Noninvasive blood glucose estimation system based on photoplethysmography is proposed in this paper. Photoplethysmography (PPG) is an optical technique which uses infrared light to determine the fluctuations in blood stream in the human body with every pulse. However literature reveals that, a functional correlation exists among pulse signal waveform and the glucose contents in the blood. Many researchers have shown estimation of glucose using different spectroscopy techniques which involves an optical sensor. The key involvement of this paper is withdrawal of different features with respect to time domain as well as frequency domain. Exploration of single pulse is also carried out with different features. A PPG data of 182 random people with varied health conditions like diabetic, blood pressure condition is recorded for about 1 min duration. The glucose estimation is done in two ways: the first one is feature extraction from complete PPG signal of a patient and the second one is single pulse wave analysis. All the features from above-mentioned methods are transformed into feature vector which acts as input to system and estimation of blood glucose is performed. We used Clarke Error Grid Analysis for blood glucose estimation which is clinically accepted. With time and frequency domain features we got 43.75% samples in region A and 50% samples in region B, so total 93.75% samples are in regions A and B. Using single pulse wave analysis we got 59.66% samples in region A and 34.61% samples in region B, so total 94.27% samples are in clinically accepted regions (A and B). As single pulse wave analysis shows significant improvement in region A (20% of actual blood glucose value), this method is good selection for glucose measurement.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    The Use of Photoplethysmography for Blood Glucose Estimation by Noninvasive Method


    Additional title:

    Lect. Notes Electrical Eng.




    Publication date :

    2023-03-12


    Size :

    13 pages





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English




    WQ1 Noninvasive Blood Glucose Monitoring in the 2.0-2.5 mum Wavelength Range

    IEEE | British Library Conference Proceedings | 2001


    Improvements on noninvasive blood glucose biosensors using wavelets for quick fault detection

    Campetelli, Germán / Zumoffen, David Alejandro Ramon / Basualdo, Marta | BASE

    Free access


    Noninvasive glucose monitoring using nearinfrared photoacoustic spectroscopy

    Quan, K.M. / Johnston, E.M. / MacKenzie, H.A. | IEEE | 1994


    Noninvasive Retinal Blood Flow Measurements

    Petrig, B. L. et al. | British Library Online Contents | 1996