Sex-Specific Digital Phenotyping of Lower Urinary Tract Symptoms: A Cross-Sectional Study of 1586 Adults From 35 Centers in China.

BACKGROUND AND PURPOSE: Lower urinary tract symptoms (LUTS) represent a complex clinical syndrome with significant sexual dimorphism. Traditional "one-size-fits-all" management often overlooks intra-sex phenotypic heterogeneity. This study utilized unsupervised machine learning to identify sex-specific digital phenotypes of LUTS within a large-scale clinical cohort to provide a foundation for precision urological care.

MATERIALS AND

METHODS

We analyzed a multicenter database of 1586 treatment-seeking adults (1098 females, 488 males) across 35 centers. Baseline clinical characteristics and LUTS parameters from standardized voiding diaries and validated scales were collected. K-means clustering was performed separately by sex to identify phenotypic subtypes, with the optimal number of clusters determined by the Elbow Method and visualized via Principal Component Analysis (PCA).

RESULTS

Significant sex-specific disparities were observed: women reported a higher symptomatic burden across almost all domains but maintained better voiding satisfaction (p < 0.001), while men exhibited poorer sleep quality (p < 0.001). Clustering identified a consistent three-subtype framework for both genders: Subtype 1 (Mild/Compensated): Characterized by the lowest symptom burden and highest tolerance. Subtype 2 (Storage-Dominant with Sleep Fragmentation): Found in older populations, defined by severe nocturia, reduced bladder capacity, and profound sleep disruption. Subtype 3 (Voiding-Dominant): Characterized by prominent dysuria and incomplete emptying.

CONCLUSIONS

This exploratory sex-stratified clustering study identified symptom-based phenotypic patterns among treatment-seeking Chinese adults with LUTS. These findings support the potential value of a subtype-plus-sex framework for LUTS assessment, but prospective validation using objective functional measures and treatment outcomes is needed.