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

Sensors ◽  
2021 ◽  
Vol 21 (23) ◽  
pp. 7815
Author(s):  
Justin Chu ◽  
Wen-Tse Yang ◽  
Wei-Ru Lu ◽  
Yao-Ting Chang ◽  
Tung-Han Hsieh ◽  
...  
Keyword(s):  
Deep Learning ◽  
Blood Glucose ◽  
Sample Size ◽  
Prediction Accuracy ◽  
Non Invasive ◽  
Physiological Diversity ◽  
Promising Solution ◽  
Glucose Prediction ◽  
Error Grid ◽  
Larger Sample

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.

Development of non-invasive blood glucose regression based on near-infrared spectroscopy combined with a deep learning method

2021 ◽  
Author(s):  
Zhuyu Wang ◽  
Linhua Zhou ◽  
Tianqing Liu ◽  
Kewei Huan ◽  
Xiaoning Jia
Keyword(s):  
Infrared Spectroscopy ◽  
Blood Glucose ◽  
Near Infrared Spectroscopy ◽  
Prediction Accuracy ◽  
Near Infrared ◽  
Blood Glucose Concentration ◽  
High Dimensional ◽  
Support Vector ◽  
Non Invasive ◽  
Blood Glucose Concentrations

Abstract Extracting micro-scale spectral features from dynamic blood glucose concentrations is extremely difficult when using non-invasive measurement methods. This work proposes a new machine-learning method based on near-infrared spectroscopy, deep belief network (DBN), and support vector machine (SVR), to improve the prediction accuracy. First, the standard oral glucose tolerance test is used to collect near-infrared spectroscopy and actual blood glucose concentration values for specific wavelengths (1200, 1300, 1350, 1450, 1600, 1610, and 1650 nm), and the blood glucose concentrations is within a clinical range of 70mg/dL~220mg/dL. Second, based on the DBN model, high-dimensional deep features of the non-invasive blood glucose spectrum are extracted. These are used to establish a support vector regression (SVR) model and to quantitatively analyze the influence of spectral sample size and corresponding feature dimensions (i.e., DBN network structure) on the prediction accuracy. Finally, based on data from six volunteers, a comparative analysis of the SVR prediction accuracy is performed both before and after using high-dimensional deep features. For volunteer 1, when the DBN-based high-dimensional deep features were used, the root mean square error (RMSE) of support vector regression (SVR) was reduced by 71.67%, the correlation coefficient (R2) and the P value of Clark grid analysis (P) were increased by 13.99% and 6.28%, respectively. Moreover, we have similar results when the proposed method was carried out on the data of other volunteers. The results show that the presented algorithm can play an important role in dynamic non-invasive blood glucose concentration prediction and can effectively improve the accuracy of the SVR model. Further, by applying the algorithm to six independent sets of data, this research also illustrates the high-precision regression and generalization capabilities of the DBN-SVR algorithm.

scholarly journals IoMT-Enabled Real-time Blood Glucose Prediction with Deep Learning and Edge Computing

2022 ◽  
pp. 1-1
Author(s):  
Taiyu Zhu ◽  
Lei Kuang ◽  
John Daniels ◽  
Pau Herrero ◽  
Kezhi Li ◽  
...  
Keyword(s):  
Deep Learning ◽  
Blood Glucose ◽  
Real Time ◽  
Edge Computing ◽  
Glucose Prediction

scholarly journals A Deep Learning Approach to Diabetic Blood Glucose Prediction

2017 ◽  
Vol 3 ◽  
Author(s):  
Hrushikesh N. Mhaskar ◽  
Sergei V. Pereverzyev ◽  
Maria D. van der Walt
Keyword(s):  
Deep Learning ◽  
Blood Glucose ◽  
Learning Approach ◽  
Glucose Prediction

scholarly journals A Model for Diabetic Blood Glucose Prediction Based on Electroencephalography Signals Using Deep Learning

2018 ◽  
Vol 12 ◽  
Author(s):  
Ali Berkol ◽  
Gokay Karayegen ◽  
Emre Tartan ◽  
Yahya Ekici ◽  
Gozde Kara ◽  
...  
Keyword(s):  
Deep Learning ◽  
Blood Glucose ◽  
Glucose Prediction

scholarly journals Commercial and Scientific Solutions for Blood Glucose Monitoring—A Review

Sensors ◽  
2022 ◽  
Vol 22 (2) ◽  
pp. 425
Author(s):  
Yirui Xue ◽  
Angelika S. Thalmayer ◽  
Samuel Zeising ◽  
Georg Fischer ◽  
Maximilian Lübke
Keyword(s):  
Blood Glucose ◽  
Glucose Monitoring ◽  
Glucose Sensing ◽  
Research Field ◽  
Blood Glucose Monitoring ◽  
Invasive Monitoring ◽  
Non Invasive ◽  
Incurable Disease ◽  
Clarke Error Grid ◽  
Error Grid

Diabetes is a chronic and, according to the state of the art, an incurable disease. Therefore, to treat diabetes, regular blood glucose monitoring is crucial since it is mandatory to mitigate the risk and incidence of hyperglycemia and hypoglycemia. Nowadays, it is common to use blood glucose meters or continuous glucose monitoring via stinging the skin, which is classified as invasive monitoring. In recent decades, non-invasive monitoring has been regarded as a dominant research field. In this paper, electrochemical and electromagnetic non-invasive blood glucose monitoring approaches will be discussed. Thereby, scientific sensor systems are compared to commercial devices by validating the sensor principle and investigating their performance utilizing the Clarke error grid. Additionally, the opportunities to enhance the overall accuracy and stability of non-invasive glucose sensing and even predict blood glucose development to avoid hyperglycemia and hypoglycemia using post-processing and sensor fusion are presented. Overall, the scientific approaches show a comparable accuracy in the Clarke error grid to that of the commercial ones. However, they are in different stages of development and, therefore, need improvement regarding parameter optimization, temperature dependency, or testing with blood under real conditions. Moreover, the size of scientific sensing solutions must be further reduced for a wearable monitoring system.

Sign in / Sign up

Export Citation Format

Share Document