Dual Optical Elastography Techniques Reveal the Regulation of Corneal Stiffness by Collagen XIV

ABSTRACT

Collagen XIV is a fibril‐associated collagen with interrupted triple helices (FACIT) that helps regulate collagen fibril spacing, alignment, and extracellular matrix organization. Its role in corneal biomechanics remains incompletely understood. Here, we investigated murine wild‐type (WT) and collagen XIV‐deficient (Col14a1 ^−/−) corneas using complementary optical elastography modalities: Brillouin microscopy and air‐coupled ultrasound optical coherence elastography (ACUS‐OCE). Brillouin microscopy generated high‐resolution maps of Brillouin frequency shift in WT and Col14a1 ^−/− corneas (N = 4 per group), while ACUS‐OCE measured elastic wave speed in WT (N = 8) and Col14a1 ^−/− (N = 7) corneas. Both methods showed reduced mechanical metrics in Col14a1 ^−/− corneas, with lower Brillouin frequency shift and lower OCE‐derived wave speed relative to WT controls. These findings support collagen XIV as a regulator of corneal stromal biomechanics and demonstrate the value of combining Brillouin microscopy with ACUS‐OCE for detecting biomechanical alterations in murine corneas.

Dual optical elastography links collagen XIV‐dependent stromal organization to corneal stiffness, showing that Col14a1 ^−/−corneas are mechanically softer than wild‐type controls.

1. Introduction

The cornea is the main refractive element of the eye, contributing to approximately two‐thirds of its total optical power. Collagen XIV, a member of the fibril‐associated collagens with interrupted triple helices (FACIT) family, is expressed in mechanically stressed tissues such as the cornea and contributes to maintaining structural integrity and mechanical function [1]. It regulates fibrillogenesis [2, 3], stromal compaction [3], and fibril diameter and packing density [4]. In collagen XIV‐deficient mice, structural analyses have revealed disorganized fibril architecture [2, 3] characterized by increased fibril diameters, greater interfibrillar spacing, and reduced stromal compaction [1]. These perturbations in stromal organization are expected to compromise corneal mechanical integrity and subsequently, vision quality, highlighting the need for biomechanical assessment to investigate optomechanical disruptions due to collagen XIV deficiency [1, 2, 5].

A wide range of advanced elasticity imaging and mechanical testing techniques has been introduced to investigate corneal mechanical properties [6, 7, 8, 9, 10, 11]. Traditional mechanical testing approaches, primarily strip extensiometry and eye globe inflation techniques, have been employed to characterize corneal biomechanics [9, 12]. However, extensiometry, besides being invasive and destructive, does not accurately reproduce the complex in vivo mechanical conditions in the eye, and assessing curvature changes or apex displacement using eye inflation cannot distinguish corneal behavior from that of the surrounding ocular structures [9, 13, 14]. Ultrasound elastography has also been applied to estimate corneal biomechanical properties [15, 16]. However, ultrasound elastography remains unsuitable for the small and thin tissue structures, such as those found in murine eyes, which are common animal models [17]. Driven by progress in cell biomechanics and biomedical imaging, optical elastography is rapidly gaining popularity for quantifying and mapping corneal mechanical properties [10]. This growth is supported by advancements in photonics and high‐resolution optical detection, allowing mechanical assessment at the microscale and mesoscale, and bridging the gap between nanoscale and tissue‐level characterization [18]. Among these techniques, optical coherence elastography (OCE) [19, 20] and Brillouin microscopy [21, 22] are particularly well suited for evaluating murine corneal biomechanics due to their noninvasive nature and high spatial resolution [10, 18].

Brillouin microscopy [23, 24, 25, 26, 27] leverages thermally induced phonon vibrations within materials, leading to inelastic scattering and enabling the estimation of longitudinal elastic modulus [28]. This all‐optical modality has emerged as a powerful tool for 2D or 3D mechanical mapping of ocular tissues [22], including the retina [29], lens [30], and cornea [31]. Despite its promise, a major challenge in Brillouin microscopy is the difficulty in disentangling the contributions of tissue hydration and viscoelasticity, complicating efforts to establish a fundamental relationship between the Brillouin‐derived longitudinal modulus and the shear elastic modulus usually used for tissue stiffness characterization [32].

