Risk Assessment for Upper Urinary Tract Deterioration in Patients With Neurogenic Bladder: A Retrospective Study Using Interpretable Machine Learning Algorithms

Article information

Int Neurourol J. 2026;30(3):266-279
Publication date (electronic) : 2026 June 28
doi : https://doi.org/10.5213/inj.2652084.042
1The Second Qilu Hospital of Shandong University, Jinan, China
2School of Rehabilitation of Capital Medical University, Department of Urology, China Rehabilitation Research Center, Beijing, China
3China Rehabilitation Science Institute, Beijing, China
4Urological and Pelvic Floor Rehabilitation Center, Beijing Bo’ai Hospital, Beijing, China
5University of Health and Rehabilitation Sciences, Qingdao, China
6Beijing Key Laboratory of Innovation and Transformation in Diagnosis, Treatment and Rehabilitation of Neurogenic Bladder, Beijing, China
Corresponding author: Limin Liao School of Rehabilitation of Capital Medical University, Department of Urology, China Rehabilitation Research Center, 10 Jiaomen North Road, Beijing 100068, China Email: lmliao1964@163.com, lmliao@263.net
Received 2026 March 18; Accepted 2026 May 19.

Abstract

Purpose

Upper urinary tract deterioration (UUTD) is a burdensome complication in patients with neurogenic bladder (NB). This study aimed to develop a practical machine learning (ML) tool for assessing UUTD risk.

Methods

The study included 2,393 patients with NB who underwent clinical and urodynamic examinations. Of these, 2,001 patients from northern regions of our country were assigned to training and validation sets, and 392 patients from southern regions formed the test set for internal-external validation. Clinical and urodynamic parameters were screened using the Boruta, least absolute shrinkage and selection operator regression, and decision tree algorithms; features identified by all 3 algorithms were used to develop the ML models. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis. SHapley Additive exPlanations analysis provided visual interpretations for individual patients.

Results

Six features, notably bladder management method and detrusor sphincter dyssynergia, were identified as risk factors for UUTD. Of the 4 ML algorithms, the Gaussian Naive Bayes (GNB) model had the best overall performance, with AUCs (95% confidence intervals) of 0.884 (0.867–0.901), 0.868 (0.829–0.907), and 0.878 (0.839–0.916) in the training, validation, and test sets, respectively. Subgroup analyses by imaging outcome, NB etiology, and age showed consistent performance of the GNB model, with AUCs ranging from 0.815 to 0.895. The GNB model incorporating the core variables was presented through a practical web-based tool for individualized risk assessment.

Conclusions

We developed a ML model to assess UUTD risk in patients with NB. The findings emphasize the importance of combining bladder management method with urodynamic evaluation when assessing UUTD risk, and the model provides a practical tool for individualized assessment. Independent external validation of the model is warranted.

INTRODUCTION

Neurogenic bladder (NB) is common among patients with neurological diseases, affecting approximately 83%–95% of those with spinal cord injury and 96% of those with spina bifida [1]. Impaired communication between the nervous system and the lower urinary tract characterizes NB and can lead to progressive upper urinary tract deterioration (UUTD). Despite advances in urological management, UUTD remains the main complication of NB, with 14.6% to 35.2% of patients developing varying degrees of renal impairment [2, 3].

Identifying UUTD early is critical to timely intervention and the prevention of irreversible renal damage. Previous studies have used traditional statistical methods to examine relationships between UUTD and urodynamic parameters [3], including detrusor leak point pressure (DLPP) [4-6] and bladder compliance [7, 8]. These studies have informed UUTD risk assessment, but their focus on individual variables or prespecified interactions limits their ability to capture complex interactions among multiple contributing factors. Furthermore, no large-scale study has comprehensively evaluated UUTD risk models across distinct NB subgroups, despite the diversity of the affected population described above.

Machine learning (ML) algorithms have emerged in recent years as powerful tools for analyzing complex data and identifying nonlinear relationships [9-11]. Given the complex effects of clinical and urodynamic parameters on UUTD [3], we aimed to develop an interpretable ML model for UUTD risk assessment using integrated clinical and urodynamic data from a large cohort of patients with NB. The study sought to provide clinicians with a practical tool for identifying patients at elevated risk of UUTD and supporting more informed, personalized management plans (Fig. 1). Reporting followed the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis+Artificial Intelligence statement [12] and the Transparency in the Reporting of Artificial Intelligence guideline [13].

Fig. 1.

Overall study workflow. SHAP, SHapley Additive exPlanations.

MATERIALS AND METHODS

Study Participants

Clinical and urodynamic data were retrospectively collected from patients with NB at our center between January 2007 and August 2025. Patients who underwent video-urodynamics (VUDS) at their first visit to our center were included. Exclusion criteria were (1) urinary diversion sur-gery before the initial visit; (2) kidney dysfunction caused by hypertension, heart disease, renal diseases, or other systemic conditions; (3) ureteral dilation or hydronephrosis caused by urinary stones; (4) congenital malformations of the renal pelvis or ureter; and (5) incomplete clinical data that would affect subsequent analysis.

Patients were classified as being from northern or southern regions according to their household registration status and permanent residential address. The 2,001 eligible patients with NB from northern regions were randomly allocated to the training and validation sets in a 7:3 ratio. The 392 patients from southern regions formed the test set for internal-external validation (Fig. 1). The institutional review board at the China Rehabilitation Research Center approved this study (No. 2023-075-01), which was conducted in accordance with the Declaration of Helsinki.

