REVIEW 4 major objections 4 minor 38 references
Glucose Levels Sensing using Whispering Gallery Modes at mm-Wave Band
T0 review · 4 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read A whispering-gallery-mode dielectric resonator, simulated at 49-70 GHz, detects 0.1 mg/mL glucose concentration steps in water through changes in its transmission coefficient.
desk verdict A credible WGM sensor design study with real dielectric measurements, but the headline sensitivity numbers are unvalidated HFSS predictions and the paper overstates what is demonstrated. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the whispering gallery mode (WGM) of a dielectric disc resonator: a high-order mode whose electric field is concentrated near the resonator boundary, making the resonance sensitive to material placed just outside it. The sensor couples this resonator to a curved alumina image waveguide carrying the dominant $E_{z11}$ mode from a rectangular waveguide feed; a plexiglass container holds the glucose sample on top of the disc. The argument runs through the Debye relaxation model, a standard frequency-dependent description of dielectric response, whose fitted coefficients for each glucose concentration supply the complex permittivity used in full-wave simulation, and through the critical-coupling condition, optimized by sweeping the waveguide-resonator gap, at which $S_{21}$ magnitude changes most sharply with glucose level.
What would settle it
Fabricate the proposed sensor and measure $|S_{21}|$ with a vector network analyzer for glucose solutions at 0.7, 0.8, 0.9, 1.0, 1.1, and 1.2 mg/mL. If the observed shift in $|S_{21}|$ at the WGH600 resonance is not near 0.76 dB per 0.1 mg/mL step and is not monotonic, the simulated sensitivity claim is contradicted. A simpler check: measure the complex permittivity of the same solutions at 60 GHz with a calibrated probe and compare it with the Debye model's prediction; a significant mismatch would invalidate the simulation's input data.
Extended reading notes
Core claim
The central claim is that a whispering-gallery-mode dielectric disc resonator, excited by a curved dielectric image waveguide, can translate small variations in the dielectric properties of a loaded glucose solution into measurable changes in the magnitude of $S_{21}$. Using a single-pole Debye model fitted to measured permittivity data for glucose-water solutions from 0.7 to 1.2 mg/mL, the paper simulates the sensor across 49-70 GHz and reports that the lower-order modes WGH600 and WGH700 give sensitivities of 0.077 and 0.025 dB/(mg/dL), respectively, with an $S_{21}$ change of 0.76 dB for a 0.1 mg/mL increase at WGH600. Higher-order modes such as WGH800 are far less sensitive, consistent with the evanescent field being too confined to sense external perturbations. The paper further demonstrates that a critical coupling gap between waveguide and resonator exists for each mode, and that loading the resonator shifts its resonance frequencies while glucose concentration only changes the transmission magnitude.
Load-bearing premise
The entire sensitivity claim rests on the assumption that the single-pole Debye model, with coefficients fitted to the measured glucose solutions, correctly predicts their electrical behaviour at every frequency in the 49-70 GHz band, and that the full-wave solver translates that behaviour into transmission values exactly as a physical device would.
Editorial extensions
If this is right
- If the simulated sensitivities are realized, a single $S_{21}$ magnitude measurement could resolve glucose concentration steps of 0.1 mg/mL in the 0.7 to 1.2 mg/mL range relevant to type 2 diabetes.
- The sensor's low-cost dielectric waveguide and disc construction, without metallic resonator structures, suggests a path to disposable or wearable mm-wave glucose monitors.
- The finding that lower-order WGH modes outperform higher-order modes provides a concrete design rule: operate at the first WGH resonances for maximum sensitivity.
- Because loading shifts resonance frequencies but glucose concentration only changes $|S_{21}|$ at fixed frequency, the sensor could separate the presence of a sample from its glucose content.
Reading between the lines
- A direct extension of this work would be to fabricate the resonator and measure $S_{21}$ for the same glucose solutions; the simulated 0.76 dB shift at WGH600 would then be checked against a physical VNA trace, which would also reveal how fabrication tolerances in the coupling gap affect sensitivity.
- Because the sensing mechanism relies on the evanescent field outside the resonator, the sensitivity likely degrades when a lossy medium such as skin is interposed; quantifying this degradation would clarify whether the sensor can work non-invasively on a finger or earlobe.
