An Impedance and Multi-Wavelength Near-Infrared Spectroscopy IC for Non-Invasive Blood Glucose Estimation

Abstract

A multi-modal spectroscopy IC combining impedance spectroscopy (IMPS) and multi-wavelength near-infrared spectroscopy (mNIRS) is proposed for high precision non-invasive glucose level estimation. A combination of IMPS and mNIRS can compensate for the glucose estimation error to improve its accuracy. The IMPS circuit measures dielectric characteristics of the tissue using the RLC resonant frequency and the resonant impedance to estimate the glucose level. To accurately find resonant frequency, a 2-step frequency sweep sinusoidal oscillator (FSSO) is proposed: 1) 8-level coarse frequency switching (f STEP = 9.4 kHz) in 10-76 kHz, and 2) fine analog frequency sweep in the range of 18.9 kHz. During the frequency sweep, the adaptive gain control loop stabilizes the output voltage swing (400 mV p-p ). To improve accuracy of mNIRS, three wavelengths, 850 nm, 950 nm, and 1,300 nm, are used. For highly accurate glucose estimation, the measurement data of the IMPS and mNIRS are combined by an artificial neural network (ANN) in external DSP. The proposed ANN method reduces the mean absolute relative difference to 8.3% from 15% of IMPS, and 15-20% of mNIRS in 80-180 mg/dL blood glucose level. The proposed multi-modal spectroscopy IC occupies 12.5 mm 2 in a 0.18 µm 1P6M CMOS technology and dissipates a peak power of 38 mW with the maximum radiant emitting power of 12.1 mW.Index Terms-Adaptive gain control, artificial neural network, frequency sweep sinusoidal oscillator, impedance spectroscopy, near-infrared spectroscopy, non-invasive glucose monitoring.

Review of Non-Invasive Glucose Sensing Techniques: Optical, Electrical and Breath Acetone

“…In order to counter these effects, multi-technology and multi-sensors can be implemented and tailored to design a non-invasive sensor that is most sensitive to glucose molecules and independent of the many factors that contribute to false readings [65,147,166,167,211]. Artificial intelligence and data analytics is necessary to analyze the data collected and to incorporate it into a prediction model to accurately quantify glucose levels [235].…”

“…A combination of methods can compensate for the error caused by each method separately and thus minimize the effect of confounding factors. A combination of two methods of (1) scattering spectroscopy at three wavelengths of 850, 950 and 1300 nm, and (2) impedance spectroscopy in the frequency range between 10 and 76 kHz was proposed in [166]. The blood glucose level of ten normal volunteers was estimated based on two methods separately and then the obtained results were combined using an artificial neural network algorithm in order to improve the accuracy of glucose estimation.…”

Recent developments in minimally and truly non-invasive blood glucose monitoring techniques

“…A watch-type transdermal sensor using microneedles to extract ISF glucose through skin [10], an ultrasound and multi-sensor-based ear-clip type non-invasive sensor for discrete measurement [11], a contact-lens-type noninvasive sensor measuring glucose concentration in tears [12], an occlusion spectroscopy-based sensor measuring blood flow [13], and electrochemical subcutaneous continuous monitoring sensors [14] have been reported. A sensor based on the combination of impedance (10 to 76 kHz range) and multiwavelength (850nm, 950nm, 1300nm) near infra-red (NIR) spectroscopy with a dedicated integrated circuit [15] and a technique measuring the refraction of visible laser light to overcome the signal to noise ratio issue related to NIR spectroscopy [16] were also reported. Wearable sensors capable of measuring glucose concentration through saliva and sweat [17] [18], and a benchtop system capable of estimating glucose level through breath have been published [19] [20].…”

Noninvasive measurement of glucose concentration on human fingertip by optical coherence tomography

“…2 Methods of the former type typically track the optical rotation angle, near-infrared (NIR)/midinfrared absorption coefficient, Raman shift, or NIR photoacoustic absorption. [3][4][5] By contrast, methods of the latter type generally measure parameters such as the light scattering coefficient of tissue, the refractive index of interstitial fluid, and the acoustic propagation speed in tissue. [6][7][8][9] The optical measurements required for glucose determination are commonly performed using optical coherence tomography (OCT).…”