Improving Accuracy of Non-Invasive Glucose Monitoring Through Non-local Data Denoising

Abstract

Correlation and clinical interpretation, in respect to the true glucose value of patient is imperative for optimum therapy and disease management. Accuracy of optical glucometer is hampered by many debilitating factors such as concentration range, sampling environment, tongue-to-spectrometer interface, changes in wavelength, polarization or intensity of light, to name a few. Regression techniques are used in such devices to build patient glucose model. This work is an extension to our previous work regarding multivariate calibration for glucose level prediction in non-invasive human tongue spectra. Here, we present our results for noise reduction and data conditioning during glucose spectrum isolation phase. We embed our 'Indicator Function (IF)' scheme into two popular techniques known as Outlier Sample Removal (OSR) and Descriptor Selection (DS). Methodology is tested on dataset 'OCATNE20' obtained from a public domain website and results are compared at both OSR and DS for a wide range of blood serum samples. Our results show that outlier samples identification and removal in early stage significantly increase the prediction of unknown samples typically in the range of 7.95% to 9.84%.

Non-Invasive Glucose Measurement Technologies: Recent Advancements and Future Challenges

“…The variance of random noise has been reduced using an adaptation of the coherent averaging approach [172]. By averaging N different individual samples from continuous frames, the influence of noise is reduced [173]. To boost SNR, frames with the highest count have been selected for [159] Serum averaging [174].…”

Everything You Wanted to Know About Noninvasive Glucose Measurement and Control

“…1) Noise Minimization and Signal Conditioning: The coherent averaging technique has been adapted to minimize the variance of random noise [167]. The impact of noise is minimized with averaging of N number of individual samples coming from the continuous frames [168]. Frames in the maximum count have been chosen for averaging to have SNR improvement [169].…”