- The Debye coefficients come from measurements up to 67 GHz, while the sensor is simulated up to 70 GHz; extrapolating the fit beyond the measured band is a source of uncertainty that direct permittivity data across 49 to 70 GHz would remove.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes a millimeter-wave (49-70 GHz) whispering-gallery-mode (WGM) dielectric disk resonator coupled to a curved image waveguide for sensing glucose concentration in aqueous solutions. The authors measure the dielectric properties of glucose-water solutions (0.7-1.2 mg/ml) with a coaxial probe, fit each concentration to a single-pole Debye model, and use Ansys HFSS to simulate S21 responses of the resonator at five WGH modes. They report sensitivities of 0.077 and 0.025 dB/(mg/dL) at WGH600 and WGH700, respectively, under critical coupling conditions, and conclude that the proposed sensor is a reliable non-invasive glucose sensor. The central claim is that the lower-order WGM modes provide high sensitivity to small glucose-induced permittivity changes.
Significance. If the reported sensitivity values are correct, the proposed WGM resonator would represent a useful contribution to millimeter-wave non-invasive glucose sensing, with simulation-based design insight into mode-order-dependent sensitivity. The strengths of the manuscript include the use of actual dielectric measurements for the glucose solutions rather than assumed literature values, a forward simulation setup with explicitly reported geometric parameters, and a falsifiable claim in the form of quantitative sensitivity predictions. However, the significance is currently limited because the central sensitivity figures rest entirely on an unvalidated HFSS model and Debye fits with no reported uncertainties, and the results are not supported by any fabricated-device measurement or independent numerical check.
major comments (4)
- [Section II, Table I] The Debye coefficients in Table I are fitted to measured dielectric data, but the manuscript reports no fit residuals, no measurement uncertainty, and no comparison of the fitted model to the raw probe measurements. At approximately 60 GHz, the fitted parameters imply only small permittivity differences between adjacent glucose concentrations, yet the claimed S21 contrasts of 0.76 dB and 0.25 dB for 0.1 mg/ml steps must be resolved against this material-model uncertainty. Please provide goodness-of-fit metrics, error bars on the Debye parameters, and a sensitivity analysis showing that the S21 contrasts exceed the propagated material-property uncertainty.
- [Section III, Fig. 3] The sensitivity values in Table III are produced by an HFSS simulation with no reported mesh-convergence study, no comparison with an independent solver, and no measurement of a fabricated sensor, despite the Fig. 3 caption reading "Measurement and simulation results of S21." Since S21 magnitude variations of a few tenths of a decibel are the entire basis for the sensor claim, a mesh-refinement study and an uncertainty analysis, or at least one experimental validation, are needed to establish that the reported WGH600 and WGH700 sensitivities are physical predictions rather than numerical artifacts.
- [Section III, Table III and text] There is a direct internal inconsistency between the text and Table III for the WGH800 mode. The text states that for a 0.1 mg/ml glucose change the S21 varies by less than 0.04 dB for WGH800, but Table III reports a sensitivity of 0.0104 dB/(mg/dL), which for 0.1 mg/ml (10 mg/dL) predicts 0.104 dB, not less than 0.04 dB. This discrepancy suggests that the reported contrasts were not cross-checked against the tabulated sensitivities and must be resolved before the quantitative claims can be trusted.
- [Sections II and Conclusion] The paper generalizes from aqueous glucose solutions to blood glucose sensing, but the sole justification is that water constitutes roughly 50% of blood volume. Blood contains proteins, cells, and salts that affect millimeter-wave permittivity, and the cited supporting measurements in [5] and [33] use saline or physiological solutions, not blood. The conclusion that the sensor is a "reliable non-invasive mm-wave integrated glucose bio-sensor" for blood is not supported by the simulations, which only model aqueous glucose. Please either temper the claims to aqueous-solution sensing or provide evidence that the dielectric response of blood in this band is dominated by the glucose-water contribution.
minor comments (4)
- [Abstract and throughout] The manuscript mixes mg/ml and mg/dL units without stating the conversion; since 0.1 mg/ml equals 10 mg/dL, please use a single unit system or explicitly define the conversion at first use.
