Machine learning-based prediction of tear osmolarity for contact lens practice.

PURPOSE: This study addressed the utilisation of machine learning techniques to estimate tear osmolarity, a clinically significant yet challenging parameter to measure accurately. Elevated tear osmolarity has been observed in contact lens wearers and is associated with contact lens-induced dry eye, a common cause of discomfort leading to discontinuation of lens wear. METHODS: The study explored machine learning, regression and classification techniques to predict tear osmolarity using routine clinical parameters. The data set consisted of 175 participants, primarily healthy subjects eligible for soft contact lens wear. Various clinical assessments were performed, including symptom assessment with the Ocular Surface Disease Index and 5-Item Dry Eye Questionnaire (DEQ-5), tear meniscus height (TMH), tear osmolarity, non-invasive keratometric tear film break-up time (NIKBUT), ocular redness, corneal and conjunctival fluorescein staining and Meibomian glands loss. RESULTS: The results revealed that simple linear regression was insufficient for accurate osmolarity prediction. Instead, more advanced regression models achieved a moderate level of predictive power, explaining approximately 32% of the osmolarity variability. Notably, key predictors for osmolarity included NIKBUT, TMH, ocular redness, Meibomian gland coverage and the DEQ-5 questionnaire. In classification tasks, distinguishing between low (<299 mOsmol/L), medium (300-307 mOsmol/L) and high osmolarity (>308 mOsmol/L) levels yielded an accuracy of approximately 80%. Key parameters for classification were similar to those in regression models, emphasising the importance of NIKBUT, TMH, ocular redness, Meibomian glands coverage and the DEQ-5 questionnaire. CONCLUSIONS: This study highlights the potential benefits of integrating machine learning into contact lens research and practice. It suggests the clinical utility of assessing Meibomian glands and NIKBUT in contact lens fitting and follow-up visits. Machine learning models can optimise contact lens prescriptions and aid in early detection of conditions like dry eye, ultimately enhancing ocular health and the contact lens wearing experience.

Description of a new method to calculate the equator of the crystalline lens using AS-OCT images: Accuracy in non-dilated measurements.

OBJECTIVE: To establish a methodology for objectively estimating the Lens Equatorial Plane (LEP) from clinical images, comparing LEP with dilated versus non-dilated pupils. METHODS: A cohort of 91 eyes from 60 patients undergoing preoperative assessments for cataract surgery was evaluated. Anterior Segment Optical Coherence Tomography (AS-OCT) images were analysed under conditions of pharmacologically induced pupil dilation versus a non-dilated pupil. Geometrical parameters, including LEP, intersection diameter (ID), lens thickness (LT), anterior and posterior lens thickness were automatically calculated by applying standard image processing techniques to clinical AS-OCT images. RESULTS: Significant differences in lens parameters, including LEP, were observed between dilated and non-dilated conditions (all p < 0.001). A strong linear correlation was found across all geometrical variables under both conditions (r[LEP] = 0.64, r[ID] = 0.78, r[LT] = 0.99, all p < 0.001); enabling reliable correction of these differences. CONCLUSION: The study introduces an objective methodology for LEP calculation, emphasising the need to consider the eye's physiological state during preoperative measurements. Incorporating LEP into future intraocular lens (IOL) power calculation formulas and replacing the habitual effective lens position may potentially improve the accuracy of IOL power estimation and thus postoperative visual outcomes.

Comparison of Scheimpflug Corneal Tomography and Anterior Segment Optical Coherence Tomography Measurements in Corneal Cystinosis: A Case Series.

PURPOSE: To report the clinical course and compare the utility of Scheimpflug tomography (ST) and anterior segment optical coherence tomography (AS-OCT) for central corneal thickness (CCT) and corneal densitometry (CD) assessment in patients with corneal crystals owing to nephropathic cystinosis. METHODS: A retrospective chart analysis of three patients with nephropathic cystinosis and the presence of corneal cystine crystals in both eyes was performed. All patients underwent clinical examination and anterior segment photography, ST, and AS-OCT scans. Corneal densitometry was exported from built-in proprietary software for ST and from custom-made validated software for AS-OCT. Anterior segment optical coherence tomography images were rescaled to grayscale units from 0 (maximum transparency) to 100 (minimum transparency) to match built-in ST densitometry readings. Furthermore, the mean pixel intensity, representative of CD, was calculated from the pixels corresponding to the segmented cornea. RESULTS: All three patients had pathognomonic cystine crystals deposits in the cornea and were treated with cysteamine medications that resulted in clinical improvement. The CCT measured using ST exhibited a range from 560 to 958 µm. Conversely, when assessed with AS-OCT, the CCT varied within the range of 548 to 610 µm. Both examinations could be performed, but in the more severe cases, AS-OCT showed far greater utility to estimate CD. In four of six eyes examined, ST showed disproportionate CCT values, compared with the AS-OCT, whereas reliable CD measurements were only available in AS-OCT. CONCLUSION: The AS-OCT could be considered a baseline ocular measurement in cystinosis and in the evaluation of disease progression and treatment efficacy.

