A Machine Learning Approach for Blood Glucose Level Prediction Using a LSTM Network

Nayeli Y. Gómez-Castillo; Pedro E. Cajilima-Cardenaz; Luis Zhinin-Vera; Belén Maldonado-Cuascota; Diana León Domínguez; Gabriela Pineda-Molina; Andrés A. Hidalgo-Parra; Fernando A. Gonzales-Zubiate

doi:10.1007/978-3-030-99170-8_8

A Machine Learning Approach for Blood Glucose Level Prediction Using a LSTM Network

Nayeli Y. Gómez-Castillo
, Pedro E. Cajilima-Cardenaz
, Luis Zhinin-Vera
, Belén Maldonado-Cuascota
, Diana León Domínguez
, Gabriela Pineda-Molina
, Andrés A. Hidalgo-Parra
, Fernando A. Gonzales-Zubiate

Universidad Yachay Tech
MIND Research Group - Model Intelligent Networks Development
Universidad de Cuenca

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

8 Scopus citations

Abstract

Diabetes is a chronic disease characterized by the elevation of glucose in blood resulting in multiple organ failure in the body. There are three types of diabetes: type 1, type 2, and gestational diabetes. Type 1 diabetes (T1D) is an autoimmune disease where insulin-producing cells are destroyed. World Health Organization latest reports indicate T1D prevalence is increasing worldwide with approximately one million new cases annually. Consequently, numerous models to predict blood glucose levels have been proposed, some of which are based on Recurrent Neural Networks (RNNs). The study presented here proposes the training of a machine learning model to predict future glucose levels with high precision using the OhioT1DM database and a Long Short-Term Memory (LSTM) network. Three variations of the dataset were used; the first one with original unprocessed data, another processed with linear interpolation, and a last one processed with a time series method. The datasets obtained were split into time prediction horizons (PH) of 5, 30, and 60 min and then fed into the proposed model. From the three variations of datasets, the one processed with time series obtained the highest prediction accuracy, followed by the one processed with linear interpolation. This study will open new ways for addressing healthcare issues related to glucose forecasting in diabetic patients, helping to avoid concomitant complications such as severe episodes of hyperglycemia.

Original language	English
Title of host publication	Smart Technologies, Systems and Applications - 2nd International Conference, SmartTech-IC 2021, Revised Selected Papers
Editors	Fabián R. Narváez, Julio Proaño, Paulina Morillo, Diego Vallejo, Daniel González Montoya, Gloria M. Díaz
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	99-113
Number of pages	15
ISBN (Print)	9783030991692
DOIs	https://doi.org/10.1007/978-3-030-99170-8_8
State	Published - 2022
Externally published	Yes
Event	2nd International Conference on Smart Technologies, Systems and Applications, SmartTech-IC 2021 - Quito, Ecuador Duration: 1 Dec 2021 → 3 Dec 2021

Publication series

Name	Communications in Computer and Information Science
Volume	1532 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	2nd International Conference on Smart Technologies, Systems and Applications, SmartTech-IC 2021
Country/Territory	Ecuador
City	Quito
Period	1/12/21 → 3/12/21

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Blood glucose level prediction
Linear interpolation
Long short-term memory
Machine learning
Time series

Access to Document

10.1007/978-3-030-99170-8_8

Cite this

Gómez-Castillo, N. Y., Cajilima-Cardenaz, P. E., Zhinin-Vera, L., Maldonado-Cuascota, B., León Domínguez, D., Pineda-Molina, G., Hidalgo-Parra, A. A., & Gonzales-Zubiate, F. A. (2022). A Machine Learning Approach for Blood Glucose Level Prediction Using a LSTM Network. In F. R. Narváez, J. Proaño, P. Morillo, D. Vallejo, D. González Montoya, & G. M. Díaz (Eds.), Smart Technologies, Systems and Applications - 2nd International Conference, SmartTech-IC 2021, Revised Selected Papers (pp. 99-113). (Communications in Computer and Information Science; Vol. 1532 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-99170-8_8

Gómez-Castillo, Nayeli Y. ; Cajilima-Cardenaz, Pedro E. ; Zhinin-Vera, Luis et al. / A Machine Learning Approach for Blood Glucose Level Prediction Using a LSTM Network. Smart Technologies, Systems and Applications - 2nd International Conference, SmartTech-IC 2021, Revised Selected Papers. editor / Fabián R. Narváez ; Julio Proaño ; Paulina Morillo ; Diego Vallejo ; Daniel González Montoya ; Gloria M. Díaz. Springer Science and Business Media Deutschland GmbH, 2022. pp. 99-113 (Communications in Computer and Information Science).

