A Prospective Pilot Study Demonstrating Noninvasive Calibration-Free Glucose Measurement
Authors' conclusion
Our findings suggest that Raman spectroscopy coupled with advanced computational methods can enable continuous, noninvasive glucose measurement without per-subject invasive calibration.
Study context
- Population
- Adults with insulin-treated diabetes; median HbA1c 7.3% (IQR 6.7–7.7)
- Participants
- 20 adults
- Exposure
- Development set vs Validation set
- Development setn = 15
- Validation setn = 5
- Outcome
- Calibration-free noninvasive glucose measurement by spectroscopy-based machine learning model
Methodological detailsexploratory, single-centre, development investigators blinded to validation analyses
- Exploratory
- Single-centre
- Development investigators blinded to validation analyses
Main finding
As reported in the abstract
Mean absolute relative difference (MARD)
14.5%
Paired glucose data points in A plus B zones of Diabetes Technology Society (DTS) error grid
96.5%
Observational data: associations, not effects.
Further findings
Correlation between average model sensitivity by wavelength and glucose spectrum
- Correlation (P < .001)
- 0.45
Original publication
A Prospective Pilot Study Demonstrating Noninvasive Calibration-Free Glucose Measurement. J Diabetes Sci Technol. Published 29 January 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →