pith. sign in

arxiv: 2605.02674 · v1 · submitted 2026-05-04 · 🧮 math.NA · cs.NA

Parameter estimation for evaporation-driven tear film model in two space dimensions

Qinying Chen , Tobin Driscoll This is my paper

Pith reviewed 2026-05-08 18:58 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords tear filmtear breakupparameter estimationevaporationproper orthogonal decompositionfluorescence imagingdry eye
0
0 comments X

The pith

Fitting an evaporation-driven tear film model to fluorescence data provides in vivo estimates of thinning parameters in tear breakup regions.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper applies a two-dimensional evaporation-driven tear film thinning model, made computationally efficient through proper orthogonal decomposition, to match experimental fluorescence images taken from the eyes of normal subjects. The match produces specific estimates for how fast evaporation occurs and how the film thins in areas prone to breakup. A reader would care because these numbers give a measurable picture of tear film behavior in living healthy eyes, which can serve as a reference when studying dry eye disease where breakup is more problematic. The dimension reduction step is key because it allows the model equations to be solved quickly enough for repeated fitting against the image sequences.

Core claim

By extending the evaporation-driven tear film model to two space dimensions using proper orthogonal decomposition and fitting the reduced model to in vivo fluorescence imaging data from normal subjects, estimates of evaporation-related and thinning parameters within tear breakup regions can be obtained. These findings enhance understanding of tear film thinning and dry-spot formation and establish a quantitative baseline for comparison with dry eye patient data.

What carries the argument

The proper orthogonal decomposition reduced-order model of the evaporation-driven tear film thinning equations in two dimensions, which enables fitting to experimental fluorescence data for parameter estimation.

If this is right

  • Estimates of evaporation and thinning parameters are derived from real imaging data of normal eyes.
  • A quantitative baseline is created for tear film dynamics in healthy subjects.
  • The method supports future comparisons with data from dry eye patients.
  • Dimension reduction makes parameter estimation from PDE models feasible on imaging data.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • These parameter values might be tested in independent experiments measuring actual evaporation rates on the ocular surface.
  • The approach could be applied to time-lapse data from other thin liquid films in biology to estimate similar physical constants.
  • If the parameters differ substantially in dry eye subjects, the model might help identify specific causes of increased breakup.

Load-bearing premise

The evaporation-driven tear film thinning model and its POD reduction accurately represent the observed fluorescence dynamics in vivo without significant unmodeled effects or data-specific biases.

What would settle it

Direct measurement of evaporation rates or film thickness changes in the tear breakup regions of the same normal subjects, followed by comparison to the fitted model predictions, would test whether the estimates hold.

read the original abstract

The tear film (TF) plays a critical role in maintaining ocular surface health, and its disruption through tear breakup (TBU) is closely associated with dry eye disease. Evaporation-driven thinning is a primary mechanism underlying TBU, yet quantitative in vivo estimates of key physical parameters remain limited. In this work, we fit an evaporation-driven TF thinning model, originally developed by Braun et al. and extended to two dimensions using proper orthogonal decomposition (POD) by Chen et al., to experimental fluorescence (FL) imaging data from normal subjects. The use of dimension reduction enables efficient solution of the governing PDEs and facilitates parameter estimation from imaging data. Our results provide in vivo estimates of evaporation-related and thinning parameters within TBU regions. These findings enhance understanding of TF thinning and dry-spot formation and establish a quantitative baseline for comparison with dry eye patient data.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript fits an evaporation-driven tear film thinning model (originally from Braun et al. and extended to 2D via POD by Chen et al.) to in vivo fluorescence imaging data from normal subjects. The goal is to obtain estimates of evaporation-related and thinning parameters within tear breakup (TBU) regions, leveraging the POD reduction for computational efficiency in solving the governing PDEs and performing the inverse problem.

