Prediction of blood glucose concentration for type 1 diabetes based on echo state networks embedded with incremental learning

Li, Ning; Tuo, Jianyong; Wang, Youqing; Wang, Menghui

doi:10.1016/j.neucom.2019.10.003

Cited by 25 publications

(14 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The mean value of the RMSE of the model was 12.38 mg/dL based on data from 10 children and only used previous BG levels to estimate upcoming values. The authors in [ 24 ] tried to learn the chaotic properties in the glucose signal obtained from CGM systems using a model based on Echo State Network’s (ESNs) and achieved RMSE values of 13.57 mg/dL for a 30 min prediction horizon when implementing subject specific variants to the model. The authors in [ 25 ] proposed an approach based on Recurrent Neural Network’s (RNNs) trained in an end-to-end fashion, using the blood glucose signal, and were able to provide an estimate of the certainty in the predictions by training the recurrent neural network to parameterize a univariate Gaussian distribution over the output.…”

Section: Related Workmentioning

confidence: 99%

Deep Physiological Model for Blood Glucose Prediction in T1DM Patients

Muñoz-Organero

2020

Sensors

View full text Add to dashboard Cite

Accurate estimations for the near future levels of blood glucose are crucial for Type 1 Diabetes Mellitus (T1DM) patients in order to be able to react on time and avoid hypo and hyper-glycemic episodes. Accurate predictions for blood glucose are the base for control algorithms in glucose regulating systems such as the artificial pancreas. Numerous research studies have already been conducted in order to provide predictions for blood glucose levels with particularities in the input signals and underlying models used. These models can be categorized into two major families: those based on tuning glucose physiological-metabolic models and those based on learning glucose evolution patterns based on machine learning techniques. This paper reviews the state of the art in blood glucose predictions for T1DM patients and proposes, implements, validates and compares a new hybrid model that decomposes a deep machine learning model in order to mimic the metabolic behavior of physiological blood glucose methods. The differential equations for carbohydrate and insulin absorption in physiological models are modeled using a Recurrent Neural Network (RNN) implemented using Long Short-Term Memory (LSTM) cells. The results show Root Mean Square Error (RMSE) values under 5 mg/dL for simulated patients and under 10 mg/dL for real patients.

show abstract

Section: Related Workmentioning

confidence: 99%

Deep Physiological Model for Blood Glucose Prediction in T1DM Patients

Muñoz-Organero

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…We found 48 relevant publications 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 , 77 , 78 , 79 , 80 , 81 , 82 , 83 , 84 , 85 , 86 , 87 , 88 , 89 , 90 , 91 , 92 , 93 , 94 , 95 , 96 , 97 , 98 , 99 , 100 that presented a prediction algorithm published between 2013 and 2020 showing the recent increasing interest for this topic. Information on these algorithms is presented in appendices.…”

Section: Algorithms For Glucose and Hypoglycaemia Predictionmentioning

confidence: 99%

Hypoglycaemia detection and prediction techniques: A systematic review on the latest developments

Diouri

Cigler

Vettoretti

et al. 2021

Diabetes Metabolism Res

View full text Add to dashboard Cite

The main objective of diabetes control is to correct hyperglycaemia while avoiding hypoglycaemia, especially in insulin-treated patients. Fear of hypoglycaemia is a hurdle to effective correction of hyperglycaemia because it promotes under-dosing of insulin. Strategies to minimise hypoglycaemia include education and training for improved hypoglycaemia awareness and the development of technologies to allow their early detection and thus minimise their occurrence. Patients with impaired hypoglycaemia awareness would benefit the most from these technologies. The purpose of this systematic review is to review currently available or in-development technologies that support detection of hypoglycaemia or hypoglycaemia risk, and identify gaps in the research. Nanomaterial use in sensors is a promising strategy to increase the accuracy of continuous glucose monitoring devices for low glucose values. Hypoglycaemia is associated with changes on vital signs, so electrocardiogram and encephalogram could also be used to detect hypoglycaemia. Accuracy improvements through multivariable measures can make already marketed galvanic skin response devices a good noninvasive alternative. Breath volatile organic compounds can be detected by dogs and devices and alert patients at hypoglycaemia onset, while near-infrared spectroscopy can also be used as a hypoglycaemia alarms. Finally, one of the main directions of research are deep learning algorithms to analyse continuous glucose monitoring data and provide earlier and more accurate prediction of hypoglycaemia. Current developments for early identification of hypoglycaemia risk combine improvements of available 'needle-type' enzymatic glucose sensors and noninvasive alternatives. Patient usability will be essential to demonstrate to allow their implementation for daily use in diabetes management.

show abstract

“…The estimation of states and unknown inputs (UIs) is important in fault diagnosis and dynamic system control. [1][2][3][4][5] This problem is frequently encountered in power system exciters, [6,7] chemical processes, [8][9][10] state estimation of a battery, [11][12][13] navigation, [14,15] and earthquake damage estimation. [16] Over the last decade, many methods have been proposed for the simultaneous estimation of the UIs and states in a linear discrete-time system, [17][18][19][20][21][22] among which Gillijns and De Moor [22] present a valuable overview.…”

Section: Introductionmentioning

confidence: 99%

“…The estimation of states and unknown inputs (UIs) is important in fault diagnosis and dynamic system control. [ 1–5 ] This problem is frequently encountered in power system exciters, [ 6,7 ] chemical processes, [ 8–10 ] state estimation of a battery, [ 11–13 ] navigation, [ 14,15 ] and earthquake damage estimation. [ 16 ]…”

Section: Introductionmentioning

confidence: 99%

Distributed semi‐cooperative filter for a nonlinear multi‐agent system with heterogeneous and homologous unknown inputs

Liu

et al. 2020

Can J Chem Eng

Self Cite

View full text Add to dashboard Cite

In this study, the simultaneous estimation of the states and unknown inputs for a nonlinear multi-agent system with homologous and heterogeneous unknown inputs is performed. The decentralized sub-filter is used to estimate the states and heterogeneous unknown inputs, whereas the distributed subfilter is used to estimate the homologous unknown inputs. The extended Kalman filter is used to solve the estimation problem for nonlinear systems. Compared with previous studies, the distributed solution is improved to relax the existence of the homologous unknown input sub-filter. Moreover, the updating method of the residual generator is improved to relax the heterogeneous unknown input sub-filter. The practical problem of estimating the state of charge and temperature of the battery pack is used to verify the effectiveness of the proposed filter.

show abstract

Prediction of blood glucose concentration for type 1 diabetes based on echo state networks embedded with incremental learning

Cited by 25 publications

References 29 publications

Deep Physiological Model for Blood Glucose Prediction in T1DM Patients

Deep Physiological Model for Blood Glucose Prediction in T1DM Patients

Hypoglycaemia detection and prediction techniques: A systematic review on the latest developments

Distributed semi‐cooperative filter for a nonlinear multi‐agent system with heterogeneous and homologous unknown inputs

Contact Info

Product

Resources

About