Clinical Assessment

VUDS followed the “Good Urodynamic Practices” recommended by the International Continence Society (ICS) [14]. The following urodynamic data were collected in accordance with ICS guidelines: maximum cystometric capacity (MCC), volume at first sensation of bladder filling, volume at strong desire to void, maximum detrusor pressure (Pdetmax) during the filling phase, end-filling pressure, presence of detrusor overactivity (DO), bladder compliance (calculated manually from urodynamic traces), DLPP, detrusor sphincter dyssynergia (DSD), detrusor underactivity (DU), acontractile detrusor (AcD), and postvoid residual urine volume. DO was defined as any involuntary detrusor contraction during the filling phase. Bladder sensation was classified as normal, increased, reduced, or absent according to the corresponding criteria [15]. Because DLPP was not observed in all patients [7], end-filling pressure was used in subsequent analyses when DLPP was unavailable [16, 17]. Maximum urethral closure pressure (MUCP) was measured using urethral profilometry.

NB etiology was recorded and classified into 3 groups: spinal cord injury, spinal dysplasia, and other etiologies (including iatrogenic pelvic neuropathy, myelitis, diabetic bladder, and others). Bladder management methods used for more than 6 months were categorized as self-voiding, abdominal pressure-assisted urination, clean intermittent catheterization (CIC), indwelling catheterization, or other methods (such as use of an external collection device). Common upper urinary tract imaging examinations included cystourethrography, urinary tract ultrasonography, magnetic resonance urography (MRU), and radioisotope renography, performed concurrently at the initial clinical visit. All included patients underwent at least 2 of these imaging modalities to determine upper urinary tract status. Vesicoureteral reflux (VUR) was determined according to the International Reflux Grading System [18]. MRU and/or urinary tract ultrasonography was used to assess hydronephrosis [19-21], and radioisotope renography was used to determine measured glomerular filtration rate (mGFR). UUTD was defined as at least one of the following upper urinary tract abnormalities: VUR, hydronephrosis, or mGFR below 60 mL/min/1.73 m2.

Data Preprocessing

Multiple imputation was used for variables with a low proportion of missing values. For variables with a high proportion of missing values (≥30%), data missing important patient information were excluded when no reasonable substitute value was available. DLPP was missing in more than 30% of cases; given its potential value and relationship to end-filling pressure, the latter was substituted for missing DLPP values. Continuous variables were standardized using z-score normalization. Binary categorical variables were coded as 0 or 1, and unordered multicategory variables were converted to dummy variables using one-hot encoding. To avoid data leakage, all preprocessing steps were applied independently to the training, validation, and test sets.

Model Development

Feature selection comprised dataset analysis and feature prioritization. Following initial screening, feature importance in the training set was assessed using 3 algorithms: Boruta, least absolute shrinkage and selection operator (LASSO), and decision tree (DT). A Venn diagram [22] was used to identify features selected by all 3 algorithms for inclusion in the UUTD risk model.

Four widely used ML algorithms were used to develop the models: support vector machine (SVM), multilayer perceptron (MLP), Gaussian Naive Bayes (GNB), and k-nearest neighbor (KNN). To address potential class imbalance in upper urinary tract outcomes, the synthetic minority oversampling technique was applied exclusively to the training set before model training. Hyperparameters for each algorithm were tuned by grid search with 10-fold cross-validation in the training set; only the optimal nondefault settings are reported below. SVM uses supervised learning to minimize structural risk and improve generalization; the optimal settings in this study were “C=0.1, max_iter=50.” MLP stacks multiple layers of neurons with nonlinear activation functions to address linearly inseparable problems [23]; its optimal settings were “activation=logistic, hidden_layer_sizes=(30, 30), max_iter=100.” GNB handled categorical features as pre-encoded variables [9, 24], with “var_ smoothing=1e-07” to maintain computational efficiency and robustness to irrelevant features. KNN used distance metrics optimized for continuous variables, with pre-encoded categorical variables also included as input features; its final settings were “n_neighbors=6, leaf_size=10.”

Statistical Analysis and Model Evaluation

Baseline continuous variables were compared using 1-way analysis of variance or Student t-test, and categorical variables were compared using Fisher exact test or the chi-square test. Models were trained on the training set and evaluated on the validation and test sets. Performance was assessed using receiver operating characteristic (ROC) curves and the area under the curve (AUC), together with calibration, decision, and precision-recall curves. Subgroup analyses by upper urinary tract imaging outcome, NB etiology, and age were performed in the validation set to assess model performance across patient subgroups (Fig. 1). SHapley Additive exPlanations (SHAP) analysis was used to explain each feature’s contribution to the model. All statistical analyses were performed in R ver. 4.2.3 (R Foundation for Statistical Computing, Austria) and Python ver. 3.11.4 (Python Software Foundation, USA), with P-values <0.05 considered statistically significant.

RESULTS

Patient Characteristics

The training, validation, and test sets included 1,400, 601, and 392 patients with NB, respective-ly; Table 1 summarizes their characteristics. Across all patients, the proportions with VUR, hydronephrosis, and UUTD were 19.64%, 24.28%, and 32.97%, respectively (Table 1). Upper urinary tract imaging outcomes did not differ significantly among the training, validation, and test sets. Age, bladder management method, MCC, and DSD differed significantly among the 3 sets, whereas the other clinical and urodynamic variables did not (Table 1). Most clinical and urodynamic variables differed significantly between patients with and without UUTD; the exceptions were sex, DO, AcD, DU, and MUCP (Table 2).

Patient characteristics of the training, validation and test sets

Patient characteristics stratified based on upper urinary tract deterioration outcome

Feature Selection

Features were selected in the training set using Boruta (Fig. 2A), LASSO regression (Fig. 2B and C), and DT (Fig. 2D). Venn diagram analysis identified 6 features common to all 3 methods: disease duration, abdominal pressure-assisted urination, bladder compliance, DSD, MCC, and Pdetmax (Fig. 2E).

Fig. 2.

Feature selection using multiple algorithms. (A) Important variables selected by the Boruta algorithm (green) according to z-scores. (B and C) LASSO regression: coefficient paths (B) and cross-validation curve (C). (D) The 10 most important variables selected by the DT algorithm according to coefficient values. (E) Venn diagram showing the overlap in features selected by the 3 algorithms. LASSO, least absolute shrinkage and selection operator; DT, decision tree.

Performance of ML Models

The 6 selected variables were used as independent factors to develop UUTD risk assessment models. Ten-fold cross-validation in the training set provided mean performance estimates for the 4 ML models (Table 3; Fig. 3). In the training set, KNN outperformed the other models, with an AUC of 0.932 and a 95% confidence interval (CI) of 0.921–0.943, compared with 0.884 (95% CI, 0.867–0.901) for GNB, 0.779 (95% CI, 0.754–0.804) for MLP, and 0.792 (95% CI, 0.770–0.815) for SVM (Fig. 3). In the validation and test sets, however, GNB outperformed the other models. Its AUCs in these sets were 0.868 (95% CI, 0.829– 0.907) and 0.878 (95% CI, 0.839–0.916), respectively. Fig. 3 compares the ROC, calibration, decision, and precision-recall curves for the 4 models. GNB had Brier scores of 0.131 and 0.134 in the validation and test sets, respectively, lower than those of the other models (Fig. 3D and G). Decision curve analysis (DCA) showed that GNB provided greater net clinical benefit than the other models across a broad range of thresholds (10%–90%) covering most clinical scenarios (Fig. 3E and H). Precision-recall curves likewise showed better performance for GNB than for the other models (Fig. 3F and I). Table 3 presents the remaining performance metrics for each model across the 3 sets.

Comparison of multiple machine learning evaluation indexes in the training, validation, and test sets

Fig. 3.

Performance of multiple machine learning algorithms in the training, validation, and test sets. (A–C) Comparison of ROC curves in the training (A), validation (B), and test (C) sets. (D) Calibration curves in the validation set. (E) Decision curve analysis in the validation set. (F) PR curves in the validation set. (G) Calibration curves in the test set. (H) Decision curve analysis in the test set. (I) PR curves in the test set. ROC, receiver operating characteristic; PR, precision-recall; GNB, Gaussian Naive Bayes; MLP, multilayer perceptron; SVM, support vector machine; KNN, K-nearest neighbor.

Subgroup analyses by upper urinary tract imaging outcome, NB etiology, and age were then performed in the validation set to assess ML model performance across patient subgroups (Table 4; Fig. 4). GNB outperformed the other models across the subgroups, with AUCs ranging from 0.815 to 0.895. For GNB, the AUCs for VUR, hydronephrosis, and the composite outcome (UUTD) were 0.815, 0.832, and 0.868, respectively, suggesting that the composite outcome might reflect upper urinary tract status more comprehensively than a single modality. Fig. 4 presents the ROC curves for GNB in each subgroup.

Subgroup analysis results of the validation set according to imaging outcome, neurogenic bladder etiology, and age

Fig. 4.

Subgroup analyses of the GNB model by imaging outcome, NB etiology, and age. (A–C) ROC curves for VUR (A), hydronephrosis (B), and UUTD (C). (D–F) ROC curves by NB etiology: spinal cord injury (D), spinal dysplasia (E), and other etiologies (F). (G and H) ROC curves by age: patients younger than 18 years (G) and aged 18 years or older (H). (I) Radar plot of the area under the ROC curve in each subgroup. GNB, Gaussian Naive Bayes; NB, neurogenic bladder; ROC, receiver operating characteristic; VUR, vesicoureteral reflux; UUTD, upper urinary tract deterioration.

Based on the ROC curves, calibration plots, DCA, precisionrecall curves, and subgroup analyses, GNB was selected as the final model for assessing UUTD risk in patients with NB. SHAP plots were used to visualize how the features affected GNB model output. Each feature’s contribution to model output was quantified and displayed in SHAP summary plots (Fig. 5A) to identify the most relevant risk factors. The SHAP importance plot (Fig. 5B) further illustrates the influence of individual features on GNB. In descending order of contribution, the features were abdominal pressure-assisted urination, DSD, disease duration, MCC, Pdetmax, and bladder compliance (Fig. 5A and B). A web-based calculator incorporating these 6 variables in the GNB model was developed to support individualized UUTD risk assessment and improve accessibility and ease of use (https://www.xsmartanalysis.com/model/list/predict/model/html?mid=31532&symbol=6177086jXjm8TO993Oo3; Fig. 5C). Fig. 5D shows SHAP force plots for 2 individual samples, with the calculated risk probabilities and corresponding explanations.

Fig. 5.

SHAP analysis of the GNB model for UUTD risk assessment. (A) SHAP summary plot. The horizontal position of each point represents its SHAP value, indicating a feature’s effect on an individual patient’s risk (red=high, blue=low). (B) SHAP importance ranking, with features ordered vertically by mean absolute SHAP values. (C) Calculator for UUTD risk assessment. (D) Individualized risk assessment for 2 samples. Horizontal plots show the magnitude of feature effects, and output values represent the probability of UUTD. These plots show variable importance and individual risk contributions within the GNB model. SHAP, SHapley Additive exPlanations; GNB, Gaussian Naive Bayes; UUTD, upper urinary tract deterioration; Pdetmax, maximum detrusor pressure; MCC, maximum cystometric capacity; DSD, detrusor sphincter dyssynergia.

DISCUSSION

Previous studies have examined several factors associated with UUTD risk, although their clinical value, applicability, and ease of use remain limited to some extent [7, 8, 25]. Comprehensive models are therefore still needed to assess UUTD risk in patients with NB. We developed an ML model for this purpose using a large cohort of patients with NB. Data-driven feature selection identified bladder management method, DSD, disease duration, Pdetmax, bladder com-pliance, and MCC as the core factors associated with UUTD risk. These findings add to the evidence on UUTD risk assessment, and the model provides a practical assessment tool.

