Noninvasive blood glucose monitoring system based on near-infrared method

Abstract

Diabetes is considered one of the life-threatening diseases in the world which need continuous monitoring to avoid the complication of diabetes. There is a need to develop a non-invasive monitoring system that avoids the risk of infection problems and pain caused by invasive monitoring techniques. This paper presents a method for developing a noninvasive technique to predict the blood glucose concentration (BCG) based on the Near-infrared (NIR) light sensor. A prototype is developed using a finger sensor based on LED of 940 nm wavelength to collect photoplethysmography (PPG) signal which is variable depending on the glucose concentration variance, a module circuit to preprocess PPG signals is realized, which includes an amplifier and analog filter circuits, an Arduino UNO is used to analog-to-digital conversion. A digital Butterworth filterer is used to remove PPG signal trends, then detect the PPG data peaks to determine the relationship between the PPG signal and (BCG) and use it as input parameters to build the calibration model based on linear regression. Experiments show that the Root Mean Squares Error (RMSE) of the prediction is between 8.264mg/dL and 13.166 mg/dL, the average of RMSE is about 10.44mg/dL with a correlation coefficient (R^2) of 0.839, it is observed that the prediction of glucose concentration is in the clinically acceptable region of the standard Clark Error Grid (CEG).

Evaluating of a Super Bright LED As a Spectrophotometer Light Source at The Clinical Laboratory

“…Generally, spectrophotometers use halogen lamps as light sources with a lifetime of certain. In addition, there are filters/wavelengths that can be selected according to the desired examination [8][9] [10]. Of course, such a spectrophotometric tool has a fairly high price, with a limited lifetime of halogen lamps.…”

Deduction learning for precise noninvasive measurements of blood glucose with a dozen rounds of data for model training

“…A summarized table of comparing previous works of PPG based NIBG with personalized models and this work is illustrated in Table 1 . Al-dhaheri et al 32 collected fifteen days of PPG signals from 10 healthy human subjects, where a linear regression model was trained from the first ten days and tested with the remaining five days of data. Rachim and Chung 17 built a PLSR model from data of every 10 min pre-carbohydrate-rich meals and every 20 min post-meal, for a total of 120 min from 12 healthy volunteers, trained by one day and tested on the next day.…”

“… References Recruited subjects Accuracy (zone A ratio of CEG) Training Rounds for modeling Time span between training and testing Input data Method Age of population V.P. Rachim et al 17 12 healthy subjects 100% ~ 20 1 day 24 features from PPG Linear partial least squares regression Not reported Al-dhaheri et al 32 10 healthy subjects > 90% > 30 Not reported PPG signal voltage Linear regression 20–36 Shu-jen Yeh et al 33 2 diabetes and 1 healthy subject 90% or less, subject dependent 3–4 day, with 15 min interval 1–13 days Temperature-modulated reflectance signal linear least square regression, retrieving training data for best model fitting 50–58 This work 30 diabetic subjects 100% (with auto-screening); 80% (w/o screening) 12 20–85 days PPG signal Deduction Learning 42–76 …”

A review of wearable sensors based monitoring with daily physical activity to manage type 2 diabetes

“…However, achieve high safety living [6][7][8] for elderly and patients is challenged. Thus, healthcare monitoring/tracking for those people are vital [9]. According the World Health Organization (WHO) reports, the number of T2D patients is 422 million in 2014.…”