@inproceedings{e66e365898ff44ee85e2aa85b9f0ff25,

title = "A Machine Learning Approach for Blood Glucose Level Prediction Using a LSTM Network",

abstract = "Diabetes is a chronic disease characterized by the elevation of glucose in blood resulting in multiple organ failure in the body. There are three types of diabetes: type 1, type 2, and gestational diabetes. Type 1 diabetes (T1D) is an autoimmune disease where insulin-producing cells are destroyed. World Health Organization latest reports indicate T1D prevalence is increasing worldwide with approximately one million new cases annually. Consequently, numerous models to predict blood glucose levels have been proposed, some of which are based on Recurrent Neural Networks (RNNs). The study presented here proposes the training of a machine learning model to predict future glucose levels with high precision using the OhioT1DM database and a Long Short-Term Memory (LSTM) network. Three variations of the dataset were used; the first one with original unprocessed data, another processed with linear interpolation, and a last one processed with a time series method. The datasets obtained were split into time prediction horizons (PH) of 5, 30, and 60 min and then fed into the proposed model. From the three variations of datasets, the one processed with time series obtained the highest prediction accuracy, followed by the one processed with linear interpolation. This study will open new ways for addressing healthcare issues related to glucose forecasting in diabetic patients, helping to avoid concomitant complications such as severe episodes of hyperglycemia.",

keywords = "Blood glucose level prediction, Linear interpolation, Long short-term memory, Machine learning, Time series",

author = "G{\'o}mez-Castillo, \{Nayeli Y.\} and Cajilima-Cardenaz, \{Pedro E.\} and Luis Zhinin-Vera and Bel{\'e}n Maldonado-Cuascota and \{Le{\'o}n Dom{\'i}nguez\}, Diana and Gabriela Pineda-Molina and Hidalgo-Parra, \{Andr{\'e}s A.\} and Gonzales-Zubiate, \{Fernando A.\}",

note = "Publisher Copyright: {\textcopyright} 2022, Springer Nature Switzerland AG.; 2nd International Conference on Smart Technologies, Systems and Applications, SmartTech-IC 2021 ; Conference date: 01-12-2021 Through 03-12-2021",

year = "2022",

doi = "10.1007/978-3-030-99170-8\_8",

language = "Ingl{\'e}s",

isbn = "9783030991692",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "99--113",

editor = "Narv{\'a}ez, \{Fabi{\'a}n R.\} and Julio Proa{\~n}o and Paulina Morillo and Diego Vallejo and \{Gonz{\'a}lez Montoya\}, Daniel and D{\'i}az, \{Gloria M.\}",

booktitle = "Smart Technologies, Systems and Applications - 2nd International Conference, SmartTech-IC 2021, Revised Selected Papers",

}

Gómez-Castillo, NY, Cajilima-Cardenaz, PE, Zhinin-Vera, L, Maldonado-Cuascota, B, León Domínguez, D, Pineda-Molina, G, Hidalgo-Parra, AA & Gonzales-Zubiate, FA 2022, A Machine Learning Approach for Blood Glucose Level Prediction Using a LSTM Network. in FR Narváez, J Proaño, P Morillo, D Vallejo, D González Montoya & GM Díaz (eds), Smart Technologies, Systems and Applications - 2nd International Conference, SmartTech-IC 2021, Revised Selected Papers. Communications in Computer and Information Science, vol. 1532 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 99-113, 2nd International Conference on Smart Technologies, Systems and Applications, SmartTech-IC 2021, Quito, Ecuador, 1/12/21. https://doi.org/10.1007/978-3-030-99170-8_8

A Machine Learning Approach for Blood Glucose Level Prediction Using a LSTM Network. / Gómez-Castillo, Nayeli Y.; Cajilima-Cardenaz, Pedro E.; Zhinin-Vera, Luis et al.
Smart Technologies, Systems and Applications - 2nd International Conference, SmartTech-IC 2021, Revised Selected Papers. ed. / Fabián R. Narváez; Julio Proaño; Paulina Morillo; Diego Vallejo; Daniel González Montoya; Gloria M. Díaz. Springer Science and Business Media Deutschland GmbH, 2022. p. 99-113 (Communications in Computer and Information Science; Vol. 1532 CCIS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A Machine Learning Approach for Blood Glucose Level Prediction Using a LSTM Network

AU - Gómez-Castillo, Nayeli Y.