Among the ML algorithms evaluated, KNN achieved the highest AUC in the training set, but its performance declined in the validation and test sets, suggesting potential overfitting. GNB, by comparison, performed consistently across all sets, with AUCs of 0.884, 0.868, and 0.878 in the training, validation, and test sets, respectively. Age, bladder management methods, MCC, and DSD differed significantly among the 3 sets. These results indicated stable GNB performance that was unaffected by the observed differences in patient characteristics under these conditions.

Bladder management method was an important risk factor for UUTD in our model. Abdominal pressure-assisted urination, in particular, was associated with a higher risk than other methods, such as CIC. This method involves applying manual pressure to the lower abdomen to aid bladder emptying, which can increase intravesical pressure during voiding. Chronic or recurrent elevations in intravesical pressure can damage the vesicoureteral junction, resulting in VUR and subsequent upper urinary tract damage. Other bladder management methods, in contrast, maintain lower, more stable intravesical pressures and thereby reduce UUTD risk.

DSD was also identified as a significant risk factor for UUTD in our model. In DSD, the urethral sphincter contracts involuntarily during detrusor contraction, causing functional outlet obstruction. The resulting increase in intravesical pressure during voiding can damage the upper urinary tract over time. Whereas abdominal pressure-assisted urination raises pressure externally, the elevation in intravesical pressure associated with DSD results from intrinsic neurological dysfunction. Early recognition of DSD and treatment with α-adrenergic receptor antagonists, intraurethral botulinum toxin injection, or other interventions can relieve DSD and prevent UUTD.

Disease duration was associated with an increased risk of UUTD in our model, consistent with previous studies [8, 26]. This association likely involves progressive remodeling and fibrosis of the bladder wall as chronic bladder dysfunction persists [27, 28]. Recurrent urinary tract infections, which become more frequent with longer NB duration, can further accelerate bladder wall fibrosis and damage the upper urinary tract [8]. These findings emphasize the importance of long-term followup and regular monitoring of patients with NB to detect UUTD early and allow prompt intervention.

Pdetmax, bladder compliance, and MCC each contributed to UUTD risk through specific pathophysiological mechanisms. In this study, Pdetmax was the value obtained during the filling phase. Elevated filling-phase Pdetmax indicates increased resting detrusor tone and can lead to persistently high intravesical pressure during bladder storage. Over time, this pressure elevation can impair ureteral peristalsis and cause hydronephrosis or renal insufficiency, thereby damaging the upper urinary tract [29]. Bladder compliance measures the bladder’s capacity to stretch and accommodate urine. With low compliance, small increases in bladder volume cause large increases in intravesical pressure, which can damage the upper urinary tract during storage [3, 7]. MCC measures the greatest volume the bladder can hold. Reduced MCC is associated with impaired bladder compliance and progressive detrusor fibrosis; these changes disrupt normal low-pressure storage, transmit higher pressure to the upper urinary tract, and consequently lead to UUTD [26]. These findings underscore the complexity of UUTD pathogenesis and the need to assess multiple aspects of bladder function comprehensively in patients with NB.

Subgroup analyses by upper urinary tract imaging outcome, NB etiology, and age demonstrated model performance across different patient groups. Earlier studies evaluated UUTD risk either in patients with a specific etiology [5, 7, 16, 17, 26] or in those with various neurological etiologies [3, 6, 8, 25], without subgroup analyses by NB etiology, imaging outcome, and age. After developing the UUTD risk assessment model for all patients with NB, we divided patients into spinal cord injury, spinal dysplasia, and other etiological subgroups; GNB outperformed the other models in each subgroup. Analyses by imaging outcome and age yielded similar results. These findings suggest that the model can assess UUTD risk across different groups of patients with NB. Because VUR, hydronephrosis, and reduced glomerular filtration rate are all clinically important manifestations of UUTD, we used a composite outcome to identify UUTD as comprehensively as possible; the relatively better model performance with the composite outcome than with a single modality indirectly reflected this more comprehensive assessment.

SHAP analysis was used to interpret model performance and assess the relative importance of each feature. By showing how individual features contribute to UUTD risk assessment, SHAP analysis may help clinicians develop personalized management plans for patients with NB. For example, patients who use ab-dominal pressure-assisted urination and have DSD require an appropriate bladder management method, such as CIC, to reduce intravesical pressure, whereas those with long disease duration and low bladder compliance can benefit from closer monitoring and timely intervention.

To our knowledge, this is the largest original study of UUTD risk assessment in patients with NB to confirm upper urinary tract outcomes using at least 2 imaging modalities and to incorporate both clinical and urodynamic variables. It is also the first to comprehensively evaluate UUTD risk models across different subgroups of patients with NB. The study nevertheless has several limitations. First, its retrospective, single-center design and restriction to patients under-going VUDS may introduce selection bias and limit generalizability. Although VUDS was indicated for routine evaluation and diagnosis of NB in this study, its use in clinical practice may vary with factors such as disease severity, economic status, and adherence. Because data were collected retrospectively from medical records, some relevant clinical parameters may not have been included. Second, although the study included a range of variables, other potential risk factors, such as radiomic features and medication use, were omitted for certain reasons. Future models should consider these factors to improve performance. Third, although patients came from multiple regions of our country, the geographically distinct test set provided internal-external validation and did not constitute independent external validation; independent multicenter or international validation is warranted. Fourth, most patients with NB underwent VUDS only once at our center, preventing assessment of changes in urinary tract function over time. Prospective longitudinal studies would help further validate the model’s capabilities and represent a key direction for future research.

In conclusion, we developed an ML model that integrates clinical and urodynamic parameters to assess UUTD risk in patients with NB. The findings emphasize the importance of bladder management method and urodynamic evaluation in this population, and the model provides a practical tool for UUTD risk assessment. Independent external validation of the model remains necessary.