Significance. If the POD reduction is shown to be accurate in the fitted parameter regime and the estimation procedure includes proper validation and uncertainty quantification, the work would deliver valuable quantitative in vivo baselines for TF thinning dynamics. This could meaningfully support comparisons with dry-eye data and advance mechanistic understanding of TBU. The approach builds directly on established prior models, but its significance is currently constrained by missing verification steps.

major comments (2)
  1. [POD reduction section] POD reduction section: No a-posteriori error bounds, residual comparisons, or direct numerical checks are reported between the POD-reduced model and the full 2D PDE at the estimated evaporation and thinning parameter values. If truncation error is comparable to data misfit inside TBU regions, the inverse problem can return systematically shifted estimates even when the reduced-model residual appears small; this verification is load-bearing for the central claim of reliable in vivo parameter estimates.
  2. [Parameter estimation and results sections] Parameter estimation and results sections: The reported estimates lack error bars, confidence intervals, goodness-of-fit metrics (e.g., residual norms or R²), data exclusion criteria, or validation on held-out images. Without these, the quality and robustness of the fits to the fluorescence data cannot be assessed, undermining the claim that the procedure yields trustworthy in vivo values.
minor comments (2)
  1. [Abstract] Abstract: The statement that 'our results provide in vivo estimates' would be strengthened by briefly indicating the order of magnitude or range of the obtained parameter values.
  2. [Notation and figures] Notation and figures: Ensure that all symbols for evaporation and thinning rates are defined consistently between the model equations and the fitting procedure; consider adding a table summarizing the final estimated values with uncertainties.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the constructive feedback. The comments highlight important verification steps that will improve the manuscript's rigor. We respond to each major comment below and outline the planned revisions.

read point-by-point responses

  1. Referee: [POD reduction section] POD reduction section: No a-posteriori error bounds, residual comparisons, or direct numerical checks are reported between the POD-reduced model and the full 2D PDE at the estimated evaporation and thinning parameter values. If truncation error is comparable to data misfit inside TBU regions, the inverse problem can return systematically shifted estimates even when the reduced-model residual appears small; this verification is load-bearing for the central claim of reliable in vivo parameter estimates.

    Authors: We agree that explicit verification of the POD approximation at the fitted parameter values is necessary to ensure the inverse problem is not biased by truncation error. The current manuscript relies on the a priori error analysis and numerical tests from Chen et al. (the POD extension paper) but does not repeat direct residual comparisons or a-posteriori bounds specifically at the estimated evaporation and thinning rates. We will add a dedicated subsection (or appendix) containing: (i) L2-norm residuals between the reduced-order and full-order solutions evaluated at the reported parameter values, (ii) comparison of fluorescence intensity profiles inside representative TBU regions, and (iii) any feasible a-posteriori error indicators. These additions will confirm that the POD truncation error is small relative to the data misfit. revision: yes

  2. Referee: [Parameter estimation and results sections] Parameter estimation and results sections: The reported estimates lack error bars, confidence intervals, goodness-of-fit metrics (e.g., residual norms or R²), data exclusion criteria, or validation on held-out images. Without these, the quality and robustness of the fits to the fluorescence data cannot be assessed, undermining the claim that the procedure yields trustworthy in vivo values.

    Authors: We accept that the current presentation of results is incomplete. We will augment the parameter estimation and results sections with: (i) the final least-squares residual norms and R² values for each fit, (ii) explicit data exclusion criteria used to identify TBU regions from the fluorescence sequences, and (iii) approximate confidence intervals obtained from the Hessian of the objective function (or bootstrap resampling where computationally feasible). For validation, we will implement and report leave-one-out cross-validation on the available normal-subject sequences. However, because the study uses all collected fluorescence data to obtain the in vivo estimates, a fully independent held-out test set is not available without new experiments. revision: partial

standing simulated objections not resolved
  • Validation on fully independent held-out image sequences, as the present study employs the entire limited collection of normal-subject fluorescence data for parameter fitting.

