Diabetes data analysis and interpretation method

TL;DR: In this article, an automated diabetes data interpretation method is provided which combines symbolic and numeric computing approaches in order to identify and highlight key clinical findings in the patient's self-recorded diabetes data.

Abstract: An automated diabetes data interpretation method is provided which combines symbolic and numeric computing approaches in order to identify and highlight key clinical findings in the patient's self-recorded diabetes data. The patient data, including blood glucose levels and insulin dosage levels, recorded by a diabetic patient over a period of time by means of a glucose meter or the like, is initially downloaded into a central processing system such as a personal computer. The accepted diabetes data is subsequently processed to (a) identify insulin dosage regimens corresponding to predefined significant changes in insulin dosage which are found to be sustained for at least a predefined segment of the overall data collection period, (b) identify statistically significant changes in blood glucose levels resulting across adjacent ones of the identified insulin regimen periods, and (c) identify clinically significant changes in blood glucose levels from within the identified statistically significant glucose level changes. The results of the diabetes data processing are generated in the form of a comprehensive yet easily understandable data interpretation report highlighting the processing results, including details pertaining to the identified insulin regimens and the associated clinically significant changes in glucose levels.