Design of a Non-invasive IOT-Based system for prediction and early detection of type 2diabetes using Fuzzy logic

Abstract

Diabetes is a chronic, metabolic disease characterized by elevated levels of blood glucose, which leads over time to serious damage to the heart, blood vessels, eyes, kidneys, and nerves. The most common is Type 2 Diabetes Mellitus (T2DM), usually found in adults, which occurs when the body becomes resistant to insulin or doesn't make enough insulin. Mostly it is a consequence of poor diet and poor lifestyle but in some cases hereditary. Diabetes can be effectively managed or avoided if it is diagnosed early. With the current trends and great strides made in technology, a lot of approaches are proposing solutions that will prevent long-term fatal conditions associated with diabetes type-2 from happening. The available solutions are mostly invasive type of testing which are uncomfortable as they require drawing blood samples for testing and is costly also to carry out as it requires constant change of consumables like test strips and needles. In this research, we use photoplethysmography (PPG) signals to get sample for blood glucose testing in a non-invasive way from the patient’s fingertip. The use of fuzzy logic system in the health domain is very beneficial as it incorporates the knowledge and experience of the medical experts which is transformed into fuzzy sets and rules. The fuzzy logic system will take the information collected from the patients in form of datasets as inputs. It then applies the rules stored in the knowledge base, which is developed by using parameters and symptoms stated by the experts who are specialist doctors in this domain to provide the prediction and early detection rates of diabetes mellitus type 2 disease. With this solution, there will be increased diabetes type 2 disease awareness among the population and improved healthy living.