Identification of Diabetes Mellitus Disease through Urine with Quartz Crystal Microbalance Sensor and Artificial Neural Network

Diabetes is a deadly disease in Indonesia. Early identification of diabetes is needed to treat this disease early. Identifying diabetes in urine with quartz crystal microbalance (QCM) sensors and artificial neural networks is a non-invasive diagnostic technique to detect diabetes mellitus, in order to provide information to people with diabetes. Currently, the common testing method used to detect diabetes is using laboratory blood tests and blood sugar monitors. However, this is often considered painful and uncomfortable for patients, especially if they do regular check-ups. In this study, the method used to classify diabetes patients uses a backpropagation artificial neural network method and 4 QCM sensors coated with carbon nanotubes (CNT) consisting of double walled, 2 multi-walled and graphene oxide. There are two classes, namely healthy and diabetes. The data sample uses patient urine. The evaluation results obtained the highest accuracy of 78%, namely with a hidden layer of 128, a learning rate of 0.2 and an epoch of 100, in the division of 80% data for training data and 20% for test data.