Answers are grounded in this paper and cite their sources.
Current screening methods for diabetic kidney disease (DKD), characterized by albumin excretion in urine, are expensive or only identify patients in late disease stages. Hence, there is need for a cost‐effective, quick, and portable screening tool which identifies patients at DKD onset. Here we report that ultracentrifugation coupled with infrared spectroscopy and machine learning can identify and quantify low level microalbuminuria in urine samples from a cohort of diabetic patients (n=155) and controls (n=22). Independent testing of the methods indicated that classification analysis discriminated between normo‐ and micro/macroalbuminuric samples with sensitivity of >91 % and specificity of >99 %. Regression methods quantified albumin concentration in the samples with error values of 17 and 44 mg/L for normo‐ and microalbuminuric patients. Using only 700 μL of sample, this approach identifies patients at an earlier stage of disease than a urinary dipstick, whilst also yielding results cheaper and faster than the albumin to creatinine ratio.
No takes yet. Share an insight, caveat, or question.
Richardson et al. (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: