A RECURRENT NEURAL NETWORK AND A SYSTEM TO PREDICT BLOOD GLUCOSE LEVEL

摘要

A circulatory neural network (RNN) is provided that predicts blood sugar levels. The cyclic neural network (RNN) includes a first long short-term memory (LSTM) network including inputs and outputs for receiving near-infrared (NIR) radiation data, an input for receiving an output of the first LSTM network, and And a second LSTM network including an output for outputting a blood glucose level based on near-infrared radiation data input to the first LSTM network.