OCE utilizes optical coherence tomography (OCT) [33, 34] to detect spatially resolved tissue displacements elicited by various mechanical loading methods [19, 20]. OCE typically involves three main steps: applying a controlled deformation to the tissue using contact or non‐contact excitation methods; capturing the resulting tissue response using OCT; and estimating mechanical properties using an appropriate biomechanical model. Wave‐based OCE methods are particularly advantageous because they can be performed completely contact‐free and they do not require prior information about the applied force magnitude [10, 35]. The ability to induce elastic waves without physical contact is a key factor driving the growing adoption of OCE in ophthalmology [10, 19, 35]. Various excitation methods have been employed in OCE to generate mechanical perturbations, such as water‐coupled acoustic radiation force (ARF) excitation [36, 37], focused micro air‐pulse stimulation [38], pulsed laser excitation [39], and air‐coupled ARF excitation applied using air‐coupled ultrasound (ACUS) transducers [40, 41]. ACUS combines the benefits of air‐pulse excitation and water‐coupled ARF, enabling truly non‐contact stimulation without a coupling medium, while offering precise spatio‐temporal control for accurate viscoelastic characterization using suitable mechanical models [42, 43]. On the other hand, the Brillouin frequency shift and the derived longitudinal (or Brillouin) modulus reflect the resistance of the material to volumetric compressional deformation. By combining Brillouin microscopy and ACUS‐OCE [44], we can obtain complementary mechanical information, enabling a comprehensive assessment of the biomechanical effects of collagen XIV deficiency in the cornea.

In this study, we quantified the mechanical properties of murine wild‐type (WT) and collagen XIV‐deficient (Col14a1 ^−/−) corneas using two complementary modalities: Brillouin microscopy and ACUS‐OCE [44]. Brillouin microscopy provides high‐resolution mapping of the Brillouin frequency shift, which is related to the tissue longitudinal elastic modulus. In contrast, ACUS‐OCE enables quantitative estimation of shear elastic properties through wave‐based analysis. This integrated approach combines the high mechanical sensitivity and resolution of Brillouin microscopy with the fast, quantitative, and wide‐field imaging capabilities of OCE, thereby enabling complementary assessment of corneal biomechanics.

2. Methods and Materials

2.1. Sample Preparation

Freshly enucleated whole mouse eyes of wild‐type (WT) and Col14a1 ^−/− on C57BL/6 backgrounds were prepared [2]. Col14a1 ^−/− mice are deficient for collagen XIV, and no collagen XIV protein was detected in these corneas [1]. Eyes were collected from 60‐day‐old mice for this study. Each genotype group (WT and Col14a1 ^−/−) consisted of six male and two female eyes. All the procedures involving animals were conducted in accordance with the protocols approved by the Institutional Animal Care and Use Committee at the University of South Florida College of Medicine. The experiments adhered to established guidelines for the ethical use of laboratory animals. After sample preparation, whole eyes were shipped overnight in corneal preservation media [45] (Optisol; Chiron Ophthalmics, Irvine, CA, USA). Samples were independently assigned to Brillouin microscopy and ACUS‐OCE measurements. For Brillouin imaging, a total of N = 8 corneal samples were analyzed, including WT (N = 4) and Col14a1 ^−/− (N = 4). For ACUS‐OCE measurements, WT (N = 8) and Col14a1 ^−/− (N = 7) corneas were imaged. Both OCE and Brillouin measurements were conducted within 24 h of enucleation. During imaging, the whole eye globe was mounted on a 3D‐printed holder. To avoid sample dehydration, the eye globe was covered with regular topical addition of 1X phosphate‐buffered saline solution [40]. Prior to the measurements, intraocular pressure (IOP) was measured using an iCare TONOLAB tonometer (iCare Finland Oy, Vantaa, Finland). The WT group had an IOP of 9.62 ± 0.74 mmHg, whereas the Col14a1 ^−/− group had an IOP of 10.25 ± 0.95 mmHg, in which the IOP changes are not statistically significant (t‐value = 1.10, df = 13, p > 0.05).

