Prediabetes detection in unconstrained conditions using wearable sensors

Summary: Prediabetes is a common health condition that often goes undetected until it progresses to type 2 diabetes. Early identification of prediabetes is essential for timely intervention and prevention of complications. This research explores the feasibility of using wearable continuous glucose m...

Full description

Saved in:
Bibliographic Details
Main Authors: Dimitra Tatli, Vasileios Papapanagiotou, Aris Liakos, Apostolos Tsapas, Anastasios Delopoulos
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:Clinical Nutrition Open Science
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2667268524000950
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850245075317030912
author Dimitra Tatli
Vasileios Papapanagiotou
Aris Liakos
Apostolos Tsapas
Anastasios Delopoulos
author_facet Dimitra Tatli
Vasileios Papapanagiotou
Aris Liakos
Apostolos Tsapas
Anastasios Delopoulos
author_sort Dimitra Tatli
collection DOAJ
description Summary: Prediabetes is a common health condition that often goes undetected until it progresses to type 2 diabetes. Early identification of prediabetes is essential for timely intervention and prevention of complications. This research explores the feasibility of using wearable continuous glucose monitoring along with smartwatches with embedded inertial sensors to collect glucose measurements and acceleration signals respectively, for the early detection of prediabetes. We propose a methodology based on signal processing and machine learning techniques. Two feature sets are extracted from the collected signals, based both on a dynamic modeling of the human glucose-homeostasis system and on the Glucose curve, inspired by three major glucose related blood tests. Features are aggregated per individual using bootstrap. Support Vector Machines are used to classify normoglycemic vs. prediabetic individuals. We collected data from 22 participants for evaluation. The results are highly encouraging, demonstrating high sensitivity and precision. This work is a proof of concept, highlighting the potential of wearable devices in prediabetes assessment. Future directions involve expanding the study to a larger, more diverse population and exploring the integration of CGM and smartwatch functionalities into a unified device. Automated eating detecting algorithms can also be used.
format Article
id doaj-art-46f4fecb0d134204aab279a072b10793
institution OA Journals
issn 2667-2685
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series Clinical Nutrition Open Science
spelling doaj-art-46f4fecb0d134204aab279a072b107932025-08-20T01:59:34ZengElsevierClinical Nutrition Open Science2667-26852024-12-015816317410.1016/j.nutos.2024.09.013Prediabetes detection in unconstrained conditions using wearable sensorsDimitra Tatli0Vasileios Papapanagiotou1Aris Liakos2Apostolos Tsapas3Anastasios Delopoulos4Multimedia Understanding Group, Dpt. of Electrical and Computer Engineering, Faculty of Engineering, AUTH, Greece; Embedded Systems Laboratory, Institute of Electrical and Micro Engineering, EPFL, Lausanne, Switzerland; Corresponding author. Embedded Systems Laboratory, Institute of Electrical and Micro Engineering, EPFL, Lausanne, Switzerland.Multimedia Understanding Group, Dpt. of Electrical and Computer Engineering, Faculty of Engineering, AUTH, Greece; IMPACT Research Group, Dpt. of Medicine, Karolinska Institutet, Huddinge, SwedenDiabetes Centre, Second Medical Department at Ippokratio General Hospital, Thessaloniki, GreeceDiabetes Centre, Second Medical Department at Ippokratio General Hospital, Thessaloniki, GreeceMultimedia Understanding Group, Dpt. of Electrical and Computer Engineering, Faculty of Engineering, AUTH, GreeceSummary: Prediabetes is a common health condition that often goes undetected until it progresses to type 2 diabetes. Early identification of prediabetes is essential for timely intervention and prevention of complications. This research explores the feasibility of using wearable continuous glucose monitoring along with smartwatches with embedded inertial sensors to collect glucose measurements and acceleration signals respectively, for the early detection of prediabetes. We propose a methodology based on signal processing and machine learning techniques. Two feature sets are extracted from the collected signals, based both on a dynamic modeling of the human glucose-homeostasis system and on the Glucose curve, inspired by three major glucose related blood tests. Features are aggregated per individual using bootstrap. Support Vector Machines are used to classify normoglycemic vs. prediabetic individuals. We collected data from 22 participants for evaluation. The results are highly encouraging, demonstrating high sensitivity and precision. This work is a proof of concept, highlighting the potential of wearable devices in prediabetes assessment. Future directions involve expanding the study to a larger, more diverse population and exploring the integration of CGM and smartwatch functionalities into a unified device. Automated eating detecting algorithms can also be used.http://www.sciencedirect.com/science/article/pii/S2667268524000950PrediabetesContinuous glucose monitorsSignal processingMachine learning
spellingShingle Dimitra Tatli
Vasileios Papapanagiotou
Aris Liakos
Apostolos Tsapas
Anastasios Delopoulos
Prediabetes detection in unconstrained conditions using wearable sensors
Clinical Nutrition Open Science
Prediabetes
Continuous glucose monitors
Signal processing
Machine learning
title Prediabetes detection in unconstrained conditions using wearable sensors
title_full Prediabetes detection in unconstrained conditions using wearable sensors
title_fullStr Prediabetes detection in unconstrained conditions using wearable sensors
title_full_unstemmed Prediabetes detection in unconstrained conditions using wearable sensors
title_short Prediabetes detection in unconstrained conditions using wearable sensors
title_sort prediabetes detection in unconstrained conditions using wearable sensors
topic Prediabetes
Continuous glucose monitors
Signal processing
Machine learning
url http://www.sciencedirect.com/science/article/pii/S2667268524000950
work_keys_str_mv AT dimitratatli prediabetesdetectioninunconstrainedconditionsusingwearablesensors
AT vasileiospapapanagiotou prediabetesdetectioninunconstrainedconditionsusingwearablesensors
AT arisliakos prediabetesdetectioninunconstrainedconditionsusingwearablesensors
AT apostolostsapas prediabetesdetectioninunconstrainedconditionsusingwearablesensors
AT anastasiosdelopoulos prediabetesdetectioninunconstrainedconditionsusingwearablesensors