Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data

Pietro Bosoni; Marco Meccariello; Valeria Calcaterra; Cristiana Larizza; Lucia Sacchi; Riccardo Bellazzi

doi:10.1007/978-3-030-59137-3_6

Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data

Pietro Bosoni, Marco Meccariello, Valeria Calcaterra, Cristiana Larizza, Lucia Sacchi, Riccardo Bellazzi

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

Abstract

Blood glucose (BG) monitoring devices play an important role in diabetes management, offering real time BG measurements, which can be analyzed to discover new knowledge. In this paper we present a multi-patient and multivariate deep learning approach, based on Long-Short Term Memory (LSTM) artificial neural networks, for building a generalized model to forecast BG levels on a short-time prediction horizon. The proposed framework is evaluated on a clinical dataset of 17 patients, receiving care at the IRCCS Policlinico San Matteo hospital in Pavia, Italy. BG profiles collected by a flash glucose monitoring system were analyzed together with information collected by an activity tracker, including heart rate, sleep, and physical activity. Results suggest that a model with good prediction performance can be obtained and that a combination of HR and lifestyle monitoring signals can help to predict BG levels.

Original language	English
Title of host publication	Artificial Intelligence in Medicine - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Proceedings
Editors	Martin Michalowski, Robert Moskovitch
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	59-63
Number of pages	5
ISBN (Print)	9783030591366
DOIs	https://doi.org/10.1007/978-3-030-59137-3_6
Publication status	Published - 2020
Event	18th International Conference on Artificial Intelligence in Medicine, AIME 2020 - Minneapolis, United States Duration: Aug 25 2020 → Aug 28 2020

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12299 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	18th International Conference on Artificial Intelligence in Medicine, AIME 2020
Country	United States
City	Minneapolis
Period	8/25/20 → 8/28/20

Keywords

Data integration
Deep learning
Diabetes
Flash glucose monitoring
Time series analysis

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-030-59137-3_6

Fingerprint Dive into the research topics of 'Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data'. Together they form a unique fingerprint.

Cite this

Bosoni, P., Meccariello, M., Calcaterra, V., Larizza, C., Sacchi, L., & Bellazzi, R. (2020). Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data. In M. Michalowski, & R. Moskovitch (Eds.), Artificial Intelligence in Medicine - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Proceedings (pp. 59-63). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12299 LNAI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-59137-3_6

Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data. / Bosoni, Pietro; Meccariello, Marco; Calcaterra, Valeria; Larizza, Cristiana; Sacchi, Lucia; Bellazzi, Riccardo.

Artificial Intelligence in Medicine - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Proceedings. ed. / Martin Michalowski; Robert Moskovitch. Springer Science and Business Media Deutschland GmbH, 2020. p. 59-63 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12299 LNAI).

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

Bosoni, P, Meccariello, M, Calcaterra, V, Larizza, C, Sacchi, L & Bellazzi, R 2020, Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data. in M Michalowski & R Moskovitch (eds), Artificial Intelligence in Medicine - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12299 LNAI, Springer Science and Business Media Deutschland GmbH, pp. 59-63, 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Minneapolis, United States, 8/25/20. https://doi.org/10.1007/978-3-030-59137-3_6

Bosoni P, Meccariello M, Calcaterra V, Larizza C, Sacchi L, Bellazzi R. Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data. In Michalowski M, Moskovitch R, editors, Artificial Intelligence in Medicine - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Proceedings. Springer Science and Business Media Deutschland GmbH. 2020. p. 59-63. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-59137-3_6