Circularity Check

0 steps flagged

No significant circularity in derivation or results

full rationale

The paper performs parameter estimation by fitting an existing evaporation-driven thinning model (cited from Braun et al.) and its POD reduction (cited from Chen et al.) to new experimental fluorescence data. The central output is the resulting parameter values, which is the explicit goal of the work rather than a derived prediction or first-principles result. No load-bearing step reduces a claimed outcome to its own inputs by construction, self-definition, or unverified self-citation chain; the fitting process is standard inverse modeling applied to external data. Minor self-citation exists for the method but is not load-bearing for the estimates themselves.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The central claim rests on the validity of the pre-existing evaporation-driven PDE model and the accuracy of the POD dimension reduction; both are imported from earlier papers rather than re-derived or independently validated here. No new physical entities are postulated.

free parameters (2)
  • evaporation-related parameters
    Fitted to match fluorescence intensity decay in TBU regions; exact symbols and values not stated in abstract.
  • thinning parameters
    Additional parameters adjusted during the fitting procedure to reproduce observed tear-film dynamics.
axioms (2)
  • domain assumption The evaporation-driven thinning model developed by Braun et al. correctly captures the dominant physics of tear-film breakup in vivo.
    The paper adopts this model without modification or re-derivation.
  • domain assumption Proper orthogonal decomposition yields a sufficiently accurate reduced-order model for both solving the PDEs and performing parameter estimation.
    POD is used to enable efficient 2D computations; its truncation error is assumed not to affect the fitted parameter values materially.

pith-pipeline@v0.9.0 · 5439 in / 1445 out tokens · 46315 ms · 2026-05-08T18:58:19.490348+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

55 extracted references · 55 canonical work pages

  1. [1]

    Doane, M.G.: Interaction of eyelids and tears in corneal wetting and the dynamics of the normal human eyeblink. Am. J. Ophthalmol.89, 507–516 (1980)

    work page 1980

  2. [2]

    Lemp, MA.: The definition and classification of dry eye disease: Report of the Def- inition and Classification Subcommittee of the International Dry Eye WorkShop (2007). Ocul. Surf.5(2), 75–92 (2007)

    work page 2007

  3. [3]

    Willcox, M.D.P., Arg¨ ueso, P., Georgiev, G.A., Holopainen, J.M., Laurie, G.W., Millar, T.J., Papas, E.B., Rolland, J.P., Schmidt, T.A., Stahl, U., Suarez, T., Subbaraman, L.N., U¸ cakhan, O.¨O., Jones, L.: TFOS DEWS II Tear Film Report. Ocul. Surf.15(3), 366–403 (2017)

    work page 2017

  4. [4]

    The ocular surface16(1), 4–30 (2018)

    King-Smith, P.E., Begley, C.G., Braun, R.J.: Mechanisms, imaging and structure of tear film breakup. The ocular surface16(1), 4–30 (2018)

    work page 2018

  5. [5]

    Stapleton, F., Alves, M., Bunya, V.Y., Jalbert, I., Lekhanont, K., Malet, F., Na, K.-S., Schaumberg, D., Uchino, M., Vehof, J., Viso, E., Vitale, S., Jones, L.: TFOS DEWS-II epidemiology report. Ocul. Surf.15(3), 334–365 (2017)

    work page 2017

  6. [6]

    Nelson, J.D., Craig, J.P., Akpek, E.K., Azar, D.T., Belmonte, C., Bron, A.J., Clayton, J.A., Dogru, M., Dua, H.S., Foulks, G.N., Gomes, J.A.P., Hammitt, K.M., Holopainen, J., Jones, L., Joo, C.-K., Liu, Z., Nichols, J.J., Nichols, K.K., Novack, G.D., Sangwan, V., Stapleton, F., Tomlinson, A., Tsubota, K., Willcox, M.D.P., Wolffsohn, J.S., Sullivan, D.A.: ...

    work page 2017

  7. [7]

    Lin, H., Yiu, S.C.: Dry eye disease: A review of diagnostic approaches and treatments. Saudi J. Ophthalmol.28(3), 173–181 (2014)

    work page 2014

  8. [8]

    Baudouin, C., Aragona, P., Messmer, E.M., Tomlinson, A., Calonge, M., Boboridis, K.G., Akova, Y.A., Geerling, G., Labetoulle, M., Rolando, M.: Role of hyperosmolarity in the pathogenesis and management of dry eye disease: Proceedings of the OCEAN group meeting. Ocul. Surf.11, 246–258 (2013)

    work page 2013

  9. [9]