Corneal tissue changes following short-term soft contact lens wear of different materials.

PURPOSE: To study the effect of different soft contact lens (CL) materials during short-term wear on corneal tissue. METHODS: Twenty-two healthy participants wore both silicone hydrogel (MyDay, CooperVision) and hydrogel soft CLs (Biomedics 1 day extra, CooperVision) for 8 h per lens. In each session, Scheimpflug images were captured before and immediately after CL removal. Images were analysed using the densitometry distribution analysis, a technique from which two parameters, α (corneal transparency) and ß (corneal homogeneity), were estimated. In addition, the central corneal thickness changes after CL wear and the influence of the CL material on corneal transparency were evaluated. RESULTS: The ß parameter (homogeneity) increased by 5% after wearing both CL materials (paired t-test, p < 0.001). However, the α parameter (transparency) only increased in half of the participants. No material was found to be more determinant in causing the corneal densitometry changes. Statistically significant but not clinically relevant changes in corneal thickness were observed. CONCLUSIONS: Biomarkers of corneal tissue integrity (α and ß) were affected by short-term soft contact lens wear. The observed changes in corneal transparency and homogeneity were not clinically relevant but support the importance of participant-material biocompatibility.

The influence of soft contact lens material on the corneoscleral profile.

PURPOSE: To objectively quantify changes in corneoscleral profile, as evaluated by the limbus position and corneoscleral junction (CSJ) angle, as a consequence of wearing different soft contact lens (CL) materials. METHODS: Twenty-two healthy participants wore silicone hydrogel (SiHy, MyDay, CooperVision) and hydrogel (Hy, Biomedics 1 day extra, CooperVision) soft CLs for 8 h per lens in their left eye. In each session, corneoscleral topography was captured before and immediately after CL removal with an Eye Surface Profiler. Previously validated automatic and objective algorithms for limbal position and CSJ angle calculation were applied to 360 semi-meridians to investigate the effect of short-term CL wear on corneoscleral topography, globally and by sectors, depending on the soft CL material worn. RESULTS: Short-term soft CL wear significantly impacted limbal position (SiHy: 120 ± 97 µm, Hy: 128 ± 85 µm) and CSJ angle (SiHy: 0.57 ± 0.36°, Hy: 0.55 ± 0.40°); all p < 0.05. A statistically significant difference was found between the sectors with regard to limbus position and CSJ angle before CL wear that remained following lens wear (all pairwise comparisons, p < 0.001). Although individual differences were observed, there was no evidence that one material caused more substantial corneoscleral alterations. CONCLUSION: Corneoscleral profile parameters were altered significantly following 8 h of soft CL wear. The observed changes in limbus position and CSJ angle support the importance of participant-material biocompatibility.

Relationship between corneal tissue and shape in short-term soft contact lens wear.

PURPOSE: To investigate which morphometric and ocular surface tissue parameters are affected by short-term soft contact lens (CL) wear and to assess whether they carry related or independent information. METHODS: Twenty-two healthy participants wore silicone hydrogel (SiHy; MyDay, CooperVision) soft CLs for 8 h in their left eye. Corneal tomography and corneoscleral topography were captured before and immediately after CL wear. Central corneal thickness (CCT), corneoscleral parameters (limbus position and corneoscleral junction [CSJ] angle) and corneal tissue parameters (corneal transparency and homogeneity) were evaluated. RESULTS: Corneoscleral parameters (limbus position and CSJ angle) were independent of corneal tissue parameters (transparency and homogeneity) at baseline and after CL wear. CCT was independent of all the other parameters examined at baseline, but baseline values of corneal tissue parameters were moderately correlated with CCT change (transparency: r = -0.51; p = 0.007), homogeneity: r = -0.46; p = 0.02). CONCLUSIONS: A complete characterisation of ocular surface changes following CL wear should consider corneoscleral topography and corneal densitometry simultaneously, since they carry complementary information.