2.2. Brillouin Microscopy System

Brillouin microscopy was carried out in both Col14a1 ^−/− and WT corneas utilizing a home‐built Brillouin microscopy system in which a 660‐nm single‐mode laser (Torus; Laser Quantum Inc., Fremont, CA, USA) illuminated the sample, with the incident power on the sample of ~17 mW. The backscattered light collected from the samples was transferred via single‐mode optical fiber to a dual‐stage virtually imaged phase array (VIPA) spectrometer [46] with ~30 GHz of free spectral range. An electron multiplying charge‐coupled device camera (iXon Andor, Belfast, United Kingdom) was employed to detect the Brillouin frequency shift. The exposure time was constant at 0.1 s per spectrum for all the measurements. Before each measurement, calibration was performed with water, acetone, and methanol to precisely and accurately characterize the free spectral range and the spectral pixel resolution. A microscope objective with 0.42 numerical aperture focused the light onto the sample, leading to the axial resolution of 10 μm and lateral resolution of 1.8 μm, which were assessed using a beam profiler (LaserCam‐HR II; Coherent Inc., Santa Clara, CA, USA). To obtain two‐dimensional Brillouin maps of the corneas, we utilized a motorized stage to scan 700 μm laterally (step size of 14 μm) and 160 μm axially (step size of 16 μm) centered at the apex. LabVIEW software (NI, Austin, TX) was employed to synchronize the operation of the translational stages, CCD camera, and data acquisition. Figure 1a shows a schematic of the Brillouin microscopy system, and Figure 1b demonstrates the mouse cornea structure overlaid with an example of Brillouin photon‐phonon interaction.

FIGURE 1.

(a) Brillouin microscopy schematic. AP: aperture, L: lens, SMF: single‐mode fiber, λ/2: half‐wave plate, λ/4: quarter‐wave plate, OL: objective lens, PBS: polarization beam splitter. (b) The mouse cornea consists of multiple layers, including the epithelium, stroma, and endothelium. The inset shows the photon‐phonon interaction that gives rise to the Brillouin frequency shift.

2.3. Brillouin Microscopy Data Analysis

Following data acquisition, the raw spectral signals representing the Brillouin spectrum were analyzed using a double Lorentzian [47] curve fitting in MATLAB R2024a (MathWorks, Natick, MA, USA). The Brillouin frequency shift at each measured point was [21]:

$${\nu }_B=\frac{2n}{\lambda }\sin \frac{\theta }{2}\sqrt{\frac{M}{{\rho }^{\prime }}}$$ (1)

where n is the refractive index, $\lambda$ is the optical wavelength, $\theta$ is the angle between the incident and scattered light, and $M$ and $\rho$ are the longitudinal elastic modulus and density, respectively. The variation in $\raisebox{1ex}{$\rho $}\!\left/ \!\raisebox{-1ex}{$n^2$}\right.$ is minimal in corneal tissues, suggesting that changes in refractive index and density partially compensate, yielding an approximately constant ratio. Since the wavelength and scattering angle are fixed, the Brillouin frequency shift is primarily governed by the longitudinal modulus. Under the epi‐detection configuration ($\sin \theta /2$ = 1), the Brillouin frequency shift reaches its maximum value. Then, a spatially resolved Brillouin image of the cornea was generated by applying a color‐coded map to the pointwise Brillouin frequency shift values. For quantitative analysis, the mean and standard deviation of the Brillouin frequency shift were computed for each sample in its corresponding group (Col14a1 ^−/− and WT).

2.4. ACUS‐OCE Experimental Setup

Figure 2a shows a schematic of the ACUS‐OCE experimental setup with murine whole eye‐globe samples, and inset Figure 2b demonstrates the confocal alignment of the OCT beam with the ACUS transducer and wave propagation through the cornea. The OCE system was based on a phase‐sensitive spectral domain OCT system, utilizing a broadband superluminescent diode (S840‐B‐I‐20; Superlum Diodes Ltd., Ireland) centered at 840 nm with a spectral bandwidth of approximately 49 nm. The axial and lateral resolutions in air were measured as ~9 and ~8 μm, respectively, with a displacement sensitivity of 0.28 nm at an OCT SNR of > 30 dB. The OCT system operated at a fixed A‐line rate of 200 kHz, yielding a temporal resolution of 5 μs.

FIGURE 2.

(a) Schematic of the ACUS‐OCE system. AP: Aperture, C: Collimator, FC: Fiber coupler, G: Grating, L: Lens, OL: Objective lens, PC: Polarized controller, SLD: superluminescent diode. (b) Demonstration of wave propagation in the cornea using ACUS excitation.

For the excitation, a noncontact approach was employed via a hemispherical ACUS transducer with a 1 MHz resonant frequency [40]. This transducer featured a circular aperture of ~10 mm diameter at the apex for OCT imaging, an external spherical diameter of 20 mm, and a focal length of ~20 mm. To co‐focus the OCT and ACUS beams, the transducer was positioned between the scanning lens and the sample. A function generator (DG4162, RIGOLTech, Beijing, China) was used to generate a continuous 1 MHz sinusoidal signal, modulated by five cycles of a 20‐kHz square pulse with a 50% duty cycle. The signal was subsequently amplified by a radio‐frequency amplifier (A150, Electronics & Innovation, Rochester, NY) prior to being delivered to the ACUS transducer.

