Machine learning approach to predicting albuminuria in persons with type 2 diabetes: An analysis of the LOOK AHEAD Cohort

Abstract

Albuminuria and estimated glomerular filtration rate (e‐GFR) are early markers of renal disease and cardiovascular outcomes in persons with diabetes. Although body composition has been shown to predict systolic blood pressure, its application in predicting albuminuria is unknown. In this study, we have used machine learning methods to assess the risk of albuminuria in persons with diabetes using body composition and other determinants of metabolic health. This study is a comparative analysis of the different methods to predict albuminuria in persons with diabetes mellitus who are older than 40 years of age, using the LOOK AHEAD study cohort‐baseline characteristics. Age, different metrics of body composition, duration of diabetes, hemoglobin A1c, serum creatinine, serum triglycerides, serum cholesterol, serum HDL, serum LDL, maximum exercise capacity, systolic blood pressure, diastolic blood pressure, and the ankle‐brachial index are used as predictors of albuminuria. We used Area under the curve (AUC) as a metric to compare the classification results of different algorithms, and we show that AUC for the different models are as follows: Random forest classifier‐0.65, gradient boost classifier‐0.61, logistic regression‐0.66, support vector classifier ‐0.61, multilayer perceptron ‐0.67, and stacking classifier‐0.62. We used the Random forest model to show that the duration of diabetes, A1C, serum triglycerides, SBP, Maximum exercise Capacity, serum creatinine, subtotal lean mass, DBP, and subtotal fat mass are important features for the classification of albuminuria. In summary, when applied to metabolic imaging (using DXA), machine learning techniques offer unique insights into the risk factors that determine the development of albuminuria in diabetes.

Identifying Distinct Risk Thresholds of Glycated Hemoglobin and Systolic Blood Pressure for Rapid Albuminuria Progression in Type 2 Diabetes From NHANES (1999–2018)

“…These, too, were in keeping with previous findings. Duration of diabetes was an unmodifiable risk factor of ACR in patients with T2DM ( 41 ) while a high protein diet can exacerbate hypertension and expedite glomerular damage ( 42 ). Additionally, our results showed central obesity and a higher Na/K intake ratio could impose an extra burden on the kidney in patients with T2DM who had HbA1c ≥ 6.4%.…”

The automatic detection of diabetic kidney disease from retinal vascular parameters combined with clinical variables using artificial intelligence in type-2 diabetes patients

“…To train models to predict DKD, 8 demographic and clinical risk factors linked to diabetic kidney disease were used as predictors, based on previous studies [ 2 , 7 , 10 ], and they were extracted from the hospital’s electronic medical records system, including: age, gender, duration of diabetes, hypertension, history of cardiovascular and cerebrovascular disease, smoking, BMI (kg/m 2 ), and glycosylated hemoglobin (%). Hypertension was defined as systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg, self-reported physician-diagnosed hypertension or use of blood pressure-lowering medications.…”

“…Early detection of diabetic kidney disease will allow appropriate interventions and thus substantially reduce the health-care burden. Detection of diabetic kidney disease depends on measurement of albuminuria or estimated glomerular filtration rate (eGFR) [ 2 ], but increasing evidence has shown that some DKD patients will not develop albuminuria [ 3 ]. Moreover, the measurements of albuminuria or eGFR are not always convenient since serum or urine samples need to be obtained and they have limited precision at an earlier DKD stage [ 2 ].…”

“…Detection of diabetic kidney disease depends on measurement of albuminuria or estimated glomerular filtration rate (eGFR) [ 2 ], but increasing evidence has shown that some DKD patients will not develop albuminuria [ 3 ]. Moreover, the measurements of albuminuria or eGFR are not always convenient since serum or urine samples need to be obtained and they have limited precision at an earlier DKD stage [ 2 ]. Although promising new markers have been reported, none have presented better performance as a screening tool for DKD than albuminuria [ 4 ].…”

Artificial Intelligence for Predicting and Diagnosing Complications of Diabetes

“…AI has been applied to ten populations of patients with type 2 diabetes to predict development of DN using various combinations of demographics, vital signs, and laboratory tests. [83][84][85][86][87][88][89][90][91][92] The predictions were compared with modified databases in a sensitivity analysis comparison in two of these reports 84,87 and outperformed algorithms derived exclusively from clinical data in four of these reports. 83,85,86,91 In addition to standard data extraction, one study used natural language processing to identify data from the EHR.…”