Cornea modelling

Abstract

Background: Biomechanics introduces numerous technologies to support clinical practice in ophthalmology, with the goal of improving surgical outcomes and to develop new advanced technologies with minimum impact on clinical training. Unfortunately, a few misconceptions on the way that computational methods should be applied to living tissues contributes to a lack of confidence towards computer-based approaches. Methods: Corneal biomechanics relies on sound theories of mechanics, including concepts of equilibrium, geometrical measurements, and complex material behaviors. The peculiarities of biological tissues require the consideration of multi-physics, typical of the eye environment, and to adopt customized geometrical models constructed on the basis of advanced optical imaging and in-vivo testing. Results: Patient-specific models are able to predict the outcomes of refractive surgery and to exploit the results of in-vivo test to characterize the material properties of the corneal tissue. Conclusions: Corneal biomechanics can become an important support to clinical practice, provided that methods are based on the actual multi-physics and use customized geometrical and mechanical models.

Numerical investigation on epi-off crosslinking effects on porcine corneas

“…The identification procedure consisted in a sequence of numerical analyses, in each of which the parameters were modified to reduce progressively the difference on the anterior NT and SI profiles between experimental images and numerical results. Interestingly, the calibrated values of the stiffness do not agree with the values obtained in previous studies [27], showing a smaller value for the isotropic stiffness (0.01 MPa versus 0.08 MPa for μ 1 and -0.009 MPa versus -0.07 MPa for μ 2 , resulting in a shear modulus 10 times smaller) and a larger value for anisotropic stiffness (0.23 MPa to 0.45 MPa versus 0.09 for k 11 and k 12 , resulting in a value 3 to 5 times larger). These differences may be related to the use of different experimental protocols, and to the accurate preservation of the eye specimens before the execution of the test, that avoided the leaking of storage fluids in the specimens in the present campaign.…”

“…Note that the volumetric compressibility and rigidity coefficients have not been modified. For reference, the very first line of Table 3 contains the values of the analyses documented in [27]. Figure 7a, b, c, d, and e compare the numerical results with the experimental data, for the five groups of corneas, in terms of pressure versus apex displacement.…”

“…Table 1 lists the average geometrical parameters of the five groups; differences can be attributed to the biological variability of the specimens and to distinct deliveries from the abattoir. The corresponding average finite element models have been constructed using a parametric solid model generator able to describe the ellipsoidal shape of porcine corneas [27]. The subdivision of the model into ten layers across the thickness has been chosen to maintain the same discretization in the five groups, for the untreated and treated cases.…”

“…The material model considered in this study, presented in previous works [26][27][28], is an advanced hyperelastic material model developed explicitly for the human cornea, which accounts the anisotropy of the material and the quasi-incompressibility of the tissue, neglecting viscosity, porosity, and other time-dependent behaviors. Anisotropy is modeled by statistically distributed (according to a von Mises distribution) collagen fibrils immersed into an isotropic elastin and proteoglycan matrix.…”

“…The attempt to adopt an orientation closer to the experimental observations for pigs was not successful, since the resulting shapes of the inflated corneas differed markedly from the experimental ones. As already mentioned in [27], the information provided in [30] is insufficient to build an accurate model of the pig cornea fibril architecture; therefore, in the present study, the human organization was considered [19,21]. The distribution of the fibrils has been kept constant in all the models.…”