Bosoni, Pietro ; Meccariello, Marco ; Calcaterra, Valeria ; Larizza, Cristiana ; Sacchi, Lucia ; Bellazzi, Riccardo. / Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data. Artificial Intelligence in Medicine - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Proceedings. editor / Martin Michalowski ; Robert Moskovitch. Springer Science and Business Media Deutschland GmbH, 2020. pp. 59-63 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{d7d1835202e940aa91d6fdf807a1f55c,

title = "Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data",

abstract = "Blood glucose (BG) monitoring devices play an important role in diabetes management, offering real time BG measurements, which can be analyzed to discover new knowledge. In this paper we present a multi-patient and multivariate deep learning approach, based on Long-Short Term Memory (LSTM) artificial neural networks, for building a generalized model to forecast BG levels on a short-time prediction horizon. The proposed framework is evaluated on a clinical dataset of 17 patients, receiving care at the IRCCS Policlinico San Matteo hospital in Pavia, Italy. BG profiles collected by a flash glucose monitoring system were analyzed together with information collected by an activity tracker, including heart rate, sleep, and physical activity. Results suggest that a model with good prediction performance can be obtained and that a combination of HR and lifestyle monitoring signals can help to predict BG levels.",

keywords = "Data integration, Deep learning, Diabetes, Flash glucose monitoring, Time series analysis",

author = "Pietro Bosoni and Marco Meccariello and Valeria Calcaterra and Cristiana Larizza and Lucia Sacchi and Riccardo Bellazzi",

year = "2020",

doi = "10.1007/978-3-030-59137-3_6",

language = "English",

isbn = "9783030591366",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

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

pages = "59--63",

editor = "Martin Michalowski and Robert Moskovitch",

booktitle = "Artificial Intelligence in Medicine - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Proceedings",

address = "Germany",

note = "18th International Conference on Artificial Intelligence in Medicine, AIME 2020 ; Conference date: 25-08-2020 Through 28-08-2020",

}

TY - GEN

T1 - Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data

AU - Bosoni, Pietro

AU - Meccariello, Marco

AU - Calcaterra, Valeria

AU - Larizza, Cristiana

AU - Sacchi, Lucia

AU - Bellazzi, Riccardo

PY - 2020

Y1 - 2020

N2 - Blood glucose (BG) monitoring devices play an important role in diabetes management, offering real time BG measurements, which can be analyzed to discover new knowledge. In this paper we present a multi-patient and multivariate deep learning approach, based on Long-Short Term Memory (LSTM) artificial neural networks, for building a generalized model to forecast BG levels on a short-time prediction horizon. The proposed framework is evaluated on a clinical dataset of 17 patients, receiving care at the IRCCS Policlinico San Matteo hospital in Pavia, Italy. BG profiles collected by a flash glucose monitoring system were analyzed together with information collected by an activity tracker, including heart rate, sleep, and physical activity. Results suggest that a model with good prediction performance can be obtained and that a combination of HR and lifestyle monitoring signals can help to predict BG levels.

AB - Blood glucose (BG) monitoring devices play an important role in diabetes management, offering real time BG measurements, which can be analyzed to discover new knowledge. In this paper we present a multi-patient and multivariate deep learning approach, based on Long-Short Term Memory (LSTM) artificial neural networks, for building a generalized model to forecast BG levels on a short-time prediction horizon. The proposed framework is evaluated on a clinical dataset of 17 patients, receiving care at the IRCCS Policlinico San Matteo hospital in Pavia, Italy. BG profiles collected by a flash glucose monitoring system were analyzed together with information collected by an activity tracker, including heart rate, sleep, and physical activity. Results suggest that a model with good prediction performance can be obtained and that a combination of HR and lifestyle monitoring signals can help to predict BG levels.

KW - Data integration

KW - Deep learning

KW - Diabetes

KW - Flash glucose monitoring

KW - Time series analysis

UR - http://www.scopus.com/inward/record.url?scp=85092229665&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85092229665&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-59137-3_6

DO - 10.1007/978-3-030-59137-3_6

M3 - Conference contribution

AN - SCOPUS:85092229665

SN - 9783030591366

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 59

EP - 63

BT - Artificial Intelligence in Medicine - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020, Proceedings

A2 - Michalowski, Martin

A2 - Moskovitch, Robert

PB - Springer Science and Business Media Deutschland GmbH

T2 - 18th International Conference on Artificial Intelligence in Medicine, AIME 2020

Y2 - 25 August 2020 through 28 August 2020

ER -