    Braun, R.J., Driscoll, T.A., Begley, C.G., King-Smith, P.E., Siddique, J.I.: On tear film breakup (TBU): Dynamics and imaging. Math. Med. Biol.: A Journal of the IMA35(2), 145–180 (2018)

    work page 2018

  10. [10]

    Braun, R.J., King-Smith, P.E., Begley, C.G., Li, L., Gewecke, N.R.: Dynamics and function of the tear film in relation to the blink cycle. Prog. Ret. Eye Res. 45, 132–164 (2015)

    work page 2015

  11. [11]

    Holly, F.J.: Formation and rupture of the tear film. Exp. Eye Res.15(5), 515–525 (1973) 20

    work page 1973

  12. [12]

    King-Smith, P.E., Fink, B.A., Hill, R.M., Koelling, K.W., Tiffany, J.M.: The thickness of the tear film. Curr. Eye Res.29(4-5), 357–368 (2004)

    work page 2004

  13. [13]

    Mishima, S., Maurice, D.M.: The oily layer of the tear film and evaporation from the corneal surface. Exp. Eye Res.1, 39–45 (1961)

    work page 1961

  14. [14]

    Gipson, I.K.: Distribution of mucins at the ocular surface. Exp. Eye Res.78(3), 379–388 (2004)

    work page 2004

  15. [15]

    Dartt, D.A.: Neural regulation of lacrimal gland secretory processes: Relevance in dry eye diseases. Prog. Ret. Eye Res.28(3), 155–177 (2009)

    work page 2009

  16. [16]

    Braun, R.J.: Dynamics of the Tear Film. Annu. Rev. Fluid Mech.44(1), 267–297 (2012)

    work page 2012

  17. [17]

    Acta Ophthalmol.48(1), 108–118 (1970)

    Norn, M.S.: Micropunctate Fluorescein Vital Staining of the Cornea. Acta Ophthalmol.48(1), 108–118 (1970)

    work page 1970

  18. [18]

    Cho, P., Brown, B., Chan, I., Conway, R., Yap, M.: Reliability of the tear break- up time technique of assessing tear stability and the locations of the tear break-up in Hong Kong Chinese. Optom. Vis. Sci.69(11), 879–885 (1992)

    work page 1992

  19. [19]

    Iranian J

    Dibajnia, P., Mohammadinia, M., Moghadasin, M., Amiri, M.A.: Tear Film Break-up Time in Bipolar Disorder. Iranian J. Psych.7(4), 191–193 (2012)

    work page 2012

  20. [20]

    Peng, C.-C., Cerretani, C., Braun, R.J., Radke, C.J.: Evaporation-driven insta- bility of the precorneal tear film. Adv. Coll. Interface Sci.206, 250–264 (2014)

    work page 2014

  21. [21]

    Zhong, L., Braun, R.J., Begley, C.G., King-Smith, P.E.: Dynamics of Fluorescent Imaging for Rapid Tear Thinning. Bull. Math. Biol.81(1), 39–80 (2019)

    work page 2019

  22. [22]

    Luke, R.A., Braun, R.J., Driscoll, T.A., Begley, C.G., Awisi-Gyau, D.: Parameter Estimation for Evaporation-Driven Tear Film Thinning. Bull. Math. Biol.82(6), 71 (2020)

    work page 2020

  23. [23]

    Luke, R.A., Braun, R.J., Driscoll, T.A., Awisi-Gyau, D., Begley, C.G.: Parameter Estimation for Mixed-Mechanism Tear Film Thinning. Bull. Math. Biol.83(5), 56 (2021)

    work page 2021

  24. [24]

    Luke, R.A., Braun, R.J., Driscoll, T.A., Sinopoli, D., Yawatkar, V., You, L., Phatak, A., Begley, C.G.: Fitting Simplified Models to Machine Learning- Identified Tear Film Breakup. Invest. Ophthalmol. Vis. Sci.62(8), 1315–1315 (2021)

    work page 2021

  25. [25]

    Driscoll, T.A., Braun, R.J., Luke, R.A., Sinopoli, D., Phatak, A., Dorsch, J., Begley, C.G., Awisi-Gyau, D.: Fitting ODE models of tear film breakup. Model. Artif. Intell. Ophthalmol.5, 1–36 (2023) 21

    work page 2023

  26. [26]

