Processing math: 100%
CGM-Based Blood Glucose Prediction Model With LSTM Encoder–Decoder Architecture | IEEE Journals & Magazine | IEEE Xplore

CGM-Based Blood Glucose Prediction Model With LSTM Encoder–Decoder Architecture


Abstract:

Accurate prediction of blood glucose levels is crucial for automated treatment in diabetic patients. This study proposes a blood glucose prediction model based on an impr...Show More

Abstract:

Accurate prediction of blood glucose levels is crucial for automated treatment in diabetic patients. This study proposes a blood glucose prediction model based on an improved attention mechanism within a long short-term memory (LSTM) encoder-decoder (Att-E-D) architecture to enhance blood glucose prediction performance significantly. Compared to traditional encoder-decoder (E-D) architectures, the core improvement of this study lies in introducing an attention mechanism combined with a dynamic time warping (DTW) similar sequence search algorithm. Specifically, during the decoding phase of the Att-E-D architecture, the DTW-based similar sequence search algorithm is first utilized to retrieve the n most matching similar sequences from historical blood glucose data for the target prediction sequence. Then, at each time step, different attention weights are assigned to the encoded information. This enables the model to selectively focus on historical feature information with higher similarity to the current step, effectively enhancing neuron memory and reducing error propagation. Finally, after processing through a fully connected layer, the model outputs a prediction sequence for blood glucose trends over a future period. To validate the generalization ability of the Att-E-D model, comparative experiments were conducted using datasets from two different continuous glucose monitoring (CGM) sensors. The results demonstrate that the Att-E-D model significantly outperforms support vector regression (SVR), LSTM, gated recurrent unit (GRU), bidirectional LSTM (Bi-LSTM), Bi-GRU, and the basic E-D model in prediction accuracy, achieving the highest {R}^{{2}} values of 0.952 and 0.972 on the two datasets, respectively, proving its superior ability to capture the long-term dependencies in blood glucose data.
Published in: IEEE Sensors Journal ( Volume: 25, Issue: 3, 01 February 2025)
Page(s): 5824 - 5839
Date of Publication: 20 December 2024

ISSN Information:

Funding Agency:

No metrics found for this document.

I. Introduction

The diabetes mellitus (DM), a global chronic metabolic disease, is characterized by hyperglycemia resulting from defects in insulin secretion or resistance [1]. This disease not only directly affects patients’ blood glucose management but also significantly increases the incidence of severe health issues such as cardiovascular disease, renal failure, disability, fatal risks, and heart failure, posing a severe threat to both personal health and the socio-economic landscape [2]. In recent years, the incidence of diabetes has been growing at an alarming rate, becoming a significant challenge for global public health. According to a 2021 report by the International Diabetes Federation (IDF), the number of people living with diabetes globally has soared from 285 million in 2009 to 537 million in 2021, and it is projected to reach 783 million by 2045, making it the third largest global public health threat after cardiovascular disease and cancer [3]. The prevention and control of diabetes and its complications are becoming increasingly urgent.

No metrics found for this document.

References

References is not available for this document.