FPGA based system for blood glucose sensing using photoplethysmography and online motion artifact correction using adaline

Mei

et al. 2020

IEEE Trans. Ind. Inf.

Blood glucose needs to be monitored on a regular basis to prevent diabetes from consuming the health of hyperglycemic patients. Currently in clinic, it is measured using an invasive technique which is uncomfortable and has risky of infection. To facilitate daily care at home, we propose an intelligent, non-invasive blood glucose monitoring system which can differentiate a user's blood glucose level into normal, borderline and warning based on smartphone PPG signals. The main implementation processes of the proposed system include: (1) a novel algorithm for acquiring photoplethysmography (PPG) signals using only smartphone camera videos; (2) a fitting-based sliding window (FSW) algorithm to remove varying degrees of baseline drifts and segment the signal into single periods; (3) extracting characteristic features from the Gaussian functions by comparing PPG signals at different blood glucose levels; (4) categorizing the valid samples into three glucose levels by applying machine learning algorithms. Our proposed system was evaluated on a data set of 80 subjects. Experimental results demonstrate that the system can separate valid signals from invalid ones at an accuracy of 97.54% and the overall accuracy of estimating the blood glucose levels reaches 81.49%. The proposed system provides a reference for the introduction of non-invasive blood glucose technology into daily or clinical applications. This research also indicates that smartphone-based PPG signals have great potentiality to assess individual's blood glucose level.

Section: Related Workmentioning

confidence: 99%

“…D IABETES is a lifelong metabolic disease characterized by chronic hyperglycemia due to multiple causes such as impaired insulin secretion and impaired biological effects [1]. Currently, there is no effective treatment of diabetes, but regular monitoring of blood glucose level can reduce or delay the incidence of complications [2].…”

Section: Introductionmentioning

confidence: 99%

A Noninvasive Blood Glucose Monitoring System Based on Smartphone PPG Signal Processing and Machine Learning

Mei

et al. 2020

IEEE Trans. Ind. Inf.

“…While finger-prick blood glucose (BG) measurement causes pain and discomfort due to its invasive nature and introduces the risk of infection [ 1 , 2 ], NIBG technology can greatly improve the quality of life in diabetic patients by alleviating the pain of regularly invasive measurement [ 3 , 4 , 5 , 6 , 7 , 8 ]. Among the investigated NIBG measurements, photoplethysmography (PPG) has been long anticipated since this technique is simple, low-cost, already commonly implanted on various wearable devices [ 9 , 10 , 11 , 12 , 13 , 14 , 15 ], and has already been successfully applied to the measurement of oxygen saturation (SpO 2 ) and pulsation rate. A PPG device measures the changes in transmittance or reflectance of near-infrared when blood passes through peripheral capillaries.…”

Section: Introductionmentioning

confidence: 99%

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

Chu

Yang

et al. 2021

Sensors

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.

IEEJ Transactions Elec Engng

“…The signal has been used in the study of detecting and diagnosing HR and HR variability (HRV) , blood oxygen saturation (SpO 2 ) [6], blood pressure , respiratory rate , vascular assessment , cardiac output , and venous refilling time . Research has also shown that multisite PPG (ears, thumbs, and big toes) measurement is useful for vascular disease detection and measuring blood glucose levels . Multisite PPG compares two PPG signal sequences recorded at anatomically symmetric sites on the human body (e.g.…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, researchers need methods to collect PPG information during long time periods and from multiple sites with high accuracy for the best diagnosis. Several PPG instruments have been developed, both custom‐made and application‐specific . Recently, communications and information technologies have aided significantly in the improvement and expansion of the PPG signal acquisition system.…”

Section: Introductionmentioning

confidence: 99%

Development of a real‐time adaptive delta compression algorithm for photoplethysmography system

Chong

Beng

Zahedi

2018

Photoplethysmography (PPG) is a simple and low-cost optical technique that measures the blood volume change of the microvascular bed. The multichannel PPG recording system enables researchers to study the characteristics of the PPG signal from multiple body sites. Long-time, continuous PPG signal monitoring helps doctors in medical diagnosis and treatment. However, the development of the PPG signal acquisition system is limited by storage capacity and computational and hardware capability. Limitation in storage capacity and communication efficiency can be addressed by introducing data compression algorithms. In this paper, a single-channel, high-resolution (24-bit) wireless PPG data acquisition system using Bluetooth connectivity with the adaptive delta compression algorithm is presented. A graphical user interface was developed using MATLAB to interface with the system using a serial port profile. The PPG signals were decompressed and stored in the local hard drive for future analysis. The performance of adaptive data compression algorithm was evaluated with 12 healthy subjects. Results show that the compression ratio achieved is 41.5. The mean percentile root-mean-square (RMS) difference at the sampling frequency of 100 Hz is 0.13%. The mean normalized percentile RMS difference value is 9.78%. The system was successfully tested in a continuous acquisition mode for up to 60 min.