Soft computing approach for predictive blood glucose management using a fuzzy neural network

Challenges in the management of blood sugar for Type 1 Diabetes mellitus (T1DM) patients have emerged to be one of the major contributors for the increase in societal cost of the health care system. The objective of this work is to develop and assess a computerized model for predicting blood glucose in patients with T1DM using the insulin pump and continuous glucose sensor. This study draws upon a soft computing approach that tolerates imprecision, uncertainty, and partial truth. An adaptive network-based fuzzy inference system (ANFIS) was implemented in the framework of fuzzy inferences and adaptive networks using an artificial neural network. The goal for the predictive approach is to provide personalized aid to patients with T1DM to better manage their glucose levels.