    King-Smith, P.E., Kimball, S.H., Nichols, J.J.: Tear film interferometry and corneal surface roughness. Invest. Ophthalmol. Vis. Sci.55(4), 2614–2618 (2014)

    work page 2014

  27. [27]

    King-Smith, P.E., Hinel, E.A., Nichols, J.J.: Application of a novel interferometric method to investigate the relation between lipid layer thickness and tear film thinning. Invest. Ophthalmol. Vis. Sci.51(5), 2418–2423 (2010)

    work page 2010

  28. [28]

    Wang, J., Fonn, D., Simpson, T.L., Jones, L.: Precorneal and Pre- and Postlens Tear Film Thickness Measured Indirectly with Optical Coherence Tomography. Invest. Ophthalmol. Vis. Sci44(6), 2524 (2003)

    work page 2003

  29. [29]

    Webber, W.R., Jones, D.P.: Continuous fluorophotometric method of measuring tear turnover rate in humans and analysis of factors affecting accuracy. Med. Biol. Eng. Comput.24(4), 386–392 (1986)

    work page 1986

  30. [30]

    Archives of Ophthalmology102(3), 410–412 (1984)

    Benedetto, D.A., Clinch, T.E., Laibson, P.R.: In Vivo Observation of Tear Dynamics Using Fluorophotometry. Archives of Ophthalmology102(3), 410–412 (1984)

    work page 1984

  31. [31]

    Begley, C.G., Simpson, T.L., Liu, H., Salvo, E., Wu, Z., Bradley, A., Situ, P.: Quantitative analysis of tear film fluorescence and discomfort during tear film instability and thinning. Invest. Ophthalmol. Vis. Sci.54(4), 2645–2653 (2013)

    work page 2013

  32. [32]

    Himebaugh, N.L., Nam, J., Bradley, A., Liu, H., Thibos, L.N., Begley, C.G.: Scale and spatial distribution of aberrations associated with tear breakup. Optom. Vis. Sci.89(11), 1590–1600 (2012)

    work page 2012

  33. [33]

    Nichols, J.J., Mitchell, G.L., King-Smith, P.E.: Thinning Rate of the Precorneal and Prelens Tear Films. Invest. Opthalmol. Vis. Sci.46(7), 2353 (2005)

    work page 2005

  34. [34]

    Journal of Engineering Mathematics149(1), 5 (2024) https://doi.org/10.1007/s10665-024-10407-6

    Chen, Q., Driscoll, T.A., Braun, R.J.: Evaporation-driven tear film thinning and breakup in two space dimensions. Journal of Engineering Mathematics149(1), 5 (2024) https://doi.org/10.1007/s10665-024-10407-6

    work page doi:10.1007/s10665-024-10407-6 2024

  35. [35]

    Stapf, M.R., Braun, R.J., King-Smith, P.E.: Duplex tear film evaporation analysis. Bull. Math. Biol.79(12), 2814–2846 (2017)

    work page 2017

  36. [36]

    Dursch, T.J., Li, W., Taraz, B., Lin, M.C., Radke, C.J.: Tear-film evaporation rate from simultaneous ocular-surface temperature and tear-breakup area. Optom. Vis. Sci.95, 5–12 (2018)

    work page 2018

  37. [37]

    Oron, A., Davis, S.H., Bankoff, S.G.: Long-scale evolution of thin liquid films. Rev. Mod. Phys.69, 931–980 (1997)

    work page 1997

  38. [38]

    Craster, R.V., Matar, O.K.: Dynamics and stability of thin liquid films. Rev. Mod. Phys.81, 1131–1198 (2009)

    work page 2009

  39. [39]

    Li, L., Braun, R.J., Henshaw, W.D., King-Smith, P.E.: Computed flow and 22 fluorescence over the ocular surface. Math. Med. Biol.35, 51–85 (2018)

    work page 2018

  40. [40]

    Jensen, O.E., Grotberg, J.B.: The spreading of heat or soluble surfactant along a thin liquid film. Phys. Fluids A75, 58–68 (1993)

    work page 1993

  41. [41]

