Machine Learning Prediction of Progression to Dialysis in Patients With Polycystic Kidney Disease: Population-Based Retrospective Cohort Study - PubMed
3 hours ago
- #Machine Learning
- #ADPKD
- #Dialysis Risk Prediction
- Study focuses on predicting dialysis risk in patients with Autosomal Dominant Polycystic Kidney Disease (ADPKD) using machine learning.
- Utilized Taiwan's National Health Insurance Research Database (2007-2018) with 1856 ADPKD patients, 16.27% progressed to dialysis.
- Key risk factors identified: age ≥66, anemia, congestive heart failure, and acute kidney injury.
- XGBoost model showed highest performance (accuracy 98.3%, AUC 0.955, F1-score 0.800, Brier score 0.022).
- Top predictors included age, comorbidities, anemia, cardiovascular markers, and medication use (e.g., anticoagulants, loop diuretics).
- ML models can aid in early identification of high-risk patients for closer monitoring and specialist evaluation.