Psychological stress exacerbates lower urinary tract symptoms (LUTS), including urinary frequency and bladder hypersensitivity. This study aimed to determine whether chronic oxytocin (OXT) administration mitigates stress-induced bladder dysfunction and modulates central stress pathways. Female Sprague–Dawley rats were assigned to control, stress, or stress/OXT groups (n = 14 each) and exposed to repeated variable stress (RVS) for one week. Mechanical sensitivity, cystometric parameters, plasma catecholamines and cortisol, bladder histology, paraventricular nucleus (PVN) c-Fos expression, and gene expression in bladder and brain regions were evaluated. RVS induced mechanical hypersensitivity, shortened intercontraction intervals, increased bladder weight, and elevated plasma adrenaline, dopamine, and cortisol levels. Bladder epithelial thickening, submucosal nerve hyperplasia, and upregulation of EP3 and ENaC were also observed. Stress significantly increased PVN activation and upregulated ChAT, CRF, and TH expression in the brainstem. OXT treatment normalized intercontraction intervals, reduced catecholamine and cortisol levels, attenuated PVN c-Fos activation, and increased OXTR and GAD expression in the hypothalamus and thalamus. IL-6 expression was unchanged by stress but decreased by OXT. Chronic OXT administration was associated with improvement of stress-induced urinary frequency and was associated with modulation of central stress pathways, with limited evidence of direct peripheral anti-inflammatory effects.
Purpose:Upper urinary tract deterioration (UUTD) is a burdensome complication in neurogenic bladder (NB) patients. In this study, we aim to develop a machine learning-based practical tool for UUTD risk assessment. Methods:A total of 2393 NB patients receiving clinical and urodynamic examinations were included in the present study. 2001 patients from northern regions of our country were divided into training and validation sets, 392 patients from southern regions were allocated into the test set for internal-external validation. A series of clinical and urodynamic parameters were analyzed and screened through Boruta, Lasso regression, and decision tree algorithms; the features consistently identified by all three algorithms were used for machine learning model development. Model performance was assessed by the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis. Shapley Additive Interpretation analysis was used as a visual interpretation for individual patient. Results:A set of six feature variables, notably bladder management method and detrusor sphincter dyssynergia (DSD), were determined as the risk factors of UUTD. Among the four machine learning algorithms, the gaussian naive bayes (GNB) model showed the best overall performance, with an AUC (95% CI) of 0.884 (0.867-0.901) in the training set, 0.868 (0.829-0.907) in the validation set, and 0.878 (0.839-0.916) in the test set. Subgroup analyses according to imaging outcome, NB etiology, and age showed the GNB model had consistent performance, with AUC ranging from 0.815 to 0.895. The GNB model incorporating the core variables was presented as a web-based practical tool for individualized risk assessment. Conclusions:Our study developed a machine learning model for UUTD risk assessment in NB patients. These findings highlight the importance of combining bladder management method and urodynamic evaluation in UUTD risk assessment and provide a practical tool for individualized assessment. Independent external validation is warranted for the model.
Bladder disorders such as overactive bladder and neurogenic bladder impose a major symptom burden, yet neuromodulation therapies are largely delivered with fixed open-loop settings and limited adaptation to time-varying bladder states. Bladder digital twins are patient-anchored computational models linked to sensing streams and updated via state estimation and/or data assimilation. They offer a pathway to closed-loop personalized therapy, but bladder implementations remain fragmented across modeling, sensing, and stimulation. This review synthesizes digital twin technologies relevant to bladder physiology, sensing, and neuromodulation and translates lessons from mature organ digital twins to bladder requirements for closed-loop control. We conducted targeted searches of PubMed, Web of Science, IEEE Xplore, and Google Scholar (2013 to 2025). We reviewed digital twin paradigms for the heart, brain, lungs, liver, and kidney and summarized bladder approaches including finite-element bladder-wall mechanics, conductivity-based torso models for wearable bioimpedance optimization, strain-to-geometry reconstruction using stretchable sensors, and sensor-informed closed-loop control. Clinically deployable twins pair mechanistic structure with limited reliable data streams and use hybrid estimation for indirect, noisy measurements. In bladder applications, feasibility has been demonstrated for state estimation and model-guided design; key barriers include ambulatory motion artifacts and impedance drift, sparse clinical anchors for personalization, limited longitudinal human validation, and incomplete safety engineering for stimulation control. A roadmap toward closed-loop bladder digital twins should prioritize ambulatory-grade sensing with artifact handling and recalibration, uncertainty-aware real-time state estimation, model-informed stimulation optimization, and fail-safe control with standardized reporting of accuracy, latency, uptime, and drift.
