Cargando…

90% Accuracy for Photoplethysmography-Based Non-Invasive Blood Glucose Prediction by Deep Learning with Cohort Arrangement and Quarterly Measured HbA1c

Previously published photoplethysmography-(PPG) based non-invasive blood glucose (NIBG) measurements have not yet been validated over 500 subjects. As illustrated in this work, we increased the number subjects recruited to 2538 and found that the prediction accuracy (the ratio in zone A of Clarke’s...

Descripción completa

Detalles Bibliográficos
Autores principales: Chu, Justin, Yang, Wen-Tse, Lu, Wei-Ru, Chang, Yao-Ting, Hsieh, Tung-Han, Yang, Fu-Liang
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8659475/
https://www.ncbi.nlm.nih.gov/pubmed/34883817
http://dx.doi.org/10.3390/s21237815
_version_ 1784612971103649792
author Chu, Justin
Yang, Wen-Tse
Lu, Wei-Ru
Chang, Yao-Ting
Hsieh, Tung-Han
Yang, Fu-Liang
author_facet Chu, Justin
Yang, Wen-Tse
Lu, Wei-Ru
Chang, Yao-Ting
Hsieh, Tung-Han
Yang, Fu-Liang
author_sort Chu, Justin
collection PubMed
description Previously published photoplethysmography-(PPG) based non-invasive blood glucose (NIBG) measurements have not yet been validated over 500 subjects. As illustrated in this work, we increased the number subjects recruited to 2538 and found that the prediction accuracy (the ratio in zone A of Clarke’s error grid) reduced to undesirable 60.6%. We suspect the low prediction accuracy induced by larger sample size might arise from the physiological diversity of subjects, and one possibility is that the diversity might originate from medication. Therefore, we split the subjects into two cohorts for deep learning: with and without medication (1682 and 856 recruited subjects, respectively). In comparison, the cohort training for subjects without any medication had approximately 30% higher prediction accuracy over the cohort training for those with medication. Furthermore, by adding quarterly (every 3 months) measured glycohemoglobin (HbA1c), we were able to significantly boost the prediction accuracy by approximately 10%. For subjects without medication, the best performing model with quarterly measured HbA1c achieved 94.3% prediction accuracy, RMSE of 12.4 mg/dL, MAE of 8.9 mg/dL, and MAPE of 0.08, which demonstrates a very promising solution for NIBG prediction via deep learning. Regarding subjects with medication, a personalized model could be a viable means of further investigation.
format Online
Article
Text
id pubmed-8659475
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-86594752021-12-10 90% Accuracy for Photoplethysmography-Based Non-Invasive Blood Glucose Prediction by Deep Learning with Cohort Arrangement and Quarterly Measured HbA1c Chu, Justin Yang, Wen-Tse Lu, Wei-Ru Chang, Yao-Ting Hsieh, Tung-Han Yang, Fu-Liang Sensors (Basel) Article Previously published photoplethysmography-(PPG) based non-invasive blood glucose (NIBG) measurements have not yet been validated over 500 subjects. As illustrated in this work, we increased the number subjects recruited to 2538 and found that the prediction accuracy (the ratio in zone A of Clarke’s error grid) reduced to undesirable 60.6%. We suspect the low prediction accuracy induced by larger sample size might arise from the physiological diversity of subjects, and one possibility is that the diversity might originate from medication. Therefore, we split the subjects into two cohorts for deep learning: with and without medication (1682 and 856 recruited subjects, respectively). In comparison, the cohort training for subjects without any medication had approximately 30% higher prediction accuracy over the cohort training for those with medication. Furthermore, by adding quarterly (every 3 months) measured glycohemoglobin (HbA1c), we were able to significantly boost the prediction accuracy by approximately 10%. For subjects without medication, the best performing model with quarterly measured HbA1c achieved 94.3% prediction accuracy, RMSE of 12.4 mg/dL, MAE of 8.9 mg/dL, and MAPE of 0.08, which demonstrates a very promising solution for NIBG prediction via deep learning. Regarding subjects with medication, a personalized model could be a viable means of further investigation. MDPI 2021-11-24 /pmc/articles/PMC8659475/ /pubmed/34883817 http://dx.doi.org/10.3390/s21237815 Text en © 2021 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Chu, Justin
Yang, Wen-Tse
Lu, Wei-Ru
Chang, Yao-Ting
Hsieh, Tung-Han
Yang, Fu-Liang
90% Accuracy for Photoplethysmography-Based Non-Invasive Blood Glucose Prediction by Deep Learning with Cohort Arrangement and Quarterly Measured HbA1c
title 90% Accuracy for Photoplethysmography-Based Non-Invasive Blood Glucose Prediction by Deep Learning with Cohort Arrangement and Quarterly Measured HbA1c
title_full 90% Accuracy for Photoplethysmography-Based Non-Invasive Blood Glucose Prediction by Deep Learning with Cohort Arrangement and Quarterly Measured HbA1c
title_fullStr 90% Accuracy for Photoplethysmography-Based Non-Invasive Blood Glucose Prediction by Deep Learning with Cohort Arrangement and Quarterly Measured HbA1c
title_full_unstemmed 90% Accuracy for Photoplethysmography-Based Non-Invasive Blood Glucose Prediction by Deep Learning with Cohort Arrangement and Quarterly Measured HbA1c
title_short 90% Accuracy for Photoplethysmography-Based Non-Invasive Blood Glucose Prediction by Deep Learning with Cohort Arrangement and Quarterly Measured HbA1c
title_sort 90% accuracy for photoplethysmography-based non-invasive blood glucose prediction by deep learning with cohort arrangement and quarterly measured hba1c
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8659475/
https://www.ncbi.nlm.nih.gov/pubmed/34883817
http://dx.doi.org/10.3390/s21237815
work_keys_str_mv AT chujustin 90accuracyforphotoplethysmographybasednoninvasivebloodglucosepredictionbydeeplearningwithcohortarrangementandquarterlymeasuredhba1c
AT yangwentse 90accuracyforphotoplethysmographybasednoninvasivebloodglucosepredictionbydeeplearningwithcohortarrangementandquarterlymeasuredhba1c
AT luweiru 90accuracyforphotoplethysmographybasednoninvasivebloodglucosepredictionbydeeplearningwithcohortarrangementandquarterlymeasuredhba1c
AT changyaoting 90accuracyforphotoplethysmographybasednoninvasivebloodglucosepredictionbydeeplearningwithcohortarrangementandquarterlymeasuredhba1c
AT hsiehtunghan 90accuracyforphotoplethysmographybasednoninvasivebloodglucosepredictionbydeeplearningwithcohortarrangementandquarterlymeasuredhba1c
AT yangfuliang 90accuracyforphotoplethysmographybasednoninvasivebloodglucosepredictionbydeeplearningwithcohortarrangementandquarterlymeasuredhba1c