Machine Learning Model for Histological Prostatitis Prediction in BPH Patients

723 patient records from a single-center study between June 2020 and June 2023 were used to develop a machine learning model for predicting histological prostatic inflammation. The dataset, created by Chunyang Meng and shared under a CC-BY-4.0 license, underpins a model achieving an AUC of 0.852. It includes features like prostate volume, neutrophil-to-lymphocyte ratio, and International Prostate Symptom Score.

Use Cases

• Developing risk stratification tools for histological prostatitis based on clinical variables like prostate volume and IPSS score.
• Benchmarking machine learning models for clinical prediction using features such as neutrophil-to-lymphocyte ratio and systemic immune–inflammation index.
• Creating interpretable clinical decision support systems using SHAP analysis on predictors like age and acute urinary retention.

Strengths

• Dataset includes 723 patient records with a clear outcome measure (53.5% had histological prostatitis).
• Model development involved internal validation and six machine learning algorithms, with the best model achieving an AUC of 0.852.
• Key predictors are explicitly identified, including prostate volume, NLR, IPSS, age, SII, and acute urinary retention.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data is from a single-center study, which may limit generalizability.

Provenance

Source

figshare

Collection Method

Single-center retrospective study of patients admitted to a urology department.

Time Range

June 2020 to June 2023

Freshness

Last updated 2026-04-21 04:22:24; freshness should be verified.