    Internat

    Tiffany, J.M.: The viscosity of human tears. Internat. Ophthalmol.15(6), 371–376 (1991)

    work page 1991

  42. [42]

    Current Eye Research19(1), 4–11 (1999)

    Nagyov´ a, B., Tiffany, J.M.: Components responsible for the surface tension of human tears. Current Eye Research19(1), 4–11 (1999)

    work page 1999

  43. [43]

    Casalini, T., Salvalaglio, M., Perale, G., Masi, M., Cavallotti, C.: Diffusion and aggregation of sodium fluorescein in aqueous solutions. J. Phys. Chem.. B 115(44), 12896–12904 (2011)

    work page 2011

  44. [44]

    Riquelme, R., Lira, I., P´ erez-L´ opez, C., Rayas, J.A., Rodr´ ıguez-Vera, R.: Interfer- ometric measurement of a diffusion coefficient: Comparison of two methods and uncertainty analysis. J. Phys. D: Appl. Phys.40(9), 2769–2776 (2007)

    work page 2007

  45. [45]

    Lemp, M.A., Bron, A.J., Baudouin, C., Ben´ ıtez Del Castillo, J.M., Geffen, D., Tauber, J., Foulks, G.N., Pepose, J.S., Sullivan, B.D.: Tear osmolarity in the diagnosis and management of dry eye disease. Am.J. Ophthalmol.151(5), 792– 7981 (2011)

    work page 2011

  46. [46]

    Determination of molar absorption coefficient

    Mota, M.C., Carvalho, P., Ramalho, J., Leite, E.: Spectrophotometric analy- sis of sodium fluorescein aqueous solutions. Determination of molar absorption coefficient. Internat. Ophthalmol.15(5), 321–326 (1991)

    work page 1991

  47. [47]

    Braun, R.J., Gewecke, N.R., Begley, C.G., King-Smith, P.E., Siddique, J.I.: A Model for Tear Film Thinning With Osmolarity and Fluorescein. Invest. Ophthalmol. Vis. Sci.55(2), 1133–1142 (2014)

    work page 2014

  48. [48]

    Awisi-Gyau, D., Begley, C.G., Situ, P., Simpson, T.L.: Changes in Corneal Detec- tion Thresholds After Repeated Tear Film Instability. Invest. Ophthalmol. Vis. Sci.60(13), 4234–4240 (2019)

    work page 2019

  49. [49]

    McGraw-Hill Education, ??? (2016)

    Carlson, N.B., Kurtz, D., Hines, C.: Clinical Procedures for Ocular Examination. McGraw-Hill Education, ??? (2016)

    work page 2016

  50. [50]

    1, (2006)

    Lakowicz, J.: Principles of Fluorescence Spectroscopy vol. 1, (2006). https://doi. org/10.1007/978-0-387-46312-4

    work page doi:10.1007/978-0-387-46312-4 2006

  51. [51]

    Wu, Z., Begley, C.G., Port, N., Bradley, A., Braun, R., King-Smith, E.: The Effects of Increasing Ocular Surface Stimulation on Blinking and Tear Secretion. Invest. Ophthalmol. Vis. Sci.56(8), 4211–4220 (2015)

    work page 2015

  52. [52]

    Software, Environments, Tools

    Trefethen, L.N.: Spectral Methods in Matlab. Software, Environments, Tools. SIAM, Philadelphia (2000) 23

    work page 2000

  53. [53]

    Journal of Open Research Software5(1), 15 (2017)

    Rackauckas, C., Nie, Q.: Differentialequations.jl–a performant and feature-rich ecosystem for solving differential equations in julia. Journal of Open Research Software5(1), 15 (2017)

    work page 2017

  54. [54]

    SIAM journal on scientific computing18(1), 1–22 (1997)

    Shampine, L.F., Reichelt, M.W.: The matlab ode suite. SIAM journal on scientific computing18(1), 1–22 (1997)

    work page 1997

  55. [55]

    https://github.com/ stevengj/nlopt (2007) 24

    Johnson, S.G.: The NLopt nonlinear-optimization package. https://github.com/ stevengj/nlopt (2007) 24

    work page 2007