- [Section II, Fig. 1] The text describing Fig. 1 mentions an exponential decrease in dielectric constant and a near-linear increase in loss tangent up to 60 GHz, but the figure itself is not visible in the manuscript and no axis labels or legends are described; please ensure the figure is readable and the measurement conditions are fully specified.
- [Section II, after Table I] The single-pole Debye equation is not written out; please define the model explicitly, including which of the fitted coefficients (epsilon_inf, epsilon_s, tau) corresponds to the standard Debye relaxation formula.
- [Section III, Fig. 3 caption] The caption "Measurement and simulation results of S21" is misleading because no measurement procedure or experimental data are described anywhere in the paper; if these are purely simulation results, the caption should say so.
Circularity Check
No significant circularity: the sensitivity values are forward HFSS simulation outputs driven by independently fitted Debye material parameters, not by construction equal to any input.
full rationale
The paper's derivation chain is a forward electromagnetic simulation: measured dielectric data for glucose solutions are fitted to single-pole Debye models (Table I), and those permittivity values are then used as material inputs to an HFSS model of the WGM resonator to compute S21 magnitude shifts. The reported sensitivity figures in Table III (e.g., 0.077 dB/(mg/dL) for WGH600) are outputs of that simulation, not quantities that were fitted or assumed. The coupling gap g is tuned by parametric sweep to obtain critical coupling, but this tuning optimizes resonance coupling rather than targeting a particular sensitivity value, so the sensitivity is not forced by the optimization. The Debye coefficients are extracted from independent coaxial-probe measurements and are not adjusted to reproduce the simulated S21 contrasts. The only self-citation to prior work by the same group, reference [34], supports the permittivity measurement step, and that measurement is an input rather than a conclusion derived from the WGM sensor model. No equation in the paper is shown to reduce to another by construction, no fitted parameter is renamed as a prediction, and no load-bearing uniqueness theorem or ansatz is imported from the authors' own prior work. The absence of mesh-convergence checks, fit residuals, or fabricated-device validation is a correctness and reporting risk about the reliability of the simulation, not a circularity in the logical derivation. Therefore the paper is not circular.
Assumptions & free parameters
free parameters (4)
- Debye coefficient set (eps_inf, eps_s, tau) per glucose concentration =
0.7 mg/mL: 5.67, 80.65, 9.42 ps; 0.8: 5.76, 81.028, 9.49; 0.9: 5.92, 81.95, 9.65; 1.0: 6.13, 82.95, 9.82; 1.1: 6.147…
- Waveguide-resonator coupling gap g =
WGH600: 0.42 mm; WGH700: 0.45 mm; WGH800: 0.45 mm
- Plexiglass container bottom thickness t2 =
1.5 mm
- Resonator and waveguide dimensions (R, h, guide cross-section, etc.) =
As in Table II (e.g., R=3.42 mm, h=0.5 mm, guide 1x0.8 mm2)
assumptions (3)
- domain assumption Single-pole Debye model accurately represents the dielectric response of glucose solutions in the 49-70 GHz band.
- domain assumption HFSS full-wave solver produces trustworthy S-parameters for the designed structure.
- domain assumption Aqueous glucose solutions are valid mimics for blood glucose sensing.
Cite this review
Pith. "Pith review of Glucose Levels Sensing using Whispering Gallery Modes at mm-Wave Band." pith.science (2026). https://pith.science/paper/XGFI5IT4
@misc{pith2026190912388,
author = {Pith},
title = {Pith review of: Glucose Levels Sensing using Whispering Gallery Modes at mm-Wave Band},
year = {2026},
howpublished = {\url{https://pith.science/paper/XGFI5IT4}},
note = {Machine review of arXiv:1909.12388}
}
read the original abstract
In this study, an integrated low-cost and complexity mm-wave structure of Whispering Gallery Mode (WGM) is proposed for sensing the glucose levels in mimicking aquatic solutions of concentrations similar to type 2 diabetics. The bio-sensor is composed of a curved dielectric waveguide coupled to a dielectric disc resonator that is loaded with the glucose sample under test. The intense WGM field induced towards the boundary of the DDR is exploited to detect the slight variations in the dielectric properties of different glucose levels via tracing the variations in the magnitude of the transmission coefficient S_{21} in the mm-wave frequency band (49-70 GHz). The WGM resonator shows a high sensitivity performance (0.025-0.077 dB/(mg/dL)) at the lower-order modes WGH_{600} and WGH_{700} as demonstrated by simulations in a 3D full-wave EM solver (Ansys HFSS).
