Non-contact acoustic micro-tapping optical coherence elastography for quantification of corneal anisotropic elasticity: in vivo rabbit study

Abstract

Purpose: To demonstrate accurate measurement of corneal elastic moduli in vivo with non-contact and non-invasive optical coherence elastography.

Methods: Elastic properties (in-plane Young's modulus E and both in-plane, u, and out-of-plane, G, shear moduli) of rabbit cornea were quantified in vivo using non-contact dynamic Acoustic micro-Tapping Optical Coherence Elastography (AuT-OCE). The IOP-dependence of measured mechanical properties was explored in extracted whole globes following in vivo measurement. A nearly-incompressible transverse isotropic (NITI) model was used to reconstruct moduli from AuT-OCE data. Independently, cornea elastic moduli were also measured ex vivo with traditional, destructive mechanical tests (tensile extensometry and shear rheometry).

Results: Our study demonstrates strong anisotropy of corneal elasticity in rabbits. The in-plane Young's modulus, computer as E=3u, was in the range of 20-44 MPa, whereas the out-of-plane shear modulus was in the range of 34-261 kPa. Both pressure-dependent ex vivo OCE and destructive mechanical tests performed on the same samples within an hour of euthanasia strongly support the results of AuT-OCE measurements.

Conclusions: Non-contact AuT-OCE can non-invasively quantify cornea anisotropic elastic properties in vivo.

Translational relevance: As OCT is broadly accepted in Ophthalmology, these results suggest the potential for rapid translation of AuT-OCE into clinical practice. In addition, AuT-OCE can likely improve diagnostic criteria of ectatic corneal diseases, leading to early diagnosis, reduced complications, customized surgical treatment, and personalized biomechanical models of the eye.

Figure 1.

a) Schematic of AμT-OCE system b) in vivo imaging of rabbit cornea c) ex vivo whole globe with needle insertion to control IOP.

Figure 2.

a) Measured corneal surface vibrations in a whole globe rabbit sample inflated to a pressure of 7 mmHg. b) Filtered and windowed x-t plot. c) Best fit of A₀ mode plotted on top of k-f plot. d) Optimization function showing best-fit and uncertainty around G and e) μ. Uncertainty intervals (red lines) calculated for the representative example where standard deviation in g_NITI was 0.3%. The best-fit solution provided G = 20.5 kPa, with uncertainty of 19.8 kPa – 21.3 kPa, and μ = 4.0 MPa, with uncertainty of 2.2 MPa – 8.7 MPa.

Figure 3.

Tono-Pen measured IOP value (mmHg) as the internal pressure was raised using a lifted water bath. Colored dots and lines correspond with each sample, black squares and error bars denote mean and standard deviation, respectively, and the black solid line corresponds with the best-fit linear function (Eq. 7). The dotted black line is the one-to-one (slope=1) line for visualization.

Figure 4.

Summary of elasticity measurements performed in vivo and ex vivo. The squares denote in vivo values, and the triangles are the mean of ex vivo measurements. Each black dot is the OCE measured ex vivo modulus for a single sample at the associated pressure (all results shown in Supplementary Materials).

Figure 5.

The mean and standard deviation for all corneas measured via OCE in vivo and ex vivo, as well as by destructive mechanical testing. a) mean shear, G, modulus and b) Young’s, E, modulus for: in vivo OCE at the fixed physiological IOP (measured with the tonopen and equal to 11 mmHg on average after the correction presented in Fig.3), ex vivo OCE for three different ranges of IOP indicated in the legend, the mean rheometry value of the storage modulus at 16 Hz, and tensile testing value of the tangential modulus at 10% strain. Standard deviation corresponds to the deviation of measured moduli across the population of 9 cornea samples.