Critical-depth Raman spectroscopy enables home-use non-invasive glucose monitoring

Abstract

One of the most ambitious endeavors in the field of diabetes technology is non-invasive glucose sensing. In the past decades, a number of different technologies have been assessed, but none of these have found its entry into general clinical use. We report on the development of a table-top confocal Raman spectrometer that was used in the home of patients with diabetes and operated for extended periods of time unsupervised and without recalibration. The system is based on measurement of glucose levels at a 'critical depth' in the skin, specifically in the interstitial fluid located below the stratum corneum but above the underlying adipose tissue layer. The region chosen for routine glucose measurements was the base of the thumb (the thenar). In a small clinical study, 35 patients with diabetes analyzed their interstitial fluid glucose for a period of 60 days using the new critical-depth Raman (CD-Raman) method and levels were correlated to reference capillary blood glucose values using a standard finger-stick and test strip product. The calibration of the CD-Raman system was stable for > 10 days. Measurement performance for glucose levels present at, or below, a depth of ~250μm below the skin surface was comparable to that reported for currently available invasive continuous glucose monitors. In summary, using the CD-Raman technology we have demonstrated the first successful use of a non-invasive glucose monitor in the home.

Fig 1. Schematic presentation of portable Raman spectrometer.

The output from the diode laser (300mW continuous power) is first collimated, unwanted spectral side lobes and fluorescence are filtered, then redirected by the dichroic mirror and finally focused (by the f/0.55 lens) just below the skin surface. Whilst both the intense reflected/scattered light, fluorescence, and generated Raman photons are collected by the f/0.55 lens, the dichroic mirror and the long-pass filter ensure that only the latter two contributions to the spectrum are focused (by the f/1.3 lens) on the entrance slit of the spectrometer.

Fig 2. Simulation of signal acquisition density distribution.

Panels (A)-(D) show the simulated density distribution of the generated Raman signal in the skin for collection depths of 150, 200, 250, and 300μm, respectively. The skin is modelled as a turbid medium featuring both bulk and surface scattering.

Fig 3. Box plot of performance versus collection depth.

35 patients were grouped into three depth ranges as indicated: 140–200 um, n = 14; 240–270 um, n = 12; 280–320 um, n = 9. The box shows upper and lower quartile and median. The threshold of 0.5 was supported by the CV results and the grading of glucose similarity for the regression vector, above this value the performance was satisfactory.

Fig 4. Independent validation of the last five days of analysis.

The measured glucose concentrations are plotted as a function of reference values in a consensus error grid. The measurement critical depth was 240–320 um and the ISUP was greater than 0.5.

Fig 5. Raman spectrum and independent predictions for an average-performing patient.

(A) Average Raman spectrum from the thenar region (blue), Raman spectrum of glucose in an aqueous solution (red), and regression vector of PLS calibration model (black). (B) Plot of the independently predicted glucose concentration and the reference value as a function of time from the last day of calibration. The dashed lines are for guidance only.