Ocular Surface Homeostasis After Scleral Lens Usage.

OBJECTIVES: The aim of this prospective study is to examine the effects of 5 hours of well-fitted, mini-scleral contact lens (mini-SL) wear on the tear film cytokine expression in healthy eyes. METHODS: Twenty-three healthy participants were included in the study. One eye of each participant was selected at random, and a mini-SL measuring 16.5 mm in diameter was fitted by an experienced contact lens specialist. The contact lens remained in place for 5 hours. Precorneal tear fluid was collected using capillary tubes at three different time points: baseline before SL insertion (T0), after 5 hours of SL wear (T1), and 3 hours after SL removal (T2). The concentration of 40 inflammatory cytokines at the three different time points was determined using multiplex bead assay. RESULTS: Mini-scleral lens wear did not result in significant changes in the cytokine-to-protein ratio after 5 hours of wear on a healthy eye. CONCLUSIONS: Although a well-fitted mini-SL reduces the rate at which the precorneal tear film is refreshed, 5 hours of lens wear did not appear to significantly affect the tears cytokine-to-protein ratio, suggesting that scleral lenses have minimal impact on corneal cytokine expression.

Influence of eye tilt on corneal densitometry.

PURPOSE: To investigate whether Pentacam densitometry readings are affected by corneal tilt. METHODS: In a prospective study, the right eyes of 86 healthy participants aged 42.8 ± 20.0 years (range 18-79 years) were imaged using Scheimpflug tomography. Elevation maps were exported to calculate corneal tilt using custom-made software, and densitometry readings were acquired directly from the corneal densitometry analysis add-on to the standard software Oculus Pentacam HR. Simple mediation analysis was applied to study age as a confounding factor in the correlation between corneal tilt and corneal densitometry. RESULTS: Corneal tilt and corneal densitometry are not independent from one another because age is significantly correlated with both corneal tilt (r = 0.50, p < 0.001) and corneal densitometry (r = 0.91, p < 0.001). Only 3.8% of the correlation between tilt and densitometry operates directly, while the remaining 96.2% depends on age. CONCLUSIONS: Corneal tilt plays a role in corneal densitometry readings, even though the interaction is strongly influenced by age. Age is a well-known factor in densitometry readings that should be taken into consideration when interpreting Scheimpflug densitometry.

Effect of Corneal Tilt on the Determination of Asphericity.

Purpose: To quantify the effect of levelling the corneal surface around the optical axis on the calculated values of corneal asphericity when conic and biconic models are used to fit the anterior corneal surface. Methods: This cross-sectional study starts with a mathematical simulation proving the concept of the effect that the eye's tilt has on the corneal asphericity calculation. Spherical, conic and biconic models are considered and compared. Further, corneal asphericity is analysed in the eyes of 177 healthy participants aged 35.4 ± 15.2. The optical axis was determined using an optimization procedure via the Levenberg-Marquardt nonlinear least-squares algorithm, before fitting the corneal surface to spherical, conic and biconic models. The influence of pupil size (aperture radii of 1.5, 3.0, 4.0 and 5.0 mm) on corneal radius and asphericity was also analysed. Results: In computer simulations, eye tilt caused an increase in the apical radii of the surface with the increase of the tilt angle in both positive and negative directions and aperture radii in all models. Fitting the cornea to spherical models did not show a significant difference between the raw-measured corneal surfaces and the levelled surfaces for right and left eyes. When the conic models were fitted to the cornea, changes in the radii of the cornea among the raw-measured corneal surfaces' data and levelled data were not significant; however, significant differences were recorded in the asphericity of the anterior surfaces at radii of aperture 1.5 mm (p < 0.01). With the biconic model, the posterior surfaces recorded significant asphericity differences at aperture radii of 1.5 mm, 3 mm, 4 mm and 5 mm (p = 0.01, p < 0.01, p < 0.01 & p < 0.01, respectively) in the nasal temporal direction of right eyes and left eyes (p < 0.01, p < 0.01, p < 0.01 & p < 0.01, respectively). In the superior-inferior direction, significant changes were only noticed at aperture radii of 1.5 mm for both right and left eyes (p = 0.05, p < 0.01). Conclusions: Estimation of human corneal asphericity from topography or tomography data using conic and biconic models of corneas are affected by eyes' natural tilt. In contrast, the apical radii of the cornea are less affected. Using corneal asphericity in certain applications such as fitting contact lenses, corneal implant design, planning for refractive surgery and mathematical modelling when a geometrical centre of the eye is needed should be implemented with caution.