Notes

Grant/Fund Support

This study was supported by the Key Technologies Research and Development Program (2023YFC3605300 and 2023YFC3605301) and the Fundamental Research Funds for Central Public Welfare Research (2023CZ-1, CRSI2025ZD-1-6).

Research Ethics

This study was approved by the institutional review board at the China Rehabilitation Research Center (No. 2023-075-01) and complied with the Declaration of Helsinki; this study was a retrospective study, and the data were processed to remove relevant private in-formation of patients, so the ethics committee agreed to exempt informed consent.

Conflict of Interest

No potential conflict of interest relevant to this article was reported.

AUTHOR CONTRIBUTION STATEMENT

· Conceptualization: LL

· Data curation: PZ, XL, TX

· Formal analysis: XL, TX

· Funding acquisition: LL

· Methodology: PZ, JW

· Project administration: LL, JW

· Visualization: XL, TX

· Writing - original draft: PZ

· Writing - review & editing: LL

References

1. Blok B, Castro-Diaz D, Del Popolo G, Groen J, Hamid R, Karsenty G, et al. EAU guidelines on neuro-urology Arnhem (The Netherlands): EAU Guidelines Office; 2025.
2. Fischer MJ, Krishnamoorthi VR, Smith BM, Evans CT, St Andre JR, Ganesh S, et al. Prevalence of chronic kidney disease in patients with spinal cord injuries/disorders. Am J Nephrol 2012;36:542–8.
3. Musco S, Padilla-Fernández B, Del Popolo G, Bonifazi M, Blok BFM, Groen J, et al. Value of urodynamic findings in predicting upper urinary tract damage in neuro-urological patients: a systematic review. Neurourol Urodyn 2018;37:1522–40.
4. McGuire EJ, Woodside JR, Borden TA, Weiss RM. Prognostic value of urodynamic testing in myelodysplastic patients. J Urol 1981;126:205–9.
5. McGuire EJ, Woodside JR, Borden TA, Weiss RM. Prognostic value of urodynamic testing in myelodysplastic patients. 1981. J Urol 2002;167(2 Pt 2):1049–53. discussion 1054.
6. Ozkan B, Demirkesen O, Durak H, Uygun N, Ismailoglu V, Cetinel B. Which factors predict upper urinary tract deterioration in overactive neurogenic bladder dysfunction? Urology 2005;66:99–104.
7. Haudebert C, Hascoet J, Samson E, Jezequel M, Voiry C, Brochard C, et al. Risk factors for upper urinary tract deterioration in adult patients with spina bifida. World J Urol 2023;41:1187–92.
8. Wang W, Fang H, Xie P, Cao Q, He L, Cai W. Create a predictive model for neurogenic bladder patients: upper urinary tract damage predictive nomogram. Int J Neurosci 2019;129:1240–6.
9. Islam MA, Majumder MZH, Miah MS, Jannaty S. Precision healthcare: a deep dive into machine learning algorithms and feature selection strategies for accurate heart disease prediction. Comput Biol Med 2024;176:108432.
10. Deivayanai VC, Swaminaathanan P, Vickram AS, Saravanan A, Bibi S, Aggarwal N, et al. Transforming healthcare: the impact of artificial intelligence on diagnostics, pharmaceuticals, and ethical considerations - a comprehensive review. Int J Surg 2025;111:4666–93.
11. Kim JM, Jung H, Kwon HE, Ko Y, Jung JH, Kwon H, et al. Predicting prognostic factors in kidney transplantation using a machine learning approach to enhance outcome predictions: a retrospective cohort study. Int J Surg 2024;110:7159–68.
12. Collins GS, Moons KGM, Dhiman P, Riley RD, Beam AL, Van Calster B, et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ 2024;385e078378.
13. Agha R, Mathew G, Rashid R, Kerwan A, Al-Jabir A, Sohrabi C, et al. Transparency In The reporting of Artificial INtelligence – the TITAN guideline. Premier J Sci 2025;10:100082.
14. Schäfer W, Abrams P, Liao L, Mattiasson A, Pesce F, Spangberg A, et al. Good urodynamic practices: uroflowmetry, filling cystometry, and pressure-flow studies. Neurourol Urodyn 2002;21:261–74.
15. Abrams P, Cardozo L, Fall M, Griffiths D, Rosier P, Ulmsten U, et al. The standardisation of terminology of lower urinary tract function: report from the Standardisation Sub-committee of the International Continence Society. Neurourol Urodyn 2002;21:167–78.
16. Dik P, Klijn AJ, van Gool JD, de Jong-de Vos van Steenwijk CC, de Jong TP. Early start to therapy preserves kidney function in spina bifida patients. Eur Urol 2006;49:908–13.
17. Veenboer PW, Bosch JL, Rosier PF, Dik P, van Asbeck FW, de Jong TP, et al. Cross-sectional study of determinants of upper and lower urinary tract outcomes in adults with spinal dysraphism--new recommendations for urodynamic followup guidelines? J Urol 2014;192:477–82.
18. O’Neil BB, Cartwright PC, Maves C, Hoeg K, Presson AP, Wallis MC. Reliability of voiding cystourethrogram for the grading of vesicoureteral reflux. J Pediatr Urol 2014;10:107–11.
19. Liao L. Evaluation and management of neurogenic bladder: what is new in China? Int J Mol Sci 2015;16:18580–600.
20. Liao L, Zhang F, Chen G. New grading system for upper urinary tract dilation using magnetic resonance urography in patients with neurogenic bladder. BMC Urol 2014;14:38.
21. Zhou Z, Liao L, Li X, Wang X, Li X, Zhao H, et al. Reliability assessment of the upper urinary tract dilation grading system based on magnetic resonance urography in patients with neurogenic bladder. BMC Urol 2022;22:97.
22. Heberle H, Meirelles GV, da Silva FR, Telles GP, Minghim R. InteractiVenn: a web-based tool for the analysis of sets through Venn diagrams. BMC Bioinformatics 2015;16:169.
23. Sathish T, Sunagar P, Singh V, Boopathi S, Sathyamurthy R, Al-Enizi AM, et al. Characteristics estimation of natural fibre reinforced plastic composites using deep multi-layer perceptron (MLP) technique. Chemosphere 2023;337:139346.
24. Wang H, Shao Y, Zhou S, Zhang C, Xiu N. Support Vector Machine Classifier via L0/1 soft-margin loss. IEEE Trans Pattern Anal Mach Intell 2022;44:7253–65.
25. Zhou Z, Hang T, Wu J, Xiang T, Zhong P, Liu X, et al. Fully end-toend automated bladder segmentation and risk assessment based on bladder morphology features for predicting upper urinary tract dysfunction. Int J Surg 2026;112:303–13.
26. Deng H, Wang Z, Liao L, Wu J, Wang Y. Risk factors predicting upper urinary tract damage in patients with myelodysplasia: data analysis of 637 cases from a single center. Int Neu-rourol J 2022;26(Suppl 1):S22–9.
27. Peyronnet B, Richard C, Bendavid C, Naudet F, Hascoet J, Brochard C, et al. Urinary TIMP-2 and MMP-2 are significantly associated with poor bladder compliance in adult patients with spina bifida. Neurourol Urodyn 2019;38:2151–8.
28. Fry CH, Kitney DG, Paniker J, Drake MJ, Kanai A, Andersson KE. Fibrosis and the bladder, implications for function ICI-RS 2017. Neurourol Urodyn 2018;37(S4):S7–12.
29. Ghobish AG. Storage detrusor pressure in bilateral hydroureteronephrosis. Eur Urol 2001;39:571–4.

