Abstract
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.
Highlights
An accurate and reliable non-invasive blood glucose (NIBG) measuring technique has long been in demand and extensively studied
The universal model of not separating subjects into cohorts nor including HbA1c was established as a baseline to examine the effects of medication and the inclusion of HbA1c on the other models
The following metrics were used to assess the performance of models: root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), coefficient of determination ( ), proportion
Summary
An accurate and reliable non-invasive blood glucose (NIBG) measuring technique has long been in demand and extensively studied. 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 (SpO2) and pulsation rate. It is known that the light absorption and reflectance of specific wavelengths are sensitive to the body’s hemodynamic properties, which are highly correlated to the health status of the cardiovascular system, and the cardiovascular system is influenced in the long term by BG levels that are directly measurable as pulse morphological profiles [9,11,13,16]. As a result, finding a correlation between PPG pulse morphology and BG levels could be a viable way towards NIBG prediction
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
