Adaptive subspace-based prediction of T1DM glycemia

Cescon, Marzia; Renard, Éric

doi:10.1109/cdc.2011.6161154

Cited by 6 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This section presents the ARX model proposed in this paper. Differently to the black box models found in the literature (for instances [28,30,26,24]), the proposed ARX model uses EE, IOB, and COB as inputs to improve IG prediction. These inputs allow to consider 1) the intensity and duration of a PA, 2) the delivered insulin which is modulated (before and during PA) to reduce the risk of hypoglycemia, and 3) the CHO, usually ingested before and during PA to prevent hypoglycemia.…”

Section: The Proposed Arx Modelmentioning

confidence: 99%

See 1 more Smart Citation

ARX model for interstitial glucose prediction during and after physical activities

Ugalde

Garnotel

Doron

et al. 2019

Control Engineering Practice

View full text Add to dashboard Cite

This paper presents the first autoregressive with exogenous input (ARX) model using energy expenditure, carbohydrates on board, and insulin on board as input to predict interstitial glucose (IG). The proposed model may be used for predicting IG even during physical activity (PA). A populationbased model, obtained from a first database composed of 14 type 1 diabetes (T1D) patients, achieved a root-mean-square error (RMSE) of 16.7 ± 15.6 mg/dL, on IG prediction (30-min ahead) at the end of a PA, on a second database (15 T1D patients). Patient-specific ARX models, obtained on the second database, improved prediction accuracy (RMSE = 7.8 ± 4.5 mg/dL), outperforming the results found in the literature.

show abstract

Section: The Proposed Arx Modelmentioning

confidence: 99%

“…System identification is an alternative solution already used for considering the effect of PA in IG prediction. For instance, in [24], a subspace-based patient-specific model is proposed for IG prediction on T1D patients during 30 min of exercise. The model receives CHO, insulin, HR, and respiration rate as inputs.…”

Section: Introductionmentioning

confidence: 99%

ARX model for interstitial glucose prediction during and after physical activities

Ugalde

Garnotel

Doron

et al. 2019

Control Engineering Practice

View full text Add to dashboard Cite

show abstract

“…Numerous approaches have been taken in the diabetes technology literature to propose glucose predictive models, the majority being based on polynomial and state-space models. [3][4][5][6][7][8][9][10] Thanks to the increased availability of real-life data collected for long-period trials, machine learning (ML) techniques have become increasingly popular and have been successfully employed to solve the BG prediction problem. In particular, early prior studies exploited random forest, multivariate adaptive regression splines, 11 k-nearest neighbor, 12 decision tree, 13 gradient-boosted regression tree, 14 and support vector regression (SVR).…”

Section: Introductionmentioning

confidence: 99%

Long-Term Prediction of Blood Glucose Levels in Type 1 Diabetes Using a CNN-LSTM-Based Deep Neural Network

Jaloli

Cescon

2022

J Diabetes Sci Technol

View full text Add to dashboard Cite

Background: In this work, we leverage state-of-the-art deep learning–based algorithms for blood glucose (BG) forecasting in people with type 1 diabetes. Methods: We propose stacks of convolutional neural network and long short-term memory units to predict BG level for 30-, 60-, and 90-minute prediction horizon (PH), given historical glucose measurements, meal information, and insulin intakes. The evaluation was performed on two data sets, Replace-BG and DIAdvisor, representative of free-living conditions and in-hospital setting, respectively. Results: For 90-minute PH, our model obtained mean absolute error of 17.30 ± 2.07 and 18.23 ± 2.97 mg/dL, root mean square error of 23.45 ± 3.18 and 25.12 ± 4.65 mg/dL, coefficient of determination of 84.13 ± 4.22% and 82.34 ± 4.54%, and in terms of the continuous glucose-error grid analysis 94.71 ± 3.89% and 91.71 ± 4.32% accurate predictions, 1.81 ± 1.06% and 2.51 ± 0.86% benign errors, and 3.47 ± 1.12% and 5.78 ± 1.72% erroneous predictions, for Replace-BG and DIAdvisor data sets, respectively. Conclusion: Our investigation demonstrated that our method achieved superior glucose forecasting compared with existing approaches in the literature, and thanks to its generalizability showed potential for real-life applications.

show abstract

“…These predictors are based on different methods and data; purely empirical [14], derived from physiological models [9] or a combination thereof [7], [10], and designed for, and validated on different usage scenarios. Selecting among the different predictors a priori is a challenging task, and considering the complex dynamics of plasma glucose metabolism, there is good reason to believe that different predictors may be experts in specific conditions, and that no single model will be able to fully capture the dynamics alone.…”

Section: Introductionmentioning

confidence: 99%

Bayesian combination of multiple plasma glucose predictors

Ståhl

Johansson

Renard

2012

2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society

View full text Add to dashboard Cite

Abstract-This paper presents a novel on-line approach of merging multiple different predictors of plasma glucose into a single optimized prediction. Various different predictors are merged by recursive weighting into a single prediction using regularized optimization. The approach is evaluated on 12 data sets of type I diabetes data, using three parallel predictors. The performance of the combined prediction is better, or in par, with the best predictor for each evaluated data set. The results suggest that the outlined method could be a suitable way to improve prediction performance when using multiple predictors, or as a means to reduce the risk associated with definite a priori model selection.

show abstract

Adaptive subspace-based prediction of T1DM glycemia

Cited by 6 publications

References 24 publications

ARX model for interstitial glucose prediction during and after physical activities

ARX model for interstitial glucose prediction during and after physical activities

Long-Term Prediction of Blood Glucose Levels in Type 1 Diabetes Using a CNN-LSTM-Based Deep Neural Network

Bayesian combination of multiple plasma glucose predictors

Contact Info

Product

Resources

About