Purpose This study aimed to evaluate the correlation between bladder outlet obstruction (BOO) and bladder trabeculation in patients with benign prostatic hyperplasia (BPH). Methods We analyzed data from consecutive BPH patients from July 2014 to June 2024, who underwent urodynamic study (UDS) and cystourethroscopy. The results of free uroflowmetry, filling cystometry, and pressure-flow study were analyzed to evaluate functional parameters. For anatomical parameters, bladder trabeculation grade, lateral lobe protrusion of prostate, and bladder neck elevation (BNE) measured in cystourethroscopy were used. BOO was defined as the BOO index of 40 or higher in UDS. Bladder trabeculation was graded using our previous studies. Results Among total of 1,452 BPH patients, 1,028 patients had trabeculation on cystoscopy. Age, serum prostate-specific antigen, postvoid residual, total prostate volume, transition zone volume, terminal type detrusor overactivity, detrusor pressure at the maximal flow rate, bladder contractility index (BCI), and BOO index increased according to increase in bladder trabeculation. Multivariable logistic analysis showed that bladder trabeculation was significantly associated with age (odds ratio [OR], 1.05; P<0.001), kissing sign (OR, 1.55; P=0.007), BNE (OR, 1.76; P<0.001), detrusor overactivity (OR, 1.88; P=0.002), BCI (OR, 1.01; P=0.037), BOO (OR, 2.27; P<0.001). BOO had the greatest correlation with bladder trabeculation. In addition, BOO index showed a positive correlation (r=0.39, P<0.001) with bladder trabeculation. BOO index well distinguished between moderate trabeculation of grade 2 or higher in receiver operating characteristic analysis (area under curve=0.72, P<0.001). Conclusions Our results showed that the severity of BOO is positively associated with the severity of bladder trabeculation.
Purpose Urodynamic studies (UDS) are the gold standard for evaluating pediatric lower urinary tract dysfunction (LUTD). However, previous studies have shown that the interpretation of cystometric findings may vary according to the observer’s level of experience and the criteria used for evaluation. This study aimed to investigate interobserver differences in the reporting of cystometric parameters in children with spina bifida. Methods Patients who underwent UDS between 2020 and 2025 for neurogenic LUTD secondary to spina bifida were retrospectively reviewed. Studies that did not reach 80% agreement for compliance with good urodynamic practice standards were excluded. Detrusor overactivity (DOA) and bladder compliance were independently evaluated by 4 pediatric urologists and 3 urologists experienced in neuro-urology. Agreement was assessed using Fleiss’ kappa test. Results A total of 100 children, including 53 females and 47 males, were included; the median age was 6.5 (range, 1–17) years. Agreement of at least 80% was reached in 86% of evaluations for DOA and 82% of evaluations for compliance. Based on the ratings of 7 observers, the Fleiss’ kappa value was 0.60 for DOA, indicating moderate agreement, and 0.46 for bladder compliance, indicating weak agreement. Conclusions This study demonstrated substantial interobserver variation in the interpretation of UDS, even among experienced centers and observers. Because UDS are essential for the diagnosis, treatment, and follow-up of patients with spina bifida, standardized criteria are needed to improve interobserver consistency in this patient group. Multicenter prospective studies involving observers with varying levels of experience are warranted to improve concordance in UDS reporting.