Article information Continued

Fig. 1.

Overall study workflow. SHAP, SHapley Additive exPlanations.

Fig. 2.

Feature selection using multiple algorithms. (A) Important variables selected by the Boruta algorithm (green) according to z-scores. (B and C) LASSO regression: coefficient paths (B) and cross-validation curve (C). (D) The 10 most important variables selected by the DT algorithm according to coefficient values. (E) Venn diagram showing the overlap in features selected by the 3 algorithms. LASSO, least absolute shrinkage and selection operator; DT, decision tree.

Fig. 3.

Performance of multiple machine learning algorithms in the training, validation, and test sets. (A–C) Comparison of ROC curves in the training (A), validation (B), and test (C) sets. (D) Calibration curves in the validation set. (E) Decision curve analysis in the validation set. (F) PR curves in the validation set. (G) Calibration curves in the test set. (H) Decision curve analysis in the test set. (I) PR curves in the test set. ROC, receiver operating characteristic; PR, precision-recall; GNB, Gaussian Naive Bayes; MLP, multilayer perceptron; SVM, support vector machine; KNN, K-nearest neighbor.

Fig. 4.

Subgroup analyses of the GNB model by imaging outcome, NB etiology, and age. (A–C) ROC curves for VUR (A), hydronephrosis (B), and UUTD (C). (D–F) ROC curves by NB etiology: spinal cord injury (D), spinal dysplasia (E), and other etiologies (F). (G and H) ROC curves by age: patients younger than 18 years (G) and aged 18 years or older (H). (I) Radar plot of the area under the ROC curve in each subgroup. GNB, Gaussian Naive Bayes; NB, neurogenic bladder; ROC, receiver operating characteristic; VUR, vesicoureteral reflux; UUTD, upper urinary tract deterioration.

Fig. 5.

SHAP analysis of the GNB model for UUTD risk assessment. (A) SHAP summary plot. The horizontal position of each point represents its SHAP value, indicating a feature’s effect on an individual patient’s risk (red=high, blue=low). (B) SHAP importance ranking, with features ordered vertically by mean absolute SHAP values. (C) Calculator for UUTD risk assessment. (D) Individualized risk assessment for 2 samples. Horizontal plots show the magnitude of feature effects, and output values represent the probability of UUTD. These plots show variable importance and individual risk contributions within the GNB model. SHAP, SHapley Additive exPlanations; GNB, Gaussian Naive Bayes; UUTD, upper urinary tract deterioration; Pdetmax, maximum detrusor pressure; MCC, maximum cystometric capacity; DSD, detrusor sphincter dyssynergia.

Table 1.

