ATR‐FTIR Spectroscopy for Early Detection of Diabetic Kidney Disease

Current screening methods for diabetic kidney disease (DKD), characterized by albumin excretion in urine, are expensive or only identify patients late stages. Hence, there is need a cost-effective, quick, and portable tool which identifies at DKD onset. Here we report that ultracentrifugation coupled with infrared spectroscopy machine learning can quantify low level microalbuminuria urine samples from cohort of (n=155) controls (n=22). Independent testing the indicated classification analysis discriminated between normo- micro/macroalbuminuric sensitivity >91 % specificity >99 %. Regression quantified concentration error values 17 44 mg/L microalbuminuric patients. Using 700 μL sample, this approach an earlier stage than urinary dipstick, whilst also yielding results cheaper faster to creatinine ratio.