Purpose Overactive bladder (OAB) symptoms may improve following surgical correction of cystocele. This study aimed to evaluate changes in urgency, the primary symptom of OAB, and identify factors associated with improvement in urgency after surgical treatment for cystocele. Methods We conducted a retrospective analysis of medical records for patients who underwent surgical treatment for cystocele and had preoperative urgency measuring ≥3 on a 5-point urinary sensation scale. Patients were categorized into 2 groups based on their urgency status 3 months postsurgery: those with improved urgency and those without improvement. Improvement was defined as a reduction of 2 or more points on the scale following surgery. We compared preoperative clinical and urodynamic factors between the 2 groups. Results A total of 137 patients were included in the study, with 98 (71.5%) showing improvement in urgency symptoms after surgery. The improved group had a significantly higher prevalence of preoperative urgency urinary incontinence (UUI) (62.2% vs. 30.8%, P=0.001) and bladder outlet obstruction (43.9% vs. 20.5%, P=0.011) compared to the nonimproved group. Urodynamic evaluations indicated that the detrusor pressure at maximum flow rate (PdetQmax) was significantly higher in the improved group than in the nonimproved group (P=0.004). Multivariate logistic regression analysis identified preoperative UUI, higher PdetQmax, and lower vaginal parity as independent predictors of improvement in urgency postoperatively. Conclusions Preoperative UUI, higher PdetQmax, and lower vaginal parity were significant predictors of postoperative improvement in urgency. Both clinical and urodynamic factors may help identify patients with cystocele and preoperative urgency who are most likely to benefit from anatomical correction.
Purpose Postmicturition dribbling (PMD) is a common but underrecognized lower urinary tract symptom (LUTS). This study evaluated the clinical and urodynamic characteristics of PMD in men with medication-refractory LUTS. Methods From 2008 to 2010, 150 male patients underwent questionnaire assessment, urodynamic study (UDS), and transrectal ultrasound of the prostate (TRUSP). PMD was defined as a Likert score ≥3. Results PMD was reported by 47% of patients and was significantly associated with higher International Prostate Symptom Score (IPSS) scores, particularly LUTS such as urgency, incomplete emptying, and weak stream. PMD also adversely affected quality of life, especially motivation, sexual function, and psychological well-being. Among the UDS and TRUSP parameters, only poor bladder compliance was significantly associated with PMD (4% vs. 17%, P=0.016), corresponding to a 4-fold higher risk. Conclusions PMD was prevalent and clinically meaningful in men with LUTS and was strongly associated with core LUTS and reduced bladder compliance. Because PMD is not included in the IPSS, it may be underrecognized in routine clinical evaluation, underscoring the need for broader symptom assessment.
To review the time course of muscle mechanics, etiological factors, and molecular mechanisms contributing to the development of postvoid residual urine (PVR), with an emphasis on transcriptomic changes in detrusor underactivity secondary to bladder outlet obstruction (BOO). PVR, the volume of urine remaining in the bladder after micturition, is a key marker of lower urinary tract dysfunction. While often overlooked in routine assessment, it may signal underlying BOO or detrusor underactivity and is associated with serious complications. PVR, bladder physiology and mechanics, and BOO-associated molecular changes were integrated with mRNA expression data from patients with bladder outlet obstruction, as well as from short- and long-term obstructed rat bladders from our own laboratory. Efficient voiding depends on tightly coordinated neural and muscular activity; disruption by elevated outlet resistance, impaired detrusor contractility, or failed coordination can produce residual urine. Chronic BOO and detrusor underactivity are interlinked, with chronic obstruction initiating time-dependent compensatory mechanisms, but ultimately the bladder decompensates. Transcriptomic studies in humans reveal fibroblast-macrophage-epithelial crosstalk driving extracellular matrix remodeling, inflammation, and tissue repair, with conserved molecular pathways across species, including transforming growth factor beta, nuclear factor-kappa B, and phosphoinositide 3-kinase/protein kinase B signaling. Rat models highlight Cthrc1 as a regulator of smooth muscle proliferation after denervation and implicate hypoxia-responsive transcription factors. PVR arises from a multifactorial interplay of timedependent mechanical, myogenic, and neurogenic factors, underpinned by conserved molecular pathways. Molecular diagnostics integrating messenger RNA/microRNA signatures hold promise for early detection of detrusor decompensation, enabling targeted interventions to preserve bladder function and prevent complications.
