International Neurourology Journal最新文献

Purpose: Stress urinary incontinence (SUI) annoyed women worldwide and surgery remain importance for those who failed to observative managements.

Methods: We retrospectively reviewed medical records of 533 female patients with mixed urinary incontinence and predominant SUI in a medical center. Some patients may have had stage 3 or higher cystocele and underwent concomitant anterior colporrhaphy. Patients were divided into four groups: pubovaginal sling (PVS) alone, PVS with colporrhaphy, transobturator suburethral sling (TOT) alone and TOT with colporrhaphy. The primary outcome was the long-term cumulative success rate in different groups and a successful outcome defined as dry or less than one pad usage per day. The secondary outcomes were subjective postoperative lower urinary tract symptoms and various perioperative complications.

Results: The long-term cumulative success rate of PVS group with or without colporrhaphy are significantly higher than those in TOT group with or without colporrhaphy (p< 0.001). The group of PVS with concurrent colporrhaphy obtained highest success rate, followed by the PVS alone, TOT with colporrhaphy and TOT alone (p=0.003). Furthermore, the highest rate of persistent overactive bladder was noted in TOT alone group (p< 0.001).

Conclusions: This study suggests PVS is superior to TOT in terms of incontinent symptom control and long-term success rate. Concurrent colporrhaphy may be also helpful for anti-incontinent effect.

Artificial intelligence (AI) has emerged as a transformative tool for advancing diagnosis, monitoring, and treatment planning in neurourology. This review synthesizes recent progress in AI-based models for predicting treatment failure in neurogenic lower urinary tract dysfunction. Machine learning and deep learning algorithms applied to urodynamic, clinical, and neuroimaging data have demonstrated strong potential to identify patients at risk of therapeutic nonresponse and improve individualized management. Automated systems now enable precise interpretation of complex bladder signals, multimodal data integration, and real-time prediction of treatment outcomes, marking a shift toward data-driven precision medicine. Nevertheless, most published studies remain limited by small, single-center datasets and a lack of external validation. Broader clinical adoption will require multicenter collaboration, adherence to standardized reporting frameworks such as TRIPOD-ML and PROBAST-AI, and integration of explainable AI to ensure transparency, reproducibility, and clinician trust.

Purpose: This study aimed to extract radiomics features from prostate magnetic resonance imaging (MRI), evaluate their reproducibility, and determine whether machine learning (ML) models built on reproducible features can noninvasively diagnose prostate cancer (PCa).

Methods: We analyzed prostate MRI from 82 subjects (41 PCa and 41 controls) in the public PROSTATEx dataset. From T2-weighted imaging (T2WI) and apparent diffusion coefficient (ADC) maps, 215 features per sequence were extracted (T2WI 215; ADC 215; total 430). Reproducibility within each sequence was quantified after repeated segmentation using the intraclass correlation coefficient (ICC) with a 2-way random-effects, absolute-agreement model. Only shared features with ICC≥0.75 in both T2WI and ADC were retained. Selected features were normalized and combined via early fusion into a single input vector. Redundant features were eliminated by Pearson correlation analysis (|r|>0.9).

Results: Reproducible radiomics features (ICC≥0.75) were key contributors to model performance. Using these features, support vector machine, neural network, and logistic regression models achieved accuracies of 80%-84% and a maximum area under the receiver operating characteristic curve of 0.85 under 5-fold cross-validation. Principal component analysis yielded the most consistent results, whereas several nonlinear dimensionality reduction methods produced variable outcomes across classifiers.

Conclusion: Combining reproducible MRI radiomics features with dimensionality reduction and ML offers a robust noninvasive approach for PCa diagnosis. Emphasizing reproducibility enhances model performance and reliability, supporting potential clinical translation.

Purpose: This study analyzes long-term changes in social perceptions and usage patterns of voiding diaries and identifies their structural characteristics. Although voiding diaries are essential instruments for evaluating lower urinary tract symptoms, their use is often limited by inconvenience and poor compliance. To clarify these issues, this research examines a decade of online community discussions to trace shifts in public perception.

Methods: Data were collected from Korean online communities between September 2015 and August 2025. Using text mining techniques, the dataset was extracted with keywords such as "urination diary," "urination log," "bladder diary," and "bladder log." To capture key terms, word associations, and temporal changes in discussion, the analysis incorporated time-series analysis, 2-gram analysis, and term frequency-inverse document frequency (TF-IDF) analysis.

Results: Mentions of voiding diaries increased markedly after 2019 and then stabilized, indicating sustained public interest. The 2-gram analysis revealed 4 categories related to recording behavior, clinical evaluation, daily life management, and disease terminology. TF-IDF analysis identified examination, treatment, recording, management, and bladder as central terms, highlighting the diary's dual relevance in both clinical and everyday contexts.

Conclusion: Voiding diaries function not only as clinical tools but also as social instruments for managing daily life. Increasing user persistence requires digital healthcare strategies such as automated reminders, data visualization, and gamification to strengthen continuity between clinical care and personal management.

Purpose: This paper proposes a technical system designed to accurately identify the stenotic site in patients with ureteral stricture and guide the surgeon. The objective of this technical solution is to improve surgical efficiency.

Methods: The proposed system applies the YOLOv5 algorithm, an artificial intelligence technology, to analyze real-time input images, detect the location of stenosis with high precision, and provide clinical support. The YOLOv5 algorithm was selected to enable rapid and accurate detection of the stenotic area.

Results: The system demonstrated high recognition accuracy, yielding an average final sensitivity of 0.95. Because sensitivity reflects the probability of a true positive result, this finding confirms that the proposed method offered precise guidance for identifying the stenotic site.