Figures
Figures from the paper (1 more)
Reference graph
Works this paper leans on
-
[5]
H. Cano-Garcia, P. Kosmas, I. Sotiriou, I. Papadopoulos-Kelidis, C. Parini, I. Gouzouasis, G. Palikaras, and E. Kallos, “Detection of glucose variability in saline solutions from transmission and reflection measurements using V-band waveguides,” Measurement Science and Technology, vol. 26, no. 12, pp. 125701-125710, Dec 2015
work page 2015
-
[33]
Meriakri, V. V., E. E. Chigrai, D. Kim, I. P. Nikitin, L. I. Pangonis, M. P. Parkhomenko, and J. H. Won, “Dielectric properties of glucose solutions in the millimetre-wave range and control of glucose content in blood,” Measurement science and technology, vol. 18, no. 4, pp. 977–982, 2007
work page 2007
-
[1]
Millimeter-wave interactions with the human body: State of knowledge and recent advances,
Zhadobov, Maxim, et al. “Millimeter-wave interactions with the human body: State of knowledge and recent advances,” International Journal of Microwave and Wireless Technologies (2011)
work page 2011
-
[2]
Sensing liquid properties using split-ring resonator in Mm-wave band,
M. Abidi, A. Elhawil, J. Stiens, R. Vounchx, J. B. Tahar and F. Choubani, “Sensing liquid properties using split-ring resonator in Mm-wave band,” IECON 2010 – 36th Annual Conference on IEEE Industrial Electronics Society, Glendale, AZ, 2010, pp. 1298-1301
work page 2010
-
[3]
Diagnosis and classification of diabetesmellitus,
Amer. Diabetes Assoc., “Diagnosis and classification of diabetesmellitus,” Diabetes Care, vol. 37, pp. S81–S90, 2014
work page 2014
-
[4]
Self-monitoring of blood glucose,
Canadian Diabetes Association, “Self-monitoring of blood glucose,” Jan. 12, 2018. [Online]. Available at http://www.diabetes.ca/clinical- practiceeducation/professional-resources/selfmonitoring-of-blood glucose
work page 2018
-
[6]
D. Cunningham and J. Stenken, In vivo glucose sensing, Hoboken, N.J: Wiley, 2010
work page 2010
-
[7]
Diabetes Technology & Therapeutics
Sieg A, Guy RH, Delgado-Charro MB, Noninvasive and minimally invasive methods for transdermal glucose monitoring. Diabetes Technology & Therapeutics. 7(1), 174-97 (Feb 2005)
work page 2005
Show all 38 references
-
[8]
Diabetes Technology & Therapeutics
Weiss R, Yegorchikov Y, Shusterman A, Raz I, Noninvasive continuous glucose monitoring using photoacoustic technology—results from the first 62 subjects. Diabetes Technology & Therapeutics. 9(1), 68-74 (Feb 2007)
2007
-
[9]
Diabetes Care
Garg S, Zisser H, Schwartz S, Bailey T, Kaplan R, Ellis S, Jovanovic L, Improvement in glycemic excursions with a transcutaneous, real-time continuous glucose sensor: a randomized controlled trial. Diabetes Care. 29(1), 44-50 (Jan 2006)
2006
-
[10]
Continuous Noninvasive Glucose Monitoring Technology Based on ‘Occlusion Spectroscopy
Amir, Orna, et al. Continuous Noninvasive Glucose Monitoring Technology Based on ‘Occlusion Spectroscopy. Journal of Diabetes Science and Technology, vol. 1(4), 463–469 (July 2007)
2007
-
[11]
S. K. Vashist, Non-invasive glucose monitoring technology in diabetes management: A review. Analytica Chimica Acta (750), 16–27 (2012)
2012
-
[12]
Dielectric Properties of Water Solutions with Small Content of Glucose in the Millimeter-Wave Band and the Determination of Glucose in Blood,
V. V. Meriakri, E. E. Chigrai, I. P. Nikitin and M. P. Parkhomenko, “Dielectric Properties of Water Solutions with Small Content of Glucose in the Millimeter-Wave Band and the Determination of Glucose in Blood,” International Kharkov Symposium Physics and Engrg. of Millimeter ...
