features with p values ​​less than 0.05 are selected to be included in the model. Random forest models were used to score and rank the importance of variables. Using these methods, eight key variables were selected for nomogram construction. The validity of the model was evaluated by its discrimination ability, calibration ability, clinical relevance, and adaptability. The discrimination accuracy was assessed by the area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Calibration was assessed by juxtaposing predicted and actual outcomes, described by calibration curves from 400 bootstrap resamples. Clinical utility was evaluated using decision curve analysis (DCA) to calculate the net benefit at different probability thresholds. Finally, the generalizability of the model was evaluated based on its performance on the test dataset.

[0049] Example 1 Demographic and clinical characteristics of patients

[0050] A total of 732 eligible subjects were included. Among them, 229 subjects (31.8%) developed urinary tract infection (UTI), including 197 males (86%) and 32 females (14%) with an average age of 19 months and a weight of 11 kg. 17 subjects developed UTI after unilateral pyeloplasty, and the remaining 499 subjects (65%) did not develop UTI.

[0051] See Table 1 for details.

[0052] Table 1 Univariate analysis comparison between UTI group and non-UTI group

[0053]

[0054]

[0055] BMI: body mass index Cr: creatinine, BUN: blood urea nitrogen, eGFR: glomerular filtration rate N / L: neutrophils

[0056] / lymphocyte ratio, APD: anteroposterior diameter of renal pelvis, OP: open pyeloplasty, LP: laparoscopic pyeloplasty.

[0057] Example 2 Determination of predictive variables for UT I after unilateral UPJO surgery by Lasso regression analysis

[0058] The present invention uses the Lasso regression method to identify significant predictor variables. By setting the penalty term λ and using the Fo ld10 cross-validation method, a set of optimal variable sets was obtained. At the minimum value of log(λ), denoted as log(λ)min, a random seed was set to ensure the reproducibility of our results. The results show that the model contains 14 predictor variables, and their coefficients are shown in Table 1. The intercept coefficient is -0.610070252. The coefficient of sex (sex 1) is 0.115231168, indicating that for every unit change in sex, the risk of UTI after unilateral UPJO changes by approximately 0.115 units, keeping other variables unchanged. The time coefficient is 0.003582231, indicating that the risk of UTI after unilateral UPJO is slightly increased. The coefficient of drainage type (catheter 1) is 1.077044923, the highest among all variables, and has a significant effect on the risk of UTI after unilateral UPJO. The coefficients for infection and fistula history were 0.351245843 and 0.71021905, respectively, indicating that these clinical characteristics were positively correlated with the risk of UTI after unilateral UPJO. The coefficients for age and height were -0.00270678 and -0.004534245, respectively, indicating a negative correlation with the risk of UTI after unilateral UPJO, and the effect of age was relatively small. Other biochemical indicators such as TP, A / G ratio, Cr, BUN, and eGFR all had negative coefficients, suggesting that their increase was associated with a decreased risk of UTI after unilateral UPJO. Finally, the coefficient for APD was 0.006521787, which was positively correlated with the risk of UTI after unilateral UPJO, while the coefficient for renal cortical thickness was -0.152582245, indicating that a thinner renal cortex was associated with a higher risk of UTI after unilateral UPJO. For detailed information, see Tables 2 and Figure 1 .

[0059] Table 2. Association of clinical and demographic characteristics with UTI after unilateral UPJO using Lasso regression analysis

[0060]

[0061]

[0062]

[0063] Example 3 Logistic regression analysis of independent risk factors

[0064] l asso regression compression features are the number of variables Lamda.1se that are 1 standard error away from λmin, as shown in Table 3.

[0065] Table 3

[0066] variable coefficient (Intercept) -0.610070252 Gender (Male) 0.115231168 Surgery takes time 0.003582231 Postoperative drainage (DJ tube) 1.077044923 History of preoperative urinary tract infection 0.351245843 History of preoperative nephrostomy 0.71021905 age -0.00270678 height -0.004534245 Total protein tp -0.003928211 White ball vs ag 0.22179642 Serum creatinine value cr -0.006363223 Blood urea nitrogen value -0.010592261 Estimated glomerular filtration rate eGFR -0.007394012 Anterior-posterior diameter of renal pelvis 0.006521787 Renal cortical thickness -0.152582245

[0067] Using Lasso regression for variable selection and shrinkage, we integrated the selected variables into the logistic regression model to identify independent risk and protective factors, and considered the features with p value less than 0.05 to be statistically significant. For details, see Table 4. After compression, the independent variables were included in the logistic regression equation, and the features with p < 0.05 were screened and included in the model.

[0068] Table 4. Logistic regression analysis of the relationship between clinical and demographic characteristics and UTI after unilateral UPJO

[0069]

[0070]

[0071] The confusion matrix further provided insights into the classification accuracy of the model. For the non-event class (0), the model correctly predicted 298 instances, while misclassifying 45 instances, resulting in a class error of approximately 13.12%. In contrast, for the event class (1), the model accurately predicted 120 instances and misclassified 72 instances, with a higher class error of 37.5%. This difference suggests that the prediction accuracy of the model varies between different classes, which may be a focus for further model refinement. Finally, combining Lasso regression analysis, univariate logistic regression, and random forest models, eight variables, including gender, surgical time, postoperative drainage method, history of urinary tract infection before surgery, history of nephrostomy before surgery, age, serum urea nitrogen value, and renal cortical thickness, were selected as the best predictors of UT I risk after unilateral UPJO.

