
Predicting the time to renal replacement therapy (RRT) is crucial for managing end-stage kidney disease. The study developed and validated a machine learning model for this purpose, comparing its accuracy with conventional methods using estimated glomerular filtration rate (eGFR) decline rates. Data from 135 adult CKD patients undergoing hemodialysis were analyzed. The machine learning model exhibited moderate accuracy (R2 = 0.60), surpassing conventional methods (R2 = -17.1). This study highlights the potential of machine learning in predicting RRT, offering new avenues for CKD treatment.
Like
Save
Share