Conclusion: The proposed method, which utilizes the YOLOv5 algorithm, can support surgery for patients with ureteral stricture. This system assists surgical procedures by accurately detecting strictures through real-time image analysis. Future research aims to provide both conservative and flexible boundary estimates for determining the stricture location.

Purpose: Recent studies on urolithiasis detection using deep learning have demonstrated promising accuracy; however, most rely on large-scale labeled imaging datasets. In clinical practice, only limited and partially labeled computed tomography (CT) scans are typically available, restricting the generalizability of conventional supervised models. This study aimed to propose a data-efficient framework for accurate stone detection from a small CT dataset by integrating self-supervised learning (SSL) and transfer learning (TL).

Methods: A total of 100 abdominal CT scans were analyzed and labeled as stone present or normal by expert radiologists. To learn generalizable feature representations from limited data, a SimCLR-based SSL framework with a ResNet50 backbone was employed. During the SSL stage, the model learned from augmented image pairs without labels to maximize similarity between positive pairs and minimize similarity between negatives. The pretrained encoder was subsequently fine-tuned using labeled data in the TL stage, with the lower layers frozen and higher blocks optimized using a linear classifier. Model training was performed with 5-fold cross-validation, and performance was evaluated using accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC).

Results: The proposed SSL+TL model achieved the best performance (AUC, 0.95; F1-score, 0.91), significantly outperforming both the random initialization and TL-only models. These findings indicate that SSL pretraining effectively learns robust and transferable representations even with limited data.

Conclusion: The proposed framework demonstrates the feasibility of artificial intelligence-based urolithiasis detection in small-data clinical environments. Combining SSL and TL alleviates data scarcity and provides a foundation for developing generalizable and resource-efficient diagnostic models for urological imaging.

This review presents a privacy-by-design-based technical and governance framework for the safe clinical deployment of large language model (LLM) chatbots in urology. Given the high sensitivity of urological data involving urinary, sexual, and reproductive health, the proposed approach integrates on-site algorithmic deidentification, federated learning with differential privacy and secure aggregation, and secure retrieval-augmented generation with source citation and audit logging. Collectively, these components establish a federated, explainable, and auditable pipeline that preserves data sovereignty while improving clinical reliability and regulatory compliance. Urology thus serves as a critical test bed for validating the safety, governance, and accountability standards required for broader adoption of LLM-based medical chatbots across clinical domains.

Purpose: This study aimed to evaluate the performance of an artificial intelligence (AI)-based analysis of uroflowmetry (UFM) curve images, enhanced with customized preprocessing techniques, to improve diagnostic accuracy for bladder outlet obstruction (BOO) and detrusor underactivity (DUA).

Methods: We retrospectively analyzed 2,579 UFM curve images from patients who underwent urodynamic study (UDS), including 725 normal and 1,854 abnormal cases (736 BOO and 1,387 DUA). A VGG16 convolutional neural network model was developed to perform 3 binary classification tasks: normal versus abnormal, BOO versus non-BOO, and DUA versus non-DUA. To improve model performance, we implemented a preprocessing pipeline consisting of denoising, cropping, axis scaling, and color-coding of clinical parameters such as voided volume and postvoid residual volume (PVR). Model performance was evaluated using 5-fold stratified cross-validation and the area under the receiver operating characteristic curve (AUROC).

Results: Abnormal cases demonstrated a lower median maximum flow rate (8.9 mL/sec vs. 14.8 mL/sec), higher PVR (60.0 mL vs. 20.0 mL), and lower voiding efficiency (78.5% vs. 92.5%) than normal cases. Within the abnormal group, the BOO subgroup showed a higher PVR (80.0 mL) than the non-BOO subgroup (30.0 mL). After applying the preprocessing pipeline, model performance improved, with AUROC increasing from 0.807±0.024 to 0.827±0.016 for normal vs. abnormal classification, from 0.749±0.019 to 0.773±0.034 for BOO classification, and from 0.693±0.016 to 0.709±0.031 for DUA classification.

Conclusion: AI-based analysis of UFM curve images, enhanced through customized preprocessing, improved diagnostic accuracy in patients with lower urinary tract symptoms, effectively identifying BOO and DUA. This noninvasive method may serve as an adjunct or screening tool to reduce reliance on invasive UDS.

Purpose: Aquaporin-2 (AQP2) and vasopressin V1a receptor (AVP-V1a) are membrane proteins involved in water transport regulation in renal and extrarenal tissues. Their presence in the urinary bladder suggests a role in local water handling. This study aimed to investigate whether pelvic nerve stimulation induces translocation of AQP2 and AVP-V1a in the bladder urothelium of rats, thereby revealing a potential neural regulatory mechanism in water permeability.

Methods: Forty female Sprague-Dawley rats were assigned to either a control (n=20) or electrical stimulation group (n=20). The pelvic nerve was exposed and stimulated at 10 V, 16 Hz, 0.8 msec for 60 seconds. Bladder tissues were harvested immediately and processed for immunohistochemistry and Western blotting to evaluate the localization and expression levels of AQP2 and AVP-V1a in cytosolic and membrane protein fractions.

Results: Immunohistochemistry revealed that AQP2 and AVP-V1a were co-localized in the urothelium. Western blotting showed that pelvic nerve stimulation led to a significant decrease in cytosolic expression and a concurrent increase in membrane- associated expression of both AQP2 and AVP-V1a (P<0.05). These findings are indicative of protein translocation in response to neural stimulation.

Conclusion: Pelvic nerve stimulation may trigger the relocalization of AQP2 and AVP-V1a from the cytosol to the membrane in bladder urothelial cells. This suggests a novel neurophysiological mechanism for modulating bladder water transport, with potential implications for understanding bladder homeostasis and dysfunction suggesting a possible role in regulating bladder water permeability.