2007
-
[13]
Karacolak T, Moreland E C and Topsakal E 2013 Cole-Cole model for glucose-dependent dielectric properties of blood plasma for continuous glucose monitoring Microw. Opt. Technol. Lett. 55 1160–4
2013
-
[14]
Dobson R, Wu R and Callaghan P 2012 Blood glucose monitoring using microwave cavity perturbation Electron. Lett. 48 905–6
2012
-
[15]
Yun F, Xiaoguang D, Qing W and WeiLian W 2010 Testing glucose concentration in aqueous solution based on microwave cavity perturbation technique BMEI 2010: 3rd Int. Conf. on Biomedical Engineering and Informatics pp 1046–9
2010
-
[16]
Wiwatwithaya S, Phasukkit P, Tungjitkusolmun S and Wongtrairat W 2011 Real-time monitoring glucose by used microwave antenna apply to biosensor BMEiCON 2011: Biomedical Engineering Int. Conf. pp 135–7
2011
-
[17]
Workshop on Medical Measurements and Applications Proc
Hofmann M, Fersch T, Weigel R, Fischer G and Kissinger D 2011 A novel approach to non-invasive blood glucose measurement based on RF transmission MeMeA 2011: IEEE Int. Workshop on Medical Measurements and Applications Proc. pp 39–42
2011
-
[18]
pp 546–9
Hofmann M, Bloss M, Weigel R, Fischer G and Kissinger D 2012 Non- invasive glucose monitoring using open electromagnetic waveguides EuMC 2012 42nd European Microwave Conf. pp 546–9
2012
-
[19]
Hofmann M, Fischer G, Weigel R and Kissinger D 2013 Microwave- based noninvasive concentration measurements for biomedical applications IEEE Trans. Microw. Theory Tech. 61 2195–2204
2013
-
[20]
Lee K, Babajanyan A, Kim C, Kim S and Friedman B 2008 Glucose aqueous solution sensing by a nearfield microwave microprobe Sensors Actuators A 148 28–32
2008
-
[21]
Microwave dielectric resonator biosensor for aqueous glucose solution,
J. Kim et al., "Microwave dielectric resonator biosensor for aqueous glucose solution," Review of Scientific Instruments, vol. 79, no. 8, August 2008
2008
-
[22]
Design and In Vitro Interference Test of Microwave Noninvasive Blood Glucose Monitoring Sensor,
H. Choi, J. Naylon, S. Luzio, J. Beutler, J. Birchall, C. Martin, and A. Porch, “Design and In Vitro Interference Test of Microwave Noninvasive Blood Glucose Monitoring Sensor,” IEEE Transactions on Microwave Theory and Techniques, vol. 63, no. 10, pp. 3016–3025, October 2015....
