Advanced Insulin Bolus Advisor based on Run-To-Run Control and Case- Based Reasoning

Advanced Insulin Bolus Advisor based on Run-To-Run Control and Case- Based Reasoning 92 97 Journal of Biomedical and Health Informatics (JBHI)

Herrero, P. Pesl, M. Reddy, N. Oliver, P. Georgiou, and C. Toumazou

Advanced Insulin Bolus Advisor based on Run-To-Run Control and Case- Based Reasoning

Insulin bolus calculators are simple decision support software tools incorporated in most commercially available insulin pumps to help people with Type 1 diabetes mellitus (T1DM) calculate the insulin doses required to control their blood glucose levels. Although their clinical benefit has been demonstrated, their performance remains suboptimal, mainly because of their lack of flexibility and adaptability to patient’s physiological changes.

This paper presents an advanced decision support system for people with T1DM which provides personalised insulin bolus recommendations. The proposed system runs on a smartphone and keeps the simplicity of a bolus calculator, while improving its performance with more adaptability and flexibility. This is achieved through enhancing a bolus calculator with Case-Based Reasoning and Run-To-Run control. Periodic use of continuous glucose monitoring data is required in order to perform a retrospective optimization of the insulin bolus calculator parameters. The validity of the proposed approach has been proven by in silico studies using the FDA-accepted UVa-Padova Type 1 Diabetes Simulator (T1DMS). Tests under more realistic in silico scenarios were achieved by modifying the simulator to incorporate intra-subject variability on insulin sensitivity, capillary blood glucose error measurements, and uncertainty in carbohydrate estimation. Compared to currently available bolus calculators, the proposed technique has the potential to provide clear superior glycaemic control in T1DM management.

Read full article at IEEE Xplore.