Purpose To evaluate the effectiveness of virtual reality glasses (VRG) intervention during urodynamic testing (UT) on procedural pain (PP), anxiety levels, and hemodynamic parameters. Methods This prospective, randomized controlled trial enrolled 84 patients undergoing UT, randomized 1:1 to VRG (n=42) or control group (n=42). The VRG group viewed immersive 3-dimensional nature videos via VRG throughout the procedure, while controls received routine care. Primary outcomes included PP (visual analogue scale) and anxiety levels (State Anxiety Inventory) assessed at baseline (pre-UT1), immediately postprocedure (post-UT2), and 15 minutes postprocedure (post-UT3). Secondary outcomes comprised hemodynamic parameters (systolic blood pressure [SBP], diastolic blood pressure [DBP], heart rate [HR], respiratory rate [RR], oxygen saturation [SpO₂]) measured at identical timepoints. Results Groups demonstrated comparable baseline characteristics and PP scores (VRG: 2.52±2.76 vs. control: 3.07±2.99, P=0.386). While baseline anxiety levels were similar between groups, the VRG group exhibited significantly lower postprocedural anxiety scores (post-UT2: 34.64±4.98 vs. 39.21±8.11, P=0.003; post-UT3: 34.45±4.68 vs. 39.45±8.27, P=0.001). Hemodynamically, the VRG group demonstrated significantly reduced DBP at post-UT3 (73.4±9.7 mmHg vs. 78.7±12.0 mmHg, P=0.028) and lower RRs at both postprocedural timepoints (P<0.001 and P=0.015, respectively). No significant between-group differences were observed for SBP, HR, or SpO₂. Conclusions VRG intervention during UT effectively reduced postprocedural anxiety and demonstrated beneficial effects on RR and DBP, without significantly impacting PP. These findings support the use of VRG by nurses as a safe, nonpharmacological adjunct for enhancing patient comfort during invasive urological procedures.
Purpose To evaluate whether time-series analysis of detrusor pressure during the pressure-flow phase provides diagnostic information that complements conventional urodynamic parameters in the assessment of bladder outlet obstruction (BOO) in men with lower urinary tract dysfunction. Methods An observational, analytical, cross-sectional study was conducted including men ≥18 years undergoing urodynamic evaluation for voiding dysfunction. Pressure-flow studies were performed according to International Continence Society standards. Detrusor pressure data recorded during voiding were exported at one-second resolution and analyzed as timeseries restricted to the active flow phase. From these segments, descriptive statistical metrics were derived, including mean detrusor pressure during voiding (mPeDet). BOO was classified using conventional indices and final clinical judgment by an experienced urologist, which served as the reference standard. Diagnostic performance was assessed using receiver operating characteristic (ROC) analysis and Cohen kappa. Results Forty-one patients were included (median age, 59 years). mPeDet showed a strong correlation with the bladder outlet obstruction index (r=0.89, P<0.001). ROC analysis demonstrated excellent diagnostic performance for mPeDet in identifying BOO, with an area under the curve of 0.95. A cutoff value of 37 cm H₂O yielded a sensitivity of 88% and specificity of 87%. Agreement between mPeDet-based classification and final clinical diagnosis was moderate (κ=0.69). Conclusions Time-series-based evaluation of detrusor pressure during voiding provides additional clinically relevant information compared to conventional point-based urodynamic parameters. mPeDet may serve as a simple and reproducible complementary marker for the assessment of BOO, particularly in cases with diagnostic uncertainty.
