Non-Invasive Prediction of Detrusor Underactivity in BPH Patients

A retrospective cohort of 538 urodynamically evaluated patients with benign prostatic hyperplasia (BPH). The dataset was used to develop an interpretable machine learning framework for predicting detrusor underactivity (DU). It was authored by Long Gao and last updated in May 2026.

Use Cases

• Training predictive models for detrusor underactivity based on clinical, anatomical, and uroflowmetry parameters.
• Benchmarking machine learning algorithms (e.g., XGBoost) against traditional logistic regression for clinical risk stratification.
• Applying SHAP analysis for global and local interpretability of model predictions in a medical context.

Strengths

• Dataset size is explicitly stated as 538 patient records.
• Model development involved a rigorous feature selection pipeline, distilling 15 parameters to a 5-feature subset.
• The resulting XGBoost model achieved a reported AUC of 0.958 in internal validation.

Limitations

• Row count for the underlying data is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 17.4 KB, indicating limited scope.

Provenance

Source

figshare, author Long Gao.

Collection Method

Retrospective cohort study of urodynamically evaluated BPH patients.

Freshness

Last updated 2026-05-01 04:16:07; freshness should be verified.