2.5. ACUS‐OCE Measurement and Processing

To acquire both structural and motion information in two dimensions (x‐z plane) within the murine cornea samples, the M‐B mode imaging paradigm was employed [35, 48]. As illustrated in Figure 2b, the x‐axis aligned horizontally along the corneal surface, while the z‐axis corresponded to the depth of the cornea sample. Each M‐mode scan included 1000 A‐lines over a 5 ms duration and covered 500 lateral positions across approximately 5 mm. The data analysis was performed in MATLAB R2024a (MathWorks, Natick, MA, USA). First, the cornea surface was identified, and the axial particle velocity $v_z\left(t\right)$ was computed using the following formula:

$$v_z\left(t\right)=\frac{{\lambda }_0}{4\mathit{\pi n\tau }}∆{\varphi }_z\left(t\right)$$ (2)

in which the ${\lambda }_0=840\mathrm{nm}$, $n=1.376$, $\tau =5\mathrm{\mu s},$ and $\mathrm{\Delta }{\varphi }_z\left(t\right)$ refer to the light source central wavelength, the corneal refractive index, the time interval, and the phase difference at each depth between successive A‐lines, respectively. To filter the wave propagation occurring at the excitation frequency of 20 kHz, a finite impulse response filter was utilized in association with a zero‐phase filtering technique. Then, the motion artifacts resulting from refractive index mismatching and surface motion were corrected [49]. To calculate the elastic wave speed, first, the space–time map was generated, and then linear regression was utilized to measure the slope of the elastic wave propagation of the space–time map. Due to the confocal excitation geometry, the generated elastic wave exhibited cylindrical propagation centered at the stimulation point. Wave speed was quantified along two opposing meridional directions, and the measurements were averaged to enhance accuracy and mitigate potential spatial variability.

2.6. Statistical Analysis

To assess statistical significance, identify trends, and interpret Brillouin microscopy measurements and the wave propagation speed in ACUS‐OCE measurements between the Col14a1 ^−/− and WT groups, a two‐sample t‐test was conducted. The results are illustrated using box‐and‐whisker plots, where the boxes indicate the interquartile range, the line through the middle of the box shows the median, the hollow square marks the mean, and the whiskers represent the 25th and 75th percentiles.

3. Results and Discussion

Figure 3a,b shows the Brillouin spectra (black profile) along with their corresponding Lorentzian curve fits (red profile) for representative WT and Col14a1 ^−/− corneas, respectively. The distance between two peaks is greater in the Col14a1 ^−/− sample, indicating a lower Brillouin frequency shift and thus softer mechanical properties compared to the WT cornea. Figure 3c,d shows exemplary 2D Brillouin maps for WT and Col14a1 ^−/− corneas. A distinct difference in Brillouin frequency shift was observed between WT and Col14a1 ^−/− corneas, reflecting a decrease in corneal stiffness associated with collagen XIV deficiency.

FIGURE 3.

Processing of Brillouin data. (a) Analysis of the Brillouin spectrum (black trace) with double Lorentzian curve fit (red trace) for a representative WT and (b) Col14a1 ^−/− cornea. The blue dash lines indicate the locations of two peaks, and the separation between these two peaks is greater in Col14a1 ^−/− samples compared to the WT. (c) 2D Brillouin frequency shift maps of a representative WT and (d) Col14a1 ^−/− cornea. Scale bars in the x and z directions are 100 and 50 μm, respectively.

The results of ACUS‐OCE are shown in Figure 4. Figure 4a shows a representative B‐mode OCT image of a cornea. Figure 4b,c demonstrate the wave propagation snapshots at t = 2.40 ms and t = 2.44 ms after excitation. Figure 4d,e show an example of the space–time maps of the wave propagation for a representative WT and Col14a1 ^−/− samples, and the white dotted rectangle is the area where the wave speed was calculated. Figure 4f represents the frequency filtering at the excitation frequency of f = 20 kHz. Figure 4g shows central corneal thickness (CCT) measurements for WT and Col14a1 ^−/− cornea samples, which were 102.08 ± 15.29 and 97.42 ± 25.42 μm, respectively, and were not significantly different (t‐value = 0.44, df = 13, p > 0.05).

FIGURE 4.