AU - Cajilima-Cardenaz, Pedro E.

AU - Zhinin-Vera, Luis

AU - Maldonado-Cuascota, Belén

AU - León Domínguez, Diana

AU - Pineda-Molina, Gabriela

AU - Hidalgo-Parra, Andrés A.

AU - Gonzales-Zubiate, Fernando A.

PY - 2022

Y1 - 2022

N2 - Diabetes is a chronic disease characterized by the elevation of glucose in blood resulting in multiple organ failure in the body. There are three types of diabetes: type 1, type 2, and gestational diabetes. Type 1 diabetes (T1D) is an autoimmune disease where insulin-producing cells are destroyed. World Health Organization latest reports indicate T1D prevalence is increasing worldwide with approximately one million new cases annually. Consequently, numerous models to predict blood glucose levels have been proposed, some of which are based on Recurrent Neural Networks (RNNs). The study presented here proposes the training of a machine learning model to predict future glucose levels with high precision using the OhioT1DM database and a Long Short-Term Memory (LSTM) network. Three variations of the dataset were used; the first one with original unprocessed data, another processed with linear interpolation, and a last one processed with a time series method. The datasets obtained were split into time prediction horizons (PH) of 5, 30, and 60 min and then fed into the proposed model. From the three variations of datasets, the one processed with time series obtained the highest prediction accuracy, followed by the one processed with linear interpolation. This study will open new ways for addressing healthcare issues related to glucose forecasting in diabetic patients, helping to avoid concomitant complications such as severe episodes of hyperglycemia.

AB - Diabetes is a chronic disease characterized by the elevation of glucose in blood resulting in multiple organ failure in the body. There are three types of diabetes: type 1, type 2, and gestational diabetes. Type 1 diabetes (T1D) is an autoimmune disease where insulin-producing cells are destroyed. World Health Organization latest reports indicate T1D prevalence is increasing worldwide with approximately one million new cases annually. Consequently, numerous models to predict blood glucose levels have been proposed, some of which are based on Recurrent Neural Networks (RNNs). The study presented here proposes the training of a machine learning model to predict future glucose levels with high precision using the OhioT1DM database and a Long Short-Term Memory (LSTM) network. Three variations of the dataset were used; the first one with original unprocessed data, another processed with linear interpolation, and a last one processed with a time series method. The datasets obtained were split into time prediction horizons (PH) of 5, 30, and 60 min and then fed into the proposed model. From the three variations of datasets, the one processed with time series obtained the highest prediction accuracy, followed by the one processed with linear interpolation. This study will open new ways for addressing healthcare issues related to glucose forecasting in diabetic patients, helping to avoid concomitant complications such as severe episodes of hyperglycemia.

KW - Blood glucose level prediction

KW - Linear interpolation

KW - Long short-term memory

KW - Machine learning

KW - Time series

UR - https://www.scopus.com/pages/publications/85128460035

U2 - 10.1007/978-3-030-99170-8_8

DO - 10.1007/978-3-030-99170-8_8

M3 - Contribución a la conferencia

AN - SCOPUS:85128460035

SN - 9783030991692

T3 - Communications in Computer and Information Science

SP - 99

EP - 113

BT - Smart Technologies, Systems and Applications - 2nd International Conference, SmartTech-IC 2021, Revised Selected Papers

A2 - Narváez, Fabián R.

A2 - Proaño, Julio

A2 - Morillo, Paulina

A2 - Vallejo, Diego

A2 - González Montoya, Daniel

A2 - Díaz, Gloria M.

PB - Springer Science and Business Media Deutschland GmbH

T2 - 2nd International Conference on Smart Technologies, Systems and Applications, SmartTech-IC 2021

Y2 - 1 December 2021 through 3 December 2021

ER -

Gómez-Castillo NY, Cajilima-Cardenaz PE, Zhinin-Vera L, Maldonado-Cuascota B, León Domínguez D, Pineda-Molina G et al. A Machine Learning Approach for Blood Glucose Level Prediction Using a LSTM Network. In Narváez FR, Proaño J, Morillo P, Vallejo D, González Montoya D, Díaz GM, editors, Smart Technologies, Systems and Applications - 2nd International Conference, SmartTech-IC 2021, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2022. p. 99-113. (Communications in Computer and Information Science). doi: 10.1007/978-3-030-99170-8_8