{"title":"Hypoglycemia Prediction in Type 1 Diabetes With Electrocardiography Beat Ensembles.","authors":"Mu-Ruei Tseng, Kathan Vyas, Anurag Das, Waris Quamer, Darpit Dave, Madhav Erranguntla, Carolina Villegas, Daniel DeSalvo, Siripoom McKay, Gerard Cote, Ricardo Gutierrez-Osuna","doi":"10.1177/19322968251319347","DOIUrl":null,"url":null,"abstract":"<p><strong>Introduction: </strong>Current methods to detect hypoglycemia in type 1 diabetes (T1D) require invasive sensors (ie, continuous glucose monitors, CGMs) that generally have low accuracy in the hypoglycemic range. A forward-looking alternative is to monitor physiological changes induced by hypoglycemia that can be measured non-invasively using, eg, electrocardiography (ECG). However, current methods require extraction of fiduciary points in the ECG signal (eg, to estimate QT interval), which is challenging in ambulatory settings.</p><p><strong>Methods: </strong>To address this issue, we present a machine-learning model that uses (1) convolutional neural networks (CNNs) to extract morphological information from raw ECG signals without the need to identify fiduciary points and (2) ensemble learning to aggregate predictions from multiple ECG beats. We evaluate the model on an experimental data set that contains ECG and CGM recordings over a period of 14 days from ten participants with T1D. We consider two testing scenarios, one that divides ECG data according to CGM readings (CGM-split) and another that divides ECG data on a day-to-day basis (day-split).</p><p><strong>Results: </strong>We find that models trained using CGM-splits tend to produce overly optimistic estimates of hypoglycemia prediction, whereas day-splits provide more realistic estimates, which are consistent with the intrinsic accuracy of CGM devices. More importantly, we find that aggregating predictions from multiple ECG beats using ensemble learning significantly improves predictions at the beat level, though these improvements have large inter-individual differences.</p><p><strong>Conclusion: </strong>Deep learning models and ensemble learning can extract and aggregate morphological information in ECG signals that is predictive of hypoglycemia. Using two validation procedures, we estimate an upper bound on the accuracy of ECG hypoglycemia prediction of 81% equal error rate and a lower bound of 60%. Further improvements may be achieved using big-data approaches that require longitudinal data from a large cohort of participants.</p>","PeriodicalId":15475,"journal":{"name":"Journal of Diabetes Science and Technology","volume":" ","pages":"19322968251319347"},"PeriodicalIF":4.1000,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11863193/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Diabetes Science and Technology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1177/19322968251319347","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}
引用次数: 0
Abstract
Introduction: Current methods to detect hypoglycemia in type 1 diabetes (T1D) require invasive sensors (ie, continuous glucose monitors, CGMs) that generally have low accuracy in the hypoglycemic range. A forward-looking alternative is to monitor physiological changes induced by hypoglycemia that can be measured non-invasively using, eg, electrocardiography (ECG). However, current methods require extraction of fiduciary points in the ECG signal (eg, to estimate QT interval), which is challenging in ambulatory settings.
Methods: To address this issue, we present a machine-learning model that uses (1) convolutional neural networks (CNNs) to extract morphological information from raw ECG signals without the need to identify fiduciary points and (2) ensemble learning to aggregate predictions from multiple ECG beats. We evaluate the model on an experimental data set that contains ECG and CGM recordings over a period of 14 days from ten participants with T1D. We consider two testing scenarios, one that divides ECG data according to CGM readings (CGM-split) and another that divides ECG data on a day-to-day basis (day-split).
Results: We find that models trained using CGM-splits tend to produce overly optimistic estimates of hypoglycemia prediction, whereas day-splits provide more realistic estimates, which are consistent with the intrinsic accuracy of CGM devices. More importantly, we find that aggregating predictions from multiple ECG beats using ensemble learning significantly improves predictions at the beat level, though these improvements have large inter-individual differences.
Conclusion: Deep learning models and ensemble learning can extract and aggregate morphological information in ECG signals that is predictive of hypoglycemia. Using two validation procedures, we estimate an upper bound on the accuracy of ECG hypoglycemia prediction of 81% equal error rate and a lower bound of 60%. Further improvements may be achieved using big-data approaches that require longitudinal data from a large cohort of participants.
期刊介绍:
The Journal of Diabetes Science and Technology (JDST) is a bi-monthly, peer-reviewed scientific journal published by the Diabetes Technology Society. JDST covers scientific and clinical aspects of diabetes technology including glucose monitoring, insulin and metabolic peptide delivery, the artificial pancreas, digital health, precision medicine, social media, cybersecurity, software for modeling, physiologic monitoring, technology for managing obesity, and diagnostic tests of glycation. The journal also covers the development and use of mobile applications and wireless communication, as well as bioengineered tools such as MEMS, new biomaterials, and nanotechnology to develop new sensors. Articles in JDST cover both basic research and clinical applications of technologies being developed to help people with diabetes.