[0072] The final calculation formula is: score

[0073] Among them, logit(P)=1.857*(postoperative drainage method)+0.767*(gender)+1.369*(history of urinary tract infection before surgery)+1.450*(history of nephrostomy before surgery)+0.010*(surgery time)0.017*(age)-0.263*(renal cortical thickness)-0.159*(serum urea nitrogen value)-2.8748.

[0074] The threshold value is 0.5. After the score is calculated according to the formula, if the score is higher than the threshold value, it is judged that the risk of urinary tract infection after unilateral pyeloplasty in children is high; if the score is lower than the threshold value, it is judged that the risk of urinary tract infection after unilateral pyeloplasty in children is low.

[0075] Example 5 Development, performance and internal validation of a predictive nomogram

[0076] Using the eight best identified predictive factors, we constructed a nomogram to assess UTI after unilateral UPJO, such as Figure 3 This nomogram assigns a score to each variable, and the total score is derived from the sum of the individual scores of the eight variables. Figure 2 Based on the data from the literature, we provided a prediction scale associated with the probability of UTI after unilateral UPJO. An increase in the composite score indicated an increased risk of UTI.

[0077] The nomogram showed that the AUC statistics for the training cohort and validation cohort were 0.844 (95% CI: 0.808-0.879) and 0.789 (95% CI: 0.723-0.855), respectively ( Figure 3 The calibration plots for both cohorts were in close agreement with the ideal curves after bias correction, indicating consistency between the predicted probabilities and the observed results ( Figure 4 and 5 ). In addition, the Hosmer-Lemeshow test (goodness of fit test) yielded a p-value of 0.6657, confirming the adequacy of the model. Decision curve analysis (DCA) showed that the predictive application of the nomogram yielded substantial net benefits over the threshold probability range of 9% to 89% ( Figure 6 ). Overall, these results highlight the discrimination ability, calibration ability, clinical utility, and generalizability of the nomogram in this study.

[0078] Example 6 Practical application of the system of the present invention.

[0079] A total of 219 subjects, including 40 patients with urinary tract infection and 179 patients without infection, were screened and tested using the system of the present invention. The results are shown in Table 5.

[0080] The calculation formulas for sensitivity, specificity and accuracy are as follows:

[0081] Sensitivity = cases tested positive among infected persons / (cases tested positive among infected persons + cases tested negative among infected persons);

[0082] Specificity = cases that tested negative among uninfected persons / (cases that tested negative among uninfected persons + cases that tested positive among uninfected persons);

[0083] Accuracy = (cases tested positive among infected persons + cases tested negative among uninfected persons) / total number of samples.

[0084] Table 5 Markers and model test results of the present invention

[0085]

[0086] The accuracy of the marker system of the present invention in predicting urinary tract infection is 78.1%.

[0087] The above description is only an embodiment of the present invention. The protection scope of the present invention is not limited by these specific embodiments, but is determined by the claims of the present invention. For those skilled in the art, the present invention may have various changes and variations. Any modification, equivalent substitution, improvement, etc. made within the technical ideas and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A system for predicting the risk of urinary tract infection after unilateral pyeloplasty in children, characterized in that: The system comprises: A detection module, at least used to detect the content or quantity data of a marker, wherein the marker includes variable data in the case data; The calculation module is at least used to calculate the content or quantity data obtained by the detection module according to the judgment formula to obtain a score; the judgment module is at least used to compare the score with the threshold to judge the risk of urinary tract infection after unilateral pyeloplasty in children. If the score is higher than the threshold, it is judged that the risk of urinary tract infection after unilateral pyeloplasty in children is high, and if the score is lower than the threshold, it is judged that the risk of urinary tract infection after unilateral pyeloplasty in children is low; The output module is at least used to output the result of the judgment module.

2. The system according to claim 1, characterized in that The variable data in the case data include: gender, operation time, postoperative drainage method, history of urinary tract infection before operation, history of nephrostomy before operation, age, serum urea nitrogen value, and renal cortical thickness value.

3. The system according to claim 1, characterized in that The judgment formula is as follows: Score Among them, logit(P)=1.857*(postoperative drainage method)+0.767*(gender)+1.369*(history of urinary tract infection before surgery)+1.450*(history of nephrostomy before surgery)+0.010*(surgery time)0.017*(age)-0.263*(renal cortical thickness)-0.159*(serum urea nitrogen value)-2.8748.

4. The system according to claim 3, characterized in that The gender, age, postoperative drainage method, history of urinary tract infection before surgery, and history of nephrostomy before surgery are all calculated in an assigned manner, and age is calculated in months.

5. The system according to claim 1, characterized in that The threshold is 0.

5.

6. The system according to claim 1, characterized in that The system further comprises a data acquisition module, which is at least used for sample extraction to obtain a sample to be detected.

7. The system according to claim 6, characterized in that The sample is a biological fluid sample.

8. The system according to any one of claims 1 to 7, characterized in that: The system is used to prepare one or more of an instrument, a device or a kit for predicting the risk of urinary tract infection after unilateral pyeloplasty in children.

9. An information data processing terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the processor executes the computer program to calculate the system described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that: The program is executed by a processor to calculate the system described in any one of claims 1-7.

Images