Stacked LSTM based deep recurrent neural network with kalman smoothing for blood glucose prediction

Abstract

Background: Blood glucose (BG) management is crucial for type-1 diabetes patients resulting in the necessity of reliable artificial pancreas or insulin infusion systems. In recent years, deep learning techniques have been utilized for a more accurate BG level prediction system. However, continuous glucose monitoring (CGM) readings are susceptible to sensor errors. As a result, inaccurate CGM readings would affect BG prediction and make it unreliable, even if the most optimal machine learning model is used.

Methods: In this work, we propose a novel approach to predicting blood glucose level with a stacked Long short-term memory (LSTM) based deep recurrent neural network (RNN) model considering sensor fault. We use the Kalman smoothing technique for the correction of the inaccurate CGM readings due to sensor error.

Results: For the OhioT1DM (2018) dataset, containing eight weeks' data from six different patients, we achieve an average RMSE of 6.45 and 17.24 mg/dl for 30 min and 60 min of prediction horizon (PH), respectively.

Conclusions: To the best of our knowledge, this is the leading average prediction accuracy for the ohioT1DM dataset. Different physiological information, e.g., Kalman smoothed CGM data, carbohydrates from the meal, bolus insulin, and cumulative step counts in a fixed time interval, are crafted to represent meaningful features used as input to the model. The goal of our approach is to lower the difference between the predicted CGM values and the fingerstick blood glucose readings-the ground truth. Our results indicate that the proposed approach is feasible for more reliable BG forecasting that might improve the performance of the artificial pancreas and insulin infusion system for T1D diabetes management.

Keywords: Blood glucose level prediction; Kalman smoothing; Recurrent neural network; Sensor fault correction; Stacked long short-term memory.

Fig. 1

Architecture of the proposed BG prediction system

Fig. 2

Kalman smoothed CGM values vs raw CGM readings for 12 h time window from the patient #563. Here red dots are raw CGM readings, where the blue line is the smoothed CGM values

Fig. 3

Two layered stacked LSTM network

Fig. 4

12-h of prediction results over PH = 30, for patient #570 (left) and #575 (right), are illustrated. Here red dots are CGM readings (unprocessed); those are ground truths. Where the blue line is the prediction curve, and the light blue region is the standard deviation

Fig. 5

12-h period prediction over PH=30, for patient #563 with the model trained with unprocessed CGM readings (Left) and Kalman-smoothed CGM readings for sensor fault correction (Right) respectively