Publication View

Automated Interpretation of Diabetes Patient Data: Detecting Temporal Changes in Insulin Therapy

Abstract
We are developing an expert system to interpret self-recorded, time-ordered diabetes patient data. One essential task in this program is the detection of significant changes in insulin therapy. This detection problem is difficult because patients usually make small adjustments in their insulin therapy based on the value of recent blood glucose measurements. Thus, we need to recognize when a systematic change in insulin therapy has occurred against a background of constantly changing insulin injections.

Publication details
Download http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2245549
Repository PubMed Central (PMC3 - NLM DTD) (United States)
Keywords Article
Type Text
Language English