Random Forest Analysis of Combined Millimeter-Wave and Near-Infrared Sensing for Noninvasive Glucose Detection

Yuyang Sun; Helena Cano-Garcia; Efthymios Kallos; Fergus O’Brien; Adeyemi Akintonde; Diana Elena Motei; Oana Ancu; Richard William Alexander Mackenzie; Panagiotis Kosmas

doi:10.1109/jsen.2023.3293248

Random Forest Analysis of Combined Millimeter-Wave and Near-Infrared Sensing for Noninvasive Glucose Detection

Yuyang Sun
, Helena Cano-Garcia
, Efthymios Kallos
, Fergus O’Brien
, Adeyemi Akintonde
, Diana Elena Motei
, Oana Ancu
, Richard William Alexander Mackenzie
, Panagiotis Kosmas

Research output: Contribution to journal › Article › peer-review

25 Citations (Scopus)

230 Downloads (Pure)

Abstract

Recently, great progress has been achieved in diabetes management through minimally invasive continuous glucose monitoring, which involves piercing the skin to identify changes in glucose levels in the interstitial fluid. Nevertheless, the development of accurate fully noninvasive (NI) glucose monitors remains a topic of great interest. This work examines the processing of data collected from a multisensor system, which combines millimeter-wave (mm-wave), near-infrared (NIR), and other auxiliary sensors' technology, coupled with a random forest machine learning (ML) model for prediction. We investigated possible correlations between mm-wave, NIR, skin temperature, pressure data, and blood glucose values and deployed an ML model for prediction. The model is based on the random forest algorithm that was trained and tested on data collected from a study involving five healthy human subjects undergoing an intravenous glucose tolerance test. Our model achieves a root mean square error of 21.06 mg/dl and a mean absolute relative difference of 7.31% for glucose prediction in the collected clinical subjects dataset. In addition, 96.1% of the model's predictions fall within the clinically acceptable zones according to the Clarke error grid analysis. The preliminary results suggest that combining multiple sensor data with ML techniques is promising for the development of NI glucose monitoring.

Original language	English
Pages (from-to)	20294-20309
Number of pages	16
Journal	IEEE Sensors Journal
Volume	23
Issue number	17
Early online date	12 Jul 2023
DOIs	https://doi.org/10.1109/jsen.2023.3293248
Publication status	Published - 1 Sept 2023
Externally published	Yes

Bibliographical note

© 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Copyright © and Moral Rights are retained by the author(s) and/ or other copyright owners. A copy can be downloaded for personal non-commercial research or study, without prior permission or charge. This item cannot be reproduced or quoted extensively from without first obtaining permission in writing from the copyright holder(s). The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the copyright holders.

This document is the author’s post-print version, incorporating any revisions agreed during the peer-review process. Some differences between the published version and this version may remain and you are advised to consult the published version if you wish to cite from it.

Funder

10.13039/501100006041-Innovate UK (Grant Number: 104554 (2019–2021))

UN SDGs

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

SDG 3 Good Health and Well-being

Keywords

Continuous glucose monitoring (CGM)
diabetes
glucose forecasting
infrared sensing
machine learning (ML)
millimeter-wave (mm-wave) radar

Access to Document

10.1109/jsen.2023.3293248

Post-PrintAccepted author manuscript, 2.06 MBLicence: CC BY

Cite this

@article{8d5db331e16143c4b62efde83e919ccf,

title = "Random Forest Analysis of Combined Millimeter-Wave and Near-Infrared Sensing for Noninvasive Glucose Detection",

abstract = "Recently, great progress has been achieved in diabetes management through minimally invasive continuous glucose monitoring, which involves piercing the skin to identify changes in glucose levels in the interstitial fluid. Nevertheless, the development of accurate fully noninvasive (NI) glucose monitors remains a topic of great interest. This work examines the processing of data collected from a multisensor system, which combines millimeter-wave (mm-wave), near-infrared (NIR), and other auxiliary sensors' technology, coupled with a random forest machine learning (ML) model for prediction. We investigated possible correlations between mm-wave, NIR, skin temperature, pressure data, and blood glucose values and deployed an ML model for prediction. The model is based on the random forest algorithm that was trained and tested on data collected from a study involving five healthy human subjects undergoing an intravenous glucose tolerance test. Our model achieves a root mean square error of 21.06 mg/dl and a mean absolute relative difference of 7.31\% for glucose prediction in the collected clinical subjects dataset. In addition, 96.1\% of the model's predictions fall within the clinically acceptable zones according to the Clarke error grid analysis. The preliminary results suggest that combining multiple sensor data with ML techniques is promising for the development of NI glucose monitoring.",

keywords = "Continuous glucose monitoring (CGM), diabetes, glucose forecasting, infrared sensing, machine learning (ML), millimeter-wave (mm-wave) radar",