Patient characteristics of the training, validation and test sets

Variable All (n = 2,393) Training set (n = 1,400) Validation set (n = 601) Test set (n = 392) P-value
Age (yr) 34.97 ± 17.56 35.32 ± 17.49 36.35 ± 17.70 31.60 ± 17.19 < 0.001
Sex 0.193
 Male 1,570 (65.61) 916 (65.43) 409 (68.05) 245 (62.50)
 Female 823 (34.39) 484 (34.57) 192 (31.95) 147 (37.50)
Disease duration (yr) 7.61 ± 10.87 7.85 ± 11.22 7.82 ± 11.13 6.43 ± 8.95 0.062
NB etiology 0.334
 Spinal cord injury 1,087 (45.42) 645 (46.07) 253 (42.10) 189 (48.21)
 Spinal dysplasia 751 (31.38) 435 (31.07) 196 (32.61) 120 (30.61)
 Others 555 (23.19) 320 (22.86) 152 (25.29) 83 (21.17)
Bladder management method 0.006
 Clean intermittent catheterization 531 (22.19) 315 (22.50) 109 (18.14) 107 (27.30)
 Indwelling catheterization 748 (31.26) 448 (32.00) 204 (33.94) 96 (24.49)
 Self-voiding 452 (18.89) 261 (18.64) 123 (20.47) 68 (17.35)
 Abdominal pressure-assisted urination 518 (21.65) 287 (20.50) 133 (22.13) 98 (25.00)
 Others 144 (6.02) 89 (6.36) 32 (5.32) 23 (5.87)
Urodynamic diagnosis 0.839
 Detrusor overactivity 868 (36.27) 511 (36.50) 216 (35.94) 141 (35.97)
 Acontractile detrusor 1,279 (53.45) 745 (53.21) 318 (52.91) 216 (55.10)
 Detrusor underactivity 246 (10.28) 144 (10.29) 67 (11.15) 35 (8.93)
Bladder compliance (mL/cm H2O) 21.89 ± 13.56 21.86 ± 13.77 22.09 ± 12.75 21.66 ± 13.99 0.888
Detrusor sphincter dyssynergia 520 (21.73) 319 (22.79) 152 (25.29) 49 (12.50) < 0.001
Maximum cystometric capacity (mL) 303.86 ± 137.00 308.61 ± 136.73 306.82 ± 138.93 282.38 ± 132.86 0.003
Bladder sensation 0.135
 Normal 886 (37.02) 531 (37.93) 219 (36.44) 136 (34.69)
 Increased 347 (14.50) 196 (14.00) 95 (15.81) 56 (14.29)
 Reduced 381 (15.92) 235 (16.79) 96 (15.97) 50 (12.76)
 Absent 779 (32.55) 438 (31.29) 191 (31.78) 150 (38.27)
Postvoid residual urine volume (mL) 243.37 ± 186.70 248.97 ± 190.98 239.41 ± 181.09 228.48 ± 177.80 0.237
Volume at first sensation of bladder filling (mL) 168.52 ± 106.38 172.05 ± 107.56 168.31 ± 109.12 154.73 ± 95.18 0.108
Volume at strong desire to void (mL) 229.25 ± 110.10 231.69 ± 107.70 229.07 ± 110.01 220.78 ± 118.10 0.513
Pdetmax during filling stage (cm H2O) 37.20 ± 23.95 37.75 ± 23.93 36.56 ± 26.05 36.20 ± 20.35 0.399
DLPP or end-filling pressure (cm H2O) 30.77 ± 24.46 31.42 ± 25.03 30.01 ± 25.26 29.61 ± 20.83 0.295
Maximum urethral closure pressure (cm H2O) 79.26 ± 45.36 79.11 ± 39.60 78.46 ± 58.65 81.22 ± 40.40 0.791
Upper urinary tract imaging outcome
 VUR 470 (19.64) 256 (18.29) 123 (20.47) 91 (23.21) 0.080
 Hydronephrosis 581 (24.28) 338 (24.14) 139 (23.13) 104 (26.53) 0.466
 UUTD 789 (32.97) 450 (32.14) 201 (33.44) 138 (35.20) 0.502

Values are presented as mean±standard deviation or number (%).

NB, neurogenic bladder; Pdetmax, maximum detrusor pressure; DLPP, detrusor leak point pressure; VUR, vesicoureteral reflux; UUTD, upper urinary tract deterioration.

Table 2.

Patient characteristics stratified based on upper urinary tract deterioration outcome

Variable UUTD (+) (n = 789) UUTD (-) (n = 1604) P-value
Age (yr) 31.04 ± 16.84 36.90 ± 17.59 < 0.001
Sex 0.605
 Male 512 (64.89) 1,058 (65.96)
 Female 277 (35.11) 546 (34.04)
Disease duration (yr) 12.64 ± 11.71 5.14 ± 9.50 < 0.001
NB etiology < 0.001
 Spinal cord injury 205 (25.98) 882 (54.99)
 Spinal dysplasia 395 (50.06) 356 (22.20)
 Others 189 (23.95) 366 (22.82)
Bladder management methods < 0.001
 Clean intermittent catheterization 69 (8.75) 462 (28.80)
 Indwelling catheterization 222 (28.14) 526 (32.79)
 Self-voiding 70 (8.87) 382 (23.82)
 Abdominal pressure-assisted urination 381 (48.29) 137 (8.54)
 Others 47 (5.96) 97 (6.05)
Urodynamic diagnosis 0.486
 Detrusor overactivity 296 (37.52) 572 (35.66)
 Acontractile detrusor 408 (51.71) 871 (54.30)
 Detrusor underactivity 85 (10.78) 161 (10.04)
Bladder compliance (mL/cm H2O) 17.54 ± 13.26 24.02 ± 13.18 < 0.001
Detrusor sphincter dyssynergia 364 (46.13) 156 (9.73) < 0.001
Maximum cystometric capacity (mL) 265.54 ± 143.57 322.71 ± 129.55 < 0.001
Bladder sensation < 0.001
 Normal 314 (39.80) 572 (35.66)
 Increased 142 (18.00) 205 (12.78)
 Reduced 111 (14.07) 270 (16.83)
 Absent 222 (28.14) 557 (34.73)
Postvoid residual urine volume (mL) 202.17 ± 165.09 265.01 ± 193.62 < 0.001
Volume at first sensation of bladder filling (mL) 156.68 ± 101.96 174.76 ± 108.11 0.002
Volume at strong desire to void (mL) 204.47 ± 108.09 242.06 ± 108.94 < 0.001
Pdetmax during filling stage (cm H2O) 43.01 ± 18.22 34.34 ± 25.84 < 0.001
DLPP or end-filling pressure (cm H2O) 35.58 ± 19.69 28.40 ± 26.17 < 0.001
Maximum urethral closure pressure (cm H2O) 80.97 ± 55.47 78.24 ± 37.99 0.298

Values are presented as mean±standard deviation or number (%).

UUTD, upper urinary tract deterioration; NB, neurogenic bladder; Pdetmax, maximum detrusor pressure; DLPP, detrusor leak point pressure.