(a) B‐mode OCT intensity image of a representative murine cornea. (b) Wave propagation snapshots at t = 2.40 ms and (c) t = 2.44 ms after excitation. (d) Example of space–time map of the wave propagation at the cornea surface for a representative WT and (e) Col14a1 ^−/− . (f) Frequency filtering at the excitation frequency of f = 20 kHz. (g) Comparison of central corneal thickness (CCT) for WT and Col14a1 ^−/− cornea samples. The boxes represent the interquartile range, the central line is the median, the central inscribed hollow square is the mean, and the whiskers are 25th and 75th percentiles.

To compare the biomechanical characteristics of WT and Col14a1 ^−/− corneas using the dual‐modality optical elastography technique, Figure 5a,b illustrate the quantification of Brillouin frequency shift and OCE‐derived wave speed for each sample, respectively. Results from both Brillouin microscopy and OCE indicate that collagen XIV deficiency decreases the cornea's overall stiffness. The Brillouin frequency shift was significantly reduced (t‐value = 4.05, df = 6, p < 0.01) in the Col14a1 ^−/− corneas compared to WT. The mean Brillouin frequency shift was 6.67 ± 0.03 GHz in the WT group and 6.44 ± 0.10 GHz in the Col14a1 ^−/− group, which was approximately 3% higher than that of the Col14a1 ^−/− group. The OCE‐measured average wave speed was 5.29 ± 0.31 m/s in wild‐type corneas and 3.55 ± 0.11 m/s in Col14a1 ^−/− corneas, which was statistically significant (t‐value = 13.61, df = 13, p < 0.001).

FIGURE 5.

(a) Quantification of mean Brillouin frequency shift (WT: N = 4, and Col14a1 ^−/− : N = 4) and (b) OCE‐measured wave speed (WT: N = 8, and Col14a1 ^−/− : N = 7). The boxes represent the interquartile range, the central line is the median, the central inscribed hollow square is the mean, and the whiskers are 25th and 75th percentiles (**p < 0.01 and ***p < 0.001).

In this study, we investigated how collagen XIV deficiency in murine corneas regulates the biomechanical properties using a dual‐modality approach. The results of both Brillouin microscopy and ACUS‐OCE show that collagen XIV deficiency leads to a reduction in stiffness. Brillouin microscopy results implied that collagen XIV deficiency resulted in a reduction in the longitudinal modulus compared to the wild‐type corneas. ACUS‐OCE revealed a similar trend, with collagen XIV deficiency in Col14a1 ^−/− corneas resulting in reduced elastic wave velocity and decreased mechanical stiffness. The results of our study provide insight into how Col14a1 ^−/− deficiency in murine corneas affects the corneal mechanical outcome. The complementary mechanical characterization by OCE is directly linked to the shear elastic modulus of the cornea, and the Brillouin frequency shift is tightly related to the longitudinal modulus (i.e., volumetric compressibility). However, the fundamental link between the Brillouin‐measured longitudinal modulus and shear (or Young's) modulus remains an open question. Nonetheless, empirical correlations between these parameters may be established for a given material [21], especially with another modality such as OCE [30].

Brillouin microscopy has been widely applied in ophthalmology to assess corneal biomechanics [24, 50, 51, 52, 53]. Scarcelli et al. used Brillouin microscopy to evaluate biomechanical changes following corneal collagen crosslinking (CXL), a clinical treatment to halt keratoconus and potentially other corneal ectasias [54]. Their work quantitatively assessed post‐CXL mechanical alterations and compared the efficiency of various treatment protocols, such as epithelium‐on versus epithelium‐off procedures [54]. In another study, Webb et al. have used Brillouin microscopy to detect corneal biomechanical anisotropy across different orientations and remain sensitive even at high hydration levels while distinguishing hydration‐independent stiffening effects by cross‐linking [55]. Mercatelli et al. targeted human corneal morphomechanics using a correlative multimodal approach employing second harmonic generation, Brillouin microscopy, and Raman spectroscopy [56]. OCE has also been employed to study the role of different collagen types in regulating corneal stiffness [44, 57]. Kling et al. reported that collagen V insufficiency in an Ehlers‐Danlos syndrome animal model resulted in increased corneal stiffness compared to WT controls, and these findings were corroborated using two‐dimensional extensometry [57].

