Machine‐learning–based early prediction of end‐stage renal disease in patients with diabetic kidney disease using clinical trials data

Abstract

Aim: To predict end-stage renal disease (ESRD) in patients with type 2 diabetes by using machine-learning models with multiple baseline demographic and clinical characteristics. Materials and methods: In total, 11 789 patients with type 2 diabetes and nephropathy from three clinical trials, RENAAL (n = 1513), IDNT (n = 1715) and ALTITUDE (n = 8561), were used in this study. Eighteen baseline demographic and clinical characteristics were used as predictors to train machine-learning models to predict ESRD (doubling of serum creatinine and/or ESRD). We used the area under the receiver operator curve (AUC) to assess the prediction performance of models and compared this with traditional Cox proportional hazard regression and kidney failure risk equation models. Results: The feed forward neural network model predicted ESRD with an AUC of 0.82 (0.76-0.87), 0.81 (0.75-0.86) and 0.84 (0.79-0.90) in the RENAAL, IDNT and ALTITUDE trials, respectively. The feed forward neural network model selected urinary albumin to creatinine ratio, serum albumin, uric acid and serum creatinine as important predictors and obtained a state-of-the-art performance for predicting long-term ESRD. Conclusions: Despite large inter-patient variability, non-linear machine-learning models can be used to predict long-term ESRD in patients with type 2 diabetes and nephropathy using baseline demographic and clinical characteristics. The proposed method has the potential to create accurate and multiple outcome prediction automated models to identify high-risk patients who could benefit from therapy in clinical practice.

Perspectives on a Way Forward to Implementation of Precision Medicine in Patients With Diabetic Kidney Disease; Results of a Stakeholder Consensus-Building Meeting

“…It was important for HCPs to identify the non-responders to standard of care. This calls for education and HCP collaboration and, possibly, using tools, such as machine learning to identify patients that may benefit from therapy ( Belur Nagaraj et al, 2020 ). Eventually, a precision medicine-to-standard of care comparison is desired for all aspects and endpoints.…”

Relevance of the Pyroptosis-Related Inflammasome Pathway in the Pathogenesis of Diabetic Kidney Disease

“…Persistent aseptic inflammatory reactions in the kidney tissue are the pathophysiological basis of diabetic nephropathy (DN) that lead to glomerular capillary damage. The clinical features of DN are a gradual decline in renal function, abnormal levels of albumin (microalbuminuria) in the urine (30 mg/day or 20 g/min), and subsequent proteinuria and end-stage renal disease (ESRD) ( 82 , 83 ). Once ESRD develops, the mortality rate is high, representing a critical clinical issue ( 84 , 85 ).…”

Application of multi-label classification models for the diagnosis of diabetic complications

“…Machine learning has great advantages when dealing with massive data with both high dimensional attributes and tremendous number of instances, which has been widely applied in disease prediction [ 7 ]. The binary classification model is typically adopted by most approaches [ 8 – 14 ] and shows promising results for the prediction of diabetic complications. However, each diabetic complication was modeled and predicted independently in these studies, making it impossible to leverage the potential correlations among diabetes complications.…”

“…While the machine learning models [ 8 – 14 ]are widely used in predictions of diabetic complications, a more suitable method, multi-label learning, has been used rarely. Single-label methods predict diabetes complications separately based on whether or not one complication occurs.…”