Article ID Journal Published Year Pages File Type
558766 Biomedical Signal Processing and Control 2014 8 Pages PDF
Abstract

•We used the Glucosafe decision support system for glycemic control in intensive.•We used virtual patients to evaluate modifications to Glucosafe advice generation.•We evaluated what modifications were needed to lower blood glucose below 6.1 mmol/l.•Virtual patients are applicable in evaluating modifications to advice generation.•In some patients decreasing nutritional intake is necessary to lower mean blood glucose.

Intensive insulin therapy has previously shown reduced mortality with lowering blood glucose to between 4.4 and 6.1 mmol/l. However presumably due to fear of hypoglycemia the current recommended glycemic target is 7.8–10 mmol/l. This study evaluates the effect of modifications to the Glucosafe system on the glycemic outcomes of an in silico cohort and which modifications are necessary to lower mean blood glucose under 6.1 mmol/l without hypoglycemic incidents. Based on data from 12 real patients from a previous clinical trial, 12 virtual patients were constructed, the groups were compared and results of the modifications evaluated. Results indicate that virtual patients are applicable in evaluating modifications to advice generation, and that it is possible to lower mean blood glucose below 6.1 mmol/l, with no hypoglycemic incidents. In some patients increased insulin use did not achieve this and decreasing nutritional intake was necessary.

Related Topics
Physical Sciences and Engineering Computer Science Signal Processing
Authors
, , ,