Stratification of Patients with Diabetes Using Continuous Glucose Monitoring Profiles and Machine Learning

Background.Continuous glucose monitoring(CGM)offers an opportunity for patients with diabetes to modify their lifestyle tobetter manage their condition and for clinicians to provide personalized healthcare and lifestyle advice.However,analytic toolsare needed to standardize and analyze the rich data that emerge from CGM devices.This would allow glucotypes of patients tobe identified to aid clinical decision-making.Methods.In this paper,we develop an analysis pipeline for CGM data and applyit to 148 diabetic patients with a total of 8632 days of follow up.The pipeline projects CGM data to a lower-dimensional spaceof features representing centrality,spread,size,and duration of glycemic excursions and the circadian cycle.We then useprincipal components analysis and k-means to cluster patients’records into one of four glucotypes and analyze clustermembership using multinomial logistic regression.Results.Glucotypes differ in the degree of control,amount of time spent inrange,and on the presence and timing of hyper-and hypoglycemia.Patients on the program had statistically significantimprovements in their glucose levels.Conclusions.This pipeline provides a fast automatic function to label raw CGM datawithout manual input.