| Summary: | Diabetes mellitus is one of the most common chronic diseases in the world. There are 2 types of diabetes. In Type 1 diabetes, the body does not produce sufficient insulin, leading to an increase of blood glucose levels. In Type 2 diabetes, the body does not use the insulin produced appropriately. For both situations, in worst case scenarios, it can lead to death from long term hyperglycemia and also from hypoglycemia due to over dosages of insulin administered to the diabetic patient.
The search for a closed-loop system becomes an important area of research as medical teams try to improve on the current continuous-glucose measurement cum insulin delivery devices. These devices provide continuous glucose measurement but manual administration of insulin dosage, which is a prevalent source of the error in insulin dosage. This is because every individual has unique glucose fluctuations and reactions towards different parameters, making manual dosages very risky and inaccurate.
This project aims to find a system to provide this closed-loop controlling solution through experimental studies using the Adaptive Neuro-Fuzzy inference System (ANFIS) and Dynamic Evolving Neuro-Fuzzy Inference System (DENFIS). Experiments have shown the validity of using these systems as the controller and results have shown to be promising as both systems capable of predicting the glucose levels at different conditions and parameters.
|
|---|