Monitoring bladder filling is crucial in patients with neurogenic lower urinary tract dysfunction, particularly for safe and timely clean intermittent catheterization. Wearable ultrasound devices have recently gained attention as potential aids for continuous bladder monitoring, but they differ fundamentally from traditional bladder scanners. While portable scanners directly measure postvoid residual volume in milliliters, first-generation wearable devices using amplitude mode (A-mode) ultrasound primarily monitor bladder fullness trends and provide threshold-based alerts rather than precise quantitative volume measurements. This narrative review examines wearable ultrasound technologies in neurourology, searching PubMed, Embase, IEEE Xplore, and Google Scholar for relevant clinical studies. Most available data come from small observational studies in pediatric populations. In children with functional voiding dysfunction (FVD), devices such as SENS-U achieved sensitivities of 85%–90% for detecting bladder fullness thresholds with good correlation to urodynamic measurements. However, performance in neurogenic bladder populations is concerning: a recent study in children with neurogenic bladder reported only 46% accuracy and 13% specificity using DFree, indicating unacceptably high false-positive rates. No prospective validation studies have been conducted in adults with spinal cord injury, multiple sclerosis, or those at risk for autonomic dysreflexia. Regulatory classifications vary significantly among devices: SENS-U holds Conformité Européenne marking as a Class IIa medical device for pediatric use, while DFree is marketed as a consumer wellness product without medical device approval. EdgeFlow UW20, which employs a deep learning-based automatic bladder contour detection, recently received Ministry of Food and Drug Safety approval in Korea as a digital medical device. Currently, evidence supports considering wearable bladder monitors only for pediatric FVD under medical supervision. For neurogenic bladder populations, these devices are not recommended for routine clinical use until population-specific validation studies demonstrate adequate diagnostic accuracy. The emergence of B-mode (brightness mode) deep learning-based devices may address the fundamental limitations of first-generation A-mode technology.
Treatment of childhood enuresis requires an integrative approach grounded in a biopsychosocial perspective. Family factors play a critical role in symptom regulation, social development, and treatment adherence. This study aimed to reconceptualize childhood enuresis and introduce medical family therapy (MedFT) as an integrative framework to enhance treatment effectiveness. A systematic literature review was conducted using international databases (PubMed, Web of Science, Scopus, and Google Scholar) and domestic databases (RISS, KISS, SCIENCEON, DBpia, and KCI). Studies addressing biological, psychological, and family-related factors in childhood enuresis were included, whereas those focusing solely on pharmacological outcomes were excluded. The selected studies were analyzed using a thematic approach. Interventions integrating biological, psychological, and social components were more effective than single-modality treatments. Family factors significantly influenced treatment adherence, children’s social development and adaptation, and overall family functioning. Key factors included parental understanding of the condition, coping strategies, parenting attitudes, and intrafamily communication. Core MedFT components were identified as problem reconceptualization, multidimensional assessment, collaborative goal setting, facilitation of adaptive family-system changes, enhancement of problem-solving capacity, outcome evaluation, and relapse prevention. These findings suggest that MedFT, as a biopsychosocial integrative intervention, addresses both child-specific symptoms and family dynamics. It contributes to symptom reduction, improved psychological stability, and sustained improvements in family functioning, thereby enhancing clinical outcomes and overall family well-being. Further empirical research across diverse clinical contexts is needed to validate and refine this approach.