Corneal tissue properties following scleral lens wear using Scheimpflug imaging.

PURPOSE: To investigate the effect of short-term scleral lens wear on the corneal stroma at a macroscopic (thickness) and microscopic (within tissue) level, including regional variations. METHODS: Fourteen young, healthy participants wore a rotationally symmetric, 16.5 mm diameter, scleral lens for 8 h. Scheimpflug images were captured before, and immediately after, lens wear, and also on a second day (without lens wear) to quantify natural corneal diurnal variations. After corneal segmentation, pixel intensities of the stromal tissue were statistically modelled using a Weibull probability density function from which parameters α and ß were derived. RESULTS: Both α and ß parameters increased significantly following scleral lens wear (by 5.7 ± 10% and 6.5 ± 6.5%, respectively, both p < 0.01). Corneal thickness also increased slightly following lens wear (mean increase 0.49 ± 1.77%, p = 0.01); however, the change in α and ß parameters did not correlate with the magnitude of corneal swelling. On the control day, small but significant corneal thinning was observed (-0.82 ± 1.1%, p = 0.03), while α and ß parameters remained stable. Both microparameters varied significantly across the cornea, with α decreasing (-15.4 ± 0.7%) and ß increasing towards the periphery (+4.4 ± 2.6%) (both p < 0.001). CONCLUSION: Corneal microparameters α and ß varied regionally across the cornea and displayed a statistically significant increase following short-term scleral lens wear, but remained stable between morning and evening measurements taken during a control day without lens wear. These corneal microparameters may be a useful metric to quantify subclinical corneal changes associated with low level hypoxia.

Scleral asymmetry as a potential predictor for scleral lens compression.

PURPOSE: To identify the position and magnitude of lens compression due to short-term miniscleral contact lens wear, as well as evaluating the usefulness of scleral asymmetry as a predictor for scleral lens decentered compression. METHODS: Fourteen healthy subjects (mean ± S.D.: 29.2 ± 6.0 years) wore a highly gas-permeable spherical haptic miniscleral contact lens during a 5-h period. Corneo-scleral height Fourier profilometry was captured using an Eye Surface Profiler (www.eaglet-eye.com) before and immediately after lens removal. Scleral asymmetry, lens compression location and magnitude were processed using custom-made algorithms, both globally and for scleral quadrants. RESULTS: Miniscleral contact lenses do not set uniformly on the ocular surface, with the largest decentration seen along the horizontal meridian. The greatest flexural stress exerted by the lens on the ocular surface occurs at the point coinciding with the inner diameter landing point of the lens and not with its overall diameter. Scleral asymmetry was significantly correlated with compression location (R = 0.71, p = 0.002) and compression magnitude (R = 0.81, p < 0.001), showing its potential as compression predictor. CONCLUSION: Larger amounts of scleral asymmetry will lead to more decentration of spherical haptic scleral lenses. Objective and accurate methods, like the one presented here, could help the practitioner prevent cases of scleral blanching or discomfort due to an excessive compression by the lens.

Scleral changes with accommodation.

PURPOSE: To assess whether the sclera changes its shape during accommodation, quantify those changes and ascertain whether they depend on age and refractive power. METHODS: Twenty-two subjects, aged between 19 and 84 years old were recruited for the study. Young subjects were grouped according to their refractive power as emmetropes (seven subjects) and myopes (seven subjects). Data were obtained with a height corneo-scleral topographer (Eye Surface Profiler) with coverage area far beyond the limbus. Lenses of different power were used to stimulate accommodative demand of 0.0 D, 2.5 D and 4.0 D. Elevation differences between measurements acquired while at the intermediate or close target and infinity where calculated for each participant for the horizontal meridian for a 16 mm chord. RESULTS: We found that sclera undergoes significant changes in its shape with accommodation in young subjects able to accommodate but not for those with limited accommodation. For a 4.0 D stimulus at the nasal side the scleral change amounted on average to 390 ± 330 µm (mean ± S.D.) for the young group. This phenomenon was more pronounced in myopes (for a 4.0 D target; nasal part: 560 ± 350 µm (mean ± S.D.)) than emmetropes (for a 4.0 D target; nasal part: 220 ± 120 µm (mean ± S.D.)). Differences were also found between different meridians. CONCLUSIONS: Scleral shape undergoes