Prediction of Blood Glucose Level Oscillations in Type1 Diabetes by Using Elman Recurrent Neural Network

One of the most dangerous symptoms of Type 1 diabetes is the frequent and great oscillation of blood glucose level that can lead the patient to unconscious and coma states. So being able to predict and finally prevent these symptoms would help the diabetic patients. This paper attempts to use Elman neural networks to predict the blood glucose levels in type1 diabetic patients. Data set used in this paper consists of the protocol of 3 Iranian type1 Diabetic women and includes parameters such as type and dosage of injected insulin, the period of time (in hour) between two consecutive measurements of the blood glucose level, carbohydrate intake, exercise and the blood glucose level measured at start of the given period of time. Finally we concluded that the use of Recurrent Neural Network such as Elman with great reduction of prediction error and also reduction of number of layers and neurons used in the construction of Neural Networks as compared with other method can be an appropriate model to predict the long term blood glucose levels in type 1 diabetes.