Please use this identifier to cite or link to this item: http://hdl.handle.net/20.500.12188/25321
Title: Comparing Time and Frequency Domain Heart Rate Variability for Deep Learning-Based Glucose Detection
Authors: Shaqiri, Ervin
Gusev, Marjan 
Poposka, Lidija 
Vavlukis, Marija 
Ahmeti, Irfan 
Keywords: ECG
HRV
Deep learning
Glucose Diabetes
Issue Date: 12-Apr-2022
Publisher: Springer International Publishing
Source: Shaqiri, E., Gusev, M., Poposka, L., Vavlukis, M., Ahmeti, I. (2022). Comparing Time and Frequency Domain Heart Rate Variability for Deep Learning-Based Glucose Detection. In: Antovski, L., Armenski, G. (eds) ICT Innovations 2021. Digital Transformation. ICT Innovations 2021. Communications in Computer and Information Science, vol 1521. Springer, Cham. https://doi.org/10.1007/978-3-031-04206-5_14
Project: Communications in computer and information science (Internet), GLUCO Project
Journal: Communications in Computer and Information Science 
Conference: ICT Innovations: International Conference on ICT Innovations
Abstract: International Conference on ICT Innovations ICT Innovations 2021: ICT Innovations 2021. Digital Transformation pp 188–197Cite as Comparing Time and Frequency Domain Heart Rate Variability for Deep Learning-Based Glucose Detection Ervin Shaqiri, Marjan Gusev, Lidija Poposka, Marija Vavlukis & Irfan Ahmeti Conference paper First Online: 12 April 2022 165 Accesses Part of the Communications in Computer and Information Science book series (CCIS,volume 1521) Abstract Many researchers have been challenged by the usage of devices related to the Internet of Things in conjunction with machine learning to anticipate and diagnose health issues. Diabetes has always been a problem that society has struggled with. Due to the simplicity with which electrocardiograms can be captured and analysed, deep learning can be used to forecast a patient’s instantaneous glucose levels. Our solution is based on a unique method for calculating heart rate variability that involves segment identification, averaging, and concatenating the data to reveal better feature engineering results. Immediate plasma glucose levels are detected using short-term heart rate variability and applied a deep learning method based on Autokeras. In this paper. we address a research question to compare the predictive capability of time and frequency domain features for instantaneous glucose values. The neural architectural search for the time domain approach provided the best results for the 15-min electrocardiogram measurements. Similarly, the Frequency Domain approach showed better results on the same time frame. Regarding the time domain the best results are as follows: RMSE (0.368), MSE (0.193), R2 (0.513), and R2 loss (0.541). The best results for the Frequency Domain approach the best results are as follows: RMSE (0.301), MSE (0.346), R2 (0.45578), and R2 loss (0.482).
URI: http://hdl.handle.net/20.500.12188/25321
DOI: 10.1007/978-3-031-04206-5_14
Appears in Collections:Faculty of Medicine: Conference papers

Show full item record

Page view(s)

39
checked on Mar 29, 2024

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.