Nair et al. employed a multimodal optical elastography approach combining Brillouin microscopy, wave‐based ACUS‐OCE, and strain‐based OCE to assess the mechanical properties of collagen XII–deficient murine corneas relative to WT controls [44]. In contrast to the trend observed in collagen V deficiency, all three optical elastography modalities demonstrated a significant decrease in stiffness in collagen XII‐deficient corneas [44]. However, these findings contradicted earlier atomic force microscopy (AFM) measurements of collagen XII‐deficient corneas, which reported approximately a twofold increase in Young's modulus compared to WT tissue [58]. Importantly, our findings in collagen XIV‐deficient corneas are consistent with previous AFM measurements [1]. In that study, Sun et al. reported a reduction in AFM‐measured Young's modulus from 59.2 kPa in WT murine corneas to 30.5 kPa in collagen XIV‐deficient samples [1]. The modest increase in fibril diameter and reduction in fibril packing density observed in collagen XIV‐deficient corneas may contribute to the decreased stromal stiffness measured in our optical elastography experiments, consistent with the microstructural alterations previously reported [1].

It should also be considered that the notable differences in corneal geometry between WT and Col14a1 ^−/− corneas may influence stiffness measurements obtained via ACUS‐OCE. Variations in curvature, thickness, or overall shape could affect the mechanical response captured by this technique, potentially contributing to the observed discrepancies [1, 59] because the calculated wavelengths for both WT and Col14a1 ^−/− corneas were longer than the corneal thickness. Under these conditions, the wave propagation cannot be considered a pure surface (Rayleigh) wave, and the propagation exhibits dispersive guided‐wave (Lamb‐like) behavior that is governed by the sample thickness in addition to intrinsic biomechanical properties. Therefore, the conventional surface wave equation cannot be strictly applied to estimate the elasticity, which is a limitation of the present analysis. Nevertheless, there was no significant difference in CCT between the WT and Col14a1 ^−/− groups as shown in Figure 4g, but the measured wave speed was significantly higher in the WT corneas compared to the Col14a1 ^−/− corneas. These values are reasonably consistent with previous ACUS‐OCE‐based estimations [44]. This suggests that the observed difference in wave speed is not influenced by corneal thickness but instead reflects a change in the elasticity of the corneas, and the present data set provides a reliable comparison and serves as a valuable supporting foundation for future studies, including advanced structural investigations using two‐photon imaging to further elucidate the microstructural mechanisms underlying the observed biomechanical alterations. In addition, sex and age‐related biological variations may act as potential confounders when comparing corneal mechanical parameters [60, 61]. Although age was matched between the two groups in this study, sex‐related factors could still introduce bias and should be considered when interpreting the results. The cornea is a hormone‐responsive tissue, and sex hormones, especially estrogen, can influence corneal collagen organization, hydration, and thickness, thereby influencing biomechanical behavior. In addition, hormonal fluctuations (e.g., during the menstrual cycle) have been reported to alter CCT [62, 63, 64]. To address these potential sources of variability, future studies will implement more strict controls for both sex and age matching, which will help reduce biological variation and improve the consistency of corneal stiffness measurements. Moreover, we plan to perform longitudinal imaging to assess how the mechanical properties of collagen deficient corneas evolve during development and ageing [65]. This methodology also paves the way for non‐invasive evaluation of corneal mechanics for additional collagen types and other critical components, such as glycosaminoglycans in the cornea.

4. Conclusions

In the current study, we have demonstrated a dual‐modality optical elastography approach to reveal the regulatory role of collagen XIV in modulating corneal biomechanics. Specifically, to assess corneal mechanical stiffness in wild‐type and Col14a1 ^−/− corneas, we pursued two complementary objectives: to quantify changes in the Brillouin frequency shift as a measure of the longitudinal elastic modulus, and to quantify elastic wave speed using OCE to evaluate the changes in shear elastic modulus. Our results show that collagen XIV deficiency results in decreased overall corneal stiffness by both approaches. These findings emphasize the capability of combined optical elastography techniques to detect subtle biomechanical alterations, underscoring their potential utility for identifying corneal abnormalities and advancing the clinical assessment of corneal disorders.

Funding

This work was supported by National Institutes of Health (R01EY034114, R01EY030063, R01EY028666, R01EY032537, R01EY033978, P30EY007551, R01EY022362).

Conflicts of Interest

M.S. and K.V.L. have a financial interest in ElastEye LLC, which is not directly related to this work. K.V.L. also has a financial interest in Shockwave Medical Inc. These financial interests have been disclosed to the University of Houston, and appropriate steps have been taken to address any potential conflicts of interest. G.S. has a financial interest in Intelon Optics, which is not directly related to this work. These financial interests have been disclosed to the University of Maryland, and appropriate steps have been taken to address any potential conflicts of interest. The other authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.