An interpretable machine learning model for the prevention of contrast-induced nephropathy in patients undergoing lower extremity endovascular interventions for peripheral arterial disease

•An interpretable machine learning model was developed to stratify risk for post-procedural acute renal failure•The top predictive factors for acute renal failure were diabetes, claudication, and blood urea nitrogen (BUN)•The model demonstrates fairness for racial and age sub cohorts•The model can be used to identify patients at risk of developing CIN and optimize decision making regarding administration of iodinated contrast media