TY - GEN
T1 - Bayesian Optimization-Based CNN Model for Blood Glucose Estimation Using Photoplethysmography Signals
AU - Alghlayini, Saifeddin
AU - Al-Betar, Mohammed Azmi
AU - Atef, Mohamed
AU - Al-Naymat, Ghazi
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
PY - 2024
Y1 - 2024
N2 - This paper presents a novel Bayesian optimization-based convolutional neural network (CNN) model for non-invasive blood glucose level (BGL) predication using photoplethysmography (PPG) signals. The Bayesian search found the optimal CNN architecture, achieving 92.85% accuracy in Clarke error grid (CEG) zone A with a wide detection range of 50–200 mg/dL and without any projected values within zones C, D, or E. This indicates that the suggested model is clinically acceptable. The model demonstrated better results in performance assessment using mean absolute error (MAE) and root mean squared error (RMSE) measures. This was achieved by utilizing only the features extracted by the CNN model, eliminating the need for additional feature extraction. This approach reduces computational demands and improves real-time performance. Also, the model achieved a much lower Standard Error of Prediction (SEP) compared with ML-based models. The proposed model is lightweight and can be easily deployed as a stand-alone device using a microcontroller like the Arduino Nano 33 BLE Sense connected with a PPG sensor.
AB - This paper presents a novel Bayesian optimization-based convolutional neural network (CNN) model for non-invasive blood glucose level (BGL) predication using photoplethysmography (PPG) signals. The Bayesian search found the optimal CNN architecture, achieving 92.85% accuracy in Clarke error grid (CEG) zone A with a wide detection range of 50–200 mg/dL and without any projected values within zones C, D, or E. This indicates that the suggested model is clinically acceptable. The model demonstrated better results in performance assessment using mean absolute error (MAE) and root mean squared error (RMSE) measures. This was achieved by utilizing only the features extracted by the CNN model, eliminating the need for additional feature extraction. This approach reduces computational demands and improves real-time performance. Also, the model achieved a much lower Standard Error of Prediction (SEP) compared with ML-based models. The proposed model is lightweight and can be easily deployed as a stand-alone device using a microcontroller like the Arduino Nano 33 BLE Sense connected with a PPG sensor.
KW - Bayesian Optimization
KW - Blood Glucose Level (BGL)
KW - Clarke Error Grid (CEG)
KW - Convolutional Neural Network (CNN)
KW - Microcontroller (MCU)
KW - Non-Invasive
KW - Photoplethysmography (PPG)
UR - http://www.scopus.com/inward/record.url?scp=85200943651&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85200943651&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-65522-7_13
DO - 10.1007/978-3-031-65522-7_13
M3 - Conference contribution
AN - SCOPUS:85200943651
SN - 9783031655210
T3 - Lecture Notes in Networks and Systems
SP - 142
EP - 152
BT - Proceedings of the 3rd International Conference on Innovations in Computing Research (ICR’24)
A2 - Daimi, Kevin
A2 - Al Sadoon, Abeer
PB - Springer Science and Business Media Deutschland GmbH
T2 - 3rd International Conference on Innovations in Computing Research, ICR 2024
Y2 - 12 August 2024 through 14 August 2024
ER -