Purpose To evaluate the pilot clinical feasibility of an automatic urinary suction device in mobility-limited older adults, focusing on usability, functional performance, and perceived safety, and a secondary exploratory assessment of agreement between device-collected and reference urine specimens. Methods A 2-cohort clinical evaluation was conducted. The primary usability cohort assessed functional performance, perceived safety, and user satisfaction using structured 5-point Likert-scale questionnaires administered before and after device implementation. A secondary diagnostic cohort compared paired urine samples obtained using the automated suction device and standard-of-care reference urine collection methods to assess bacterial detection agreement, species concordance, and urine pH agreement. Paired statistical analyses with nonparametric sensitivity testing were used for usability outcomes, and diagnostic metrics were calculated with Wilson confidence intervals and Bland-Altman analysis. Results Seven participants completed paired usability assessments. Significant improvements were observed in operation and functionality (2.97±0.52 vs. 3.83±0.18, P=0.007) and ease of use (2.14±0.48 vs. 3.71±0.39, P=0.001), suggesting improved practicality of urinary management. Scores in the infection/skin complication domain decreased significantly. Product safety and overall satisfaction showed favorable but nonsignificant trends. In a secondary exploratory diagnostic analysis of 12 paired urine samples, bacterial detection agreement was acceptable at the presence/absence level (sensitivity 80.0%, specificity 100.0%, accuracy 91.7%), whereas species-level concordance was limited (40.0%), indicating restricted microbiologic reliability. Urine pH showed close numerical agreement. However, these findings should be interpreted cautiously, as key preanalytical variables were not controlled, and therefore should be regarded as exploratory rather than confirmatory. Conclusions This pilot study suggests that the automated urinary suction device may improve usability and functional handling in mobility-limited older adults. Skin-related findings remain preliminary, and the diagnostic applicability of device-collected urine should be interpreted cautiously pending further validation.
Purpose This study aimed to quantitatively evaluate the effect of simulated radiation dose reduction on deep learning-based renal segmentation performance and to identify a clinically acceptable minimum dose threshold. Methods Using the KiTS21 (2021 Kidney and Kidney Tumor Segmentation Challenge) dataset, which included 299 contrastenhanced computed tomography volumes with expert segmentation labels, 4 dose levels were simulated: 100%, 50%, 25%, and 10%. Dose reduction was simulated using Poisson noise modeling. A 2-dimensional U-Net with a ResNet34 encoder was trained exclusively on standard-dose images and evaluated across all dose levels using 5-fold cross-validation. Case-level performance was assessed using the Dice similarity coefficient (DSC), intersection over union, 95th-percentile Hausdorff distance (HD95), and volumetric error. Statistical significance was evaluated using the Wilcoxon signed-rank test with effect-size analysis. Results At the standard dose, the model achieved a case-level DSC of 0.948±0.044. Performance remained stable at 50% dose (0.945±0.046), declined moderately at 25% dose (0.939±0.052), and decreased more substantially at 10% dose (0.921±0.069). The Wilcoxon signed-rank test showed statistically significant differences between 100% dose and all reduced dose levels (P<0.001). HD95 increased from 4.73±4.82 pixels at 100% dose to 6.58±6.74 pixels at 10% dose. Conclusions Deep learning-based renal segmentation demonstrated substantial robustness to simulated dose reduction. Performance remained clinically acceptable, with a DSC>0.93, down to 25% of the standard dose, suggesting that substantial dose reduction may be feasible without critically compromising artificial intelligence–assisted renal segmentation. The marked performance decline at 10% dose identifies a potential lower bound for clinical dose optimization.
Artificial intelligence (AI) chatbots are transforming the delivery of urological care, evolving from simple digital assistants into emerging clinical partners that support patients and clinicians across the care continuum. This review traces this transformation by examining the development, clinical applications, and persistent challenges of AI-driven chatbots in urology. Across the patient journey, these systems are reshaping how urological care is accessed and experienced, from early symptom screening and patient education to lifestyle management, clinical decision support, and postoperative follow-up. Although these advances show considerable promise, important challenges remain regarding accuracy, data privacy, and empathic communication. Looking ahead, next-generation multimodal and on-device AI systems may further advance this transformation, positioning chatbots as increasingly important clinical partners in the delivery of high-quality, personalized, and patient-centered urological care.