Table 3.

Comparison of multiple machine learning evaluation indexes in the training, validation, and test sets

Data set Models AUC (95% CI) Accuracy Sensitivity Specificity PPV NPV F1 score
Training set GNB 0.884 (0.867–0.901) 0.787 0.873 0.747 0.617 0.926 0.723
MLP 0.779 (0.754–0.804) 0.746 0.635 0.797 0.594 0.824 0.614
SVM 0.792 (0.770–0.815) 0.715 0.843 0.655 0.533 0.899 0.653
KNN 0.932 (0.921–0.943) 0.840 0.880 0.821 0.697 0.936 0.778
Validation set GNB 0.868 (0.829–0.907) 0.756 0.837 0.712 0.611 0.889 0.707
MLP 0.761 (0.730–0.825) 0.731 0.638 0.781 0.612 0.799 0.625
SVM 0.783 (0.737–0.829) 0.726 0.830 0.669 0.576 0.879 0.680
KNN 0.788 (0.740–0.835) 0.731 0.674 0.762 0.605 0.811 0.638
Test set GNB 0.878 (0.839–0.916) 0.797 0.877 0.759 0.629 0.930 0.733
MLP 0.760 (0.707–0.813) 0.713 0.746 0.698 0.535 0.855 0.623
SVM 0.739 (0.686–0.792) 0.685 0.807 0.629 0.503 0.875 0.620
KNN 0.791 (0.741–0.841) 0.719 0.728 0.714 0.542 0.850 0.622

AUC, area under the receiver operating characteristic curve; CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value; GNB, Gaussian Naive Bayes; MLP, multilayer perceptron; SVM, support vector machine; KNN, K-nearest neighbor.

Table 4.

Subgroup analysis results of the validation set according to imaging outcome, neurogenic bladder etiology, and age

Subgroup Models AUC (95% CI) Accuracy Sensitivity Specificity PPV NPV F1 score
Upper urinary tract imaging outcome
 VUR GNB 0.815 (0.770–0.859) 0.719 0.813 0.691 0.443 0.924 0.574
MLP 0.757 (0.700–0.813) 0.696 0.758 0.678 0.416 0.903 0.537
SVM 0.687 (0.624–0.750) 0.582 0.791 0.518 0.332 0.891 0.468
KNN 0.729 (0.668–0.789) 0.656 0.747 0.628 0.378 0.892 0.502
 Hydronephrosis GNB 0.832 (0.790–0.874) 0.742 0.817 0.715 0.509 0.916 0.627
MLP 0.742 (0.689–0.796) 0.699 0.731 0.688 0.458 0.876 0.563
SVM 0.692 (0.635–0.749) 0.615 0.817 0.542 0.392 0.891 0.530
KNN 0.746 (0.691–0.801) 0.673 0.750 0.646 0.433 0.877 0.549
 UUTD GNB 0.868 (0.829–0.907) 0.756 0.837 0.712 0.611 0.889 0.707
MLP 0.761 (0.730–0.825) 0.731 0.638 0.781 0.612 0.799 0.625
SVM 0.783 (0.737–0.829) 0.726 0.830 0.669 0.576 0.879 0.680
KNN 0.788 (0.740–0.835) 0.731 0.674 0.762 0.605 0.811 0.638
NB etiology
 Spinal cord injury GNB 0.851 (0.792–0.910) 0.723 0.843 0.693 0.410 0.946 0.551
MLP 0.812 (0.747–0.878) 0.826 0.627 0.876 0.561 0.903 0.593
SVM 0.796 (0.724–0.869) 0.755 0.765 0.752 0.438 0.927 0.557
KNN 0.784 (0.707–0.861) 0.810 0.627 0.856 0.525 0.901 0.571
 Spinal dysplasia GNB 0.855 (0.804–0.907) 0.776 0.850 0.685 0.765 0.792 0.805
MLP 0.670 (0.594–0.746) 0.638 0.822 0.416 0.629 0.661 0.713
SVM 0.672 (0.595–0.748) 0.617 0.860 0.326 0.605 0.659 0.710
KNN 0.781 (0.717–0.846) 0.724 0.813 0.618 0.719 0.733 0.763
 Others GNB 0.895 (0.842–0.949) 0.842 0.814 0.853 0.686 0.921 0.745
MLP 0.767 (0.681–0.853) 0.743 0.605 0.798 0.542 0.837 0.571
SVM 0.757 (0.677–0.837) 0.704 0.767 0.679 0.485 0.881 0.595
KNN 0.831 (0.758–0.904) 0.770 0.698 0.798 0.577 0.870 0.632
Age (yr)
 < 18 GNB 0.870 (0.796–0.944) 0.800 0.829 0.780 0.725 0.867 0.773
MLP 0.735 (0.626–0.844) 0.659 0.800 0.560 0.560 0.800 0.659
SVM 0.722 (0.614–0.829) 0.624 1.000 0.360 0.522 1.000 0.686
KNN 0.850 (0.771–0.930) 0.741 0.829 0.680 0.644 0.850 0.725
 ≥ 18 GNB 0.866 (0.833–0.900) 0.765 0.843 0.728 0.596 0.907 0.698
MLP 0.792 (0.750–0.833) 0.757 0.711 0.779 0.605 0.850 0.654
SVM 0.786 (0.746–0.827) 0.709 0.777 0.676 0.533 0.864 0.632
KNN 0.819 (0.778–0.859) 0.777 0.723 0.802 0.635 0.859 0.676

AUC, area under the receiver operating characteristic curve; CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value; VUR, vesicoureteral reflux; UUTD, upper urinary tract deterioration; NB, neurogenic bladder; GNB, Gaussian Naive Bayes; MLP, multilayer perceptron; SVM, support vector machine; KNN, K-nearest neighbor.