Optical polarimetry in the anterior chamber of the eye has emerged as a potential technique to non-invasively measure glucose levels for diabetes. Time varying corneal birefringence due to eye motion artifact confounds the optical signal ultimately limiting the polarimetric technique from accurately predicting glucose concentrations. In this study, a high speed dual-wavelength optical polarimetric approach was developed and in vitro phantom studies were performed with and without motion. The glucose concentrations ranged from 0-600 mg/dL at 100 mg/dL increments. The polarimeter produced glucose measurements with less than a 10 msec stabilization time and yielding standard errors of less than 10 mg/dL without motion and standard errors less than 26 mg/dL with motion. The results indicate a high speed dual-wavelength polarimetric approach has the potential to be used for non-invasive glucose measurements.
Optical polarimetry as a method to monitor glucose levels in the aqueous humor has shown promise as a way
to noninvasively ascertain blood glucose concentration. A major limiting factor to polarimetric approaches for
glucose monitoring in the aqueous humor is time varying birefringence due to motion artifact. Here, we present a
modulation approach for real-time polarimetry that is capable of glucose monitoring in vitro at optical modulation
frequencies of tens of kHz and includes the DC-compensation in a single device. Such higher frequency modulation
has the potential benefit of improving the signal-to-noise ratio of the system in the presence of motion artifacts. In
this report we present a near real-time closed-loop single wavelength polarimeter capable of glucose sensing in vitro
at an optical modulation frequency of 32 kHz. The single wavelength polarimetric setup and in vitro glucose
measurements will be presented demonstrating the sensitivity and accuracy of the system. Our PID control system
can reach stability in less than 10 ms which is fast enough to overcome motion artifact due to heart beat and
respiration. The the system can predict the glucose concentration with a standard error of less than 18.5 mg/dL and a
MARD of less than 6.65% over the physiologic glucose range of 0-600 mg/dL. Our results indicate that this optical
modulation approach coupled with dual-wavelength polarimetry has the potential to improve the of the dual-wavelength
approach for in vivo glucose detection applications.