2015
-
[23]
Reflection and transmission measurements using 60 GHz patch antennas in the presence of animal tissue for non-invasive glucose sensing,
H. Cano-Garcia et al., "Reflection and transmission measurements using 60 GHz patch antennas in the presence of animal tissue for non-invasive glucose sensing," 2016 10th European Conference on Antennas and Propagation (EuCAP), Davos, 2016, pp. 1-3. doi: 10.1109/EuCAP.2016.7481178
2016
-
[24]
A Glucose Sensing System Based on Transmission Measurements at Millimetre Waves using Micro strip Patch Antennas,
S. Saha et al. “A Glucose Sensing System Based on Transmission Measurements at Millimetre Waves using Micro strip Patch Antennas,” Scientific Reports, vol. 7, no. 1, 2017. doi:10.1038/s41598-017-06926-1
2017 doi
-
[25]
Millimeter-wave non-invasive monitoring of glucose in anesthetized rats,
P. H. Siegel, Y. Lee, and V. Pikov, "Millimeter-wave non-invasive monitoring of glucose in anesthetized rats," 2014 39th International Conference on Infrared, Millimeter, and Terahertz waves (IRMMW- THz), Tucson, AZ, 2014, pp. 1-2. doi: 10.1109/IRMMW- THz.2014.6956294
2014
-
[26]
Non- invasive glucose monitoring using open electromagnetic waveguides,
M. Hofmann, M. Bloss, R. Weigel, G. Fischer, and D. Kissinger, "Non- invasive glucose monitoring using open electromagnetic waveguides," 2012 42nd European Microwave Conference, Amsterdam, 2012, pp. 546-
2012
-
[27]
Detection of glucose variability in saline solutions from transmission and reflection measurements using V-band waveguides,
H. Cano-Garcia et al., “Detection of glucose variability in saline solutions from transmission and reflection measurements using V-band waveguides,” Measurement Science & Technology, vol. 26, no. 12, pp. 125701-125710, 2015. doi:10.1088/0957-0233/26/12/125701
2015 doi
-
[28]
IEEE Trans
Annino, G.; Cassettari, M.; Longo, I.; Martinelli, M.: Whispering gallery modes in a dielectric resonator: characterization at millimeter wavelength. IEEE Trans. Microw. Theory Tech., 45 (1997), 2025–2034
1997
-
[29]
Creedon, D.L.; Reshitnyk, Y.; Farr, W.; Martinis, J.M.; Duty, T.L.; Tobar, M.E.: High Q-factor sapphire whispering gallery mode microwave resonator at single photon energies and millikelvin temperature. Appl. Phys. Lett., 98 (2011), 222903(3)
2011
-
[30]
Fundamentals of medicalsurgical nursing,
A. Brady, C. McCabe and M. McCann, “Fundamentals of medicalsurgical nursing,” John Wiley & Sons, 2013
2013
-
[31]
Electromagnetic properties of tissue in the optical region,
K. Yaws, D. Mixon, and W. Roach, “Electromagnetic properties of tissue in the optical region,” in Biomedical Optics (BiOS). International Society for Optics and Photonics, 2007, pp. 507–643
2007
-
[32]
Millimeter-wave non-invasive monitoring of glucose in anesthetized rats,
P. H. Siegel, Y. Lee, and V. Pikov, "Millimeter-wave non-invasive monitoring of glucose in anesthetized rats," 2014 39th International Conference on Infrared, Millimeter, and Terahertz waves (IRMMW- THz), Tucson, AZ, 2014, pp. 1-2
2014
-
[34]
EM Measurements of Glucose-Aqueous Solutions
A. E. Omer, G. Shaker, S. Safavi-Naeini, and R. M. Shubair, “EM Measurements of Glucose-Aqueous Solutions”, 2019 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting
2019
-
[35]
Microstrip Line-Based Glucose Sensor for Noninvasive Continuous Monitoring Using the Main Field for Sensing and Multivariable Crosschecking,
S. Y. Huang et al., “Microstrip Line-Based Glucose Sensor for Noninvasive Continuous Monitoring Using the Main Field for Sensing and Multivariable Crosschecking,” in IEEE Sensors Journal, vol. 19, no. 2, pp. 535-547, 15 Jan.15, 2019
2019
-
[36]
Temperature- Corrected Fluidic Glucose Sensor Based on Microwave Resonator. Sensors,
C. Jang, J.-K. Park, H.-J. Lee, G.-H. Yun, J.-G. Yook, “Temperature- Corrected Fluidic Glucose Sensor Based on Microwave Resonator. Sensors,” Sensors 2018, 18(11), 3850
2018
-
[37]
Millimeter-wave Adaptive Glucose Concentration Estimation with Complex-Valued Neural Networks,
S. Hu, S. Nagae and A. Hirose, “Millimeter-wave Adaptive Glucose Concentration Estimation with Complex-Valued Neural Networks,” in IEEE Transactions on Biomedical Engineering. doi: 10.1109/TBME.2018.2883085
2018
-
[549]
doi: 10.23919/EuMC.2012.6459152
2012
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.