author = "Yuyang Sun and Helena Cano-Garcia and Efthymios Kallos and Fergus O{\textquoteright}Brien and Adeyemi Akintonde and Motei, \{Diana Elena\} and Oana Ancu and Mackenzie, \{Richard William Alexander\} and Panagiotis Kosmas",

note = "{\textcopyright} 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Copyright {\textcopyright} and Moral Rights are retained by the author(s) and/ or other copyright owners. A copy can be downloaded for personal non-commercial research or study, without prior permission or charge. This item cannot be reproduced or quoted extensively from without first obtaining permission in writing from the copyright holder(s). The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the copyright holders. This document is the author{\textquoteright}s post-print version, incorporating any revisions agreed during the peer-review process. Some differences between the published version and this version may remain and you are advised to consult the published version if you wish to cite from it. ",

year = "2023",

month = sep,

day = "1",

doi = "10.1109/jsen.2023.3293248",

language = "English",

volume = "23",

pages = "20294--20309",

journal = "IEEE Sensors Journal",

issn = "1530-437X",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "17",

}

TY - JOUR

T1 - Random Forest Analysis of Combined Millimeter-Wave and Near-Infrared Sensing for Noninvasive Glucose Detection

AU - Sun, Yuyang

AU - Cano-Garcia, Helena

AU - Kallos, Efthymios

AU - O’Brien, Fergus

AU - Akintonde, Adeyemi

AU - Motei, Diana Elena

AU - Ancu, Oana

AU - Mackenzie, Richard William Alexander

AU - Kosmas, Panagiotis

N1 - © 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Copyright © and Moral Rights are retained by the author(s) and/ or other copyright owners. A copy can be downloaded for personal non-commercial research or study, without prior permission or charge. This item cannot be reproduced or quoted extensively from without first obtaining permission in writing from the copyright holder(s). The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the copyright holders. This document is the author’s post-print version, incorporating any revisions agreed during the peer-review process. Some differences between the published version and this version may remain and you are advised to consult the published version if you wish to cite from it.

PY - 2023/9/1

Y1 - 2023/9/1

N2 - Recently, great progress has been achieved in diabetes management through minimally invasive continuous glucose monitoring, which involves piercing the skin to identify changes in glucose levels in the interstitial fluid. Nevertheless, the development of accurate fully noninvasive (NI) glucose monitors remains a topic of great interest. This work examines the processing of data collected from a multisensor system, which combines millimeter-wave (mm-wave), near-infrared (NIR), and other auxiliary sensors' technology, coupled with a random forest machine learning (ML) model for prediction. We investigated possible correlations between mm-wave, NIR, skin temperature, pressure data, and blood glucose values and deployed an ML model for prediction. The model is based on the random forest algorithm that was trained and tested on data collected from a study involving five healthy human subjects undergoing an intravenous glucose tolerance test. Our model achieves a root mean square error of 21.06 mg/dl and a mean absolute relative difference of 7.31% for glucose prediction in the collected clinical subjects dataset. In addition, 96.1% of the model's predictions fall within the clinically acceptable zones according to the Clarke error grid analysis. The preliminary results suggest that combining multiple sensor data with ML techniques is promising for the development of NI glucose monitoring.

AB - Recently, great progress has been achieved in diabetes management through minimally invasive continuous glucose monitoring, which involves piercing the skin to identify changes in glucose levels in the interstitial fluid. Nevertheless, the development of accurate fully noninvasive (NI) glucose monitors remains a topic of great interest. This work examines the processing of data collected from a multisensor system, which combines millimeter-wave (mm-wave), near-infrared (NIR), and other auxiliary sensors' technology, coupled with a random forest machine learning (ML) model for prediction. We investigated possible correlations between mm-wave, NIR, skin temperature, pressure data, and blood glucose values and deployed an ML model for prediction. The model is based on the random forest algorithm that was trained and tested on data collected from a study involving five healthy human subjects undergoing an intravenous glucose tolerance test. Our model achieves a root mean square error of 21.06 mg/dl and a mean absolute relative difference of 7.31% for glucose prediction in the collected clinical subjects dataset. In addition, 96.1% of the model's predictions fall within the clinically acceptable zones according to the Clarke error grid analysis. The preliminary results suggest that combining multiple sensor data with ML techniques is promising for the development of NI glucose monitoring.

KW - Continuous glucose monitoring (CGM)

KW - diabetes

KW - glucose forecasting

KW - infrared sensing

KW - machine learning (ML)

KW - millimeter-wave (mm-wave) radar

UR - https://www.scopus.com/pages/publications/85164691059

U2 - 10.1109/jsen.2023.3293248

DO - 10.1109/jsen.2023.3293248

M3 - Article

SN - 1530-437X

VL - 23

SP - 20294

EP - 20309

JO - IEEE Sensors Journal

JF - IEEE Sensors Journal

IS - 17

ER -

Random Forest Analysis of Combined Millimeter-Wave and Near-Infrared Sensing for Noninvasive Glucose Detection

Abstract

Bibliographical note

Funder

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this