Purpose Current therapies using midurethral sling mesh to treat stress urinary incontinence (SUI) are often considered insufficient. This study aims to evaluate the therapeutic efficacies of a minimally invasive implantable wireless electrical stimulator as an alternative treatment of SUI. Methods SUI was induced in 12 female, 12-week-old Sprague-Dawley rats with a deteriorated urethral sphincter complex by mechanical dilatation of the vagina. Six rats underwent sham operations. Four weeks after inducing SUI, 12 rats with SUI were divided into 2 groups. One group received a transplantation of an implantable wireless electrical stimulator, and the other midline incision and suturing. A total of 4 electrical stimulations were applied for 30 minutes twice weekly for 4 weeks in implantation model. Cystometric parameters including leak point pressure, urethral pressure and electromyography, histological examinations, immunostaining for Masson trichrome stain were measured at 4 weeks after intervention. Results After SUI induction, mean leak point pressure (LPP) was significantly lower in SUI groups than in controls (25.0±6.9 cm H2O vs. 62.6±4.1 cm H2O, P=0.012). At 4-week postintervention, the +ES group (SUI rat model with stimulation) demonstrated significantly higher LPP than the −ES group (48.8±8.4 cm H2O vs. 26.0±6.6 cm H2O, P=0.002). Urethral pressure showed a nonsignificant upward trend in the +ES group (SUI rat model without stimulation) compared with −ES (20.5±1.4 cm H2O vs. 18.1±1.1 cm H2O, P=0.070). Histology showed improved tissue organization with reduced inflammatory infiltration, and Masson staining demonstrated a trend toward higher collagen density in +ES (bladder: 28.3%±5.2% vs. 32.5%±7.6%, P=0.292; urethra: 44.2%±8.6% vs. 51.7%±8.2%, P=0.153). Conclusions These results indicate that an implantation of a wireless electrical stimulator can help to improve urethral sphincter function in SUI rat model. Moreover, the minimally invasive wireless electrical strategy may represent a promising therapeutic modality for pelvic floor muscle rehabilitation in the treatment of SUI with potential for clinical translation.
Artificial intelligence (AI) is being applied across healthcare, including disease diagnosis, personalized treatment, and rehabilitation management. In urology, AI research has evolved from traditional machine learning to advanced deep learning and generative models. This review categorizes these developments into 3 primary domains. First, machine learning has been applied to mobile-based platforms for continuous urinary health monitoring and personalized healthcare. Second, deep learning- based diagnostic systems have improved the identification of conditions such as prostatic hyperplasia, ureteral strictures, and urinary stones, thereby supporting clinical decision-making. Finally, generative AI, including large language models and vision transformers, is reshaping medical image analysis through data augmentation and expanding patient education through natural language interactions. This review examines integrated applications of AI in urological diseases and discusses current research trends and future prospects for a digital healthcare ecosystem in urology.
Purpose The maximum urinary flow rate (Qmax) is the most clinically relevant uroflowmetry parameter and is strongly associated with the International Prostate Symptom Score. However, traditional predictors such as age, prostate volume, and voided volume (VV) only partially explain the wide variability in Qmax among men with lower urinary tract symptoms (LUTS). Given the recognized role of male hypogonadism in LUTS progression and the potential of testicular volume as a surrogate marker of long-term androgenic status, this study investigated whether testicular volume could serve as a novel predictor of Qmax. Methods We retrospectively analyzed 170 men with LUTS evaluated between April 2012 and March 2017. All patients underwent uroflowmetry, transrectal ultrasonography, and testicular volume assessment using a Prader orchidometer. Serum prostate-specific antigen and testosterone levels were measured. Pearson correlation analysis was used to explore associations between clinical variables and Qmax, followed by multivariate linear regression to identify independent predictors. Results The mean age was 69.3 years, mean total prostate volume (TPV) 66.1 mL, and mean total testicular volume (TTV) 32.2 mL. In univariate analysis, age, height, TPV, TTV, VV, and residual urine (RU) were significantly correlated with Qmax. Multivariate regression demonstrated that TTV (β=0.268, P=0.001), VV (β=0.370, P<0.001), and RU (β=-0.193, P=0.009) were independent predictors, whereas prostate volume, age, and serum testosterone were not. Conclusions The TTV is a novel independent predictor of Qmax in men with LUTS. Smaller testicular volume was associated with a lower rate of urinary flow, potentially reflecting the cumulative endocrine decline associated with male hypogonadism. As a simple and noninvasive measurement, TTV can serve as a valuable clinical marker to complement existing assessments, helping to identify patients at higher risk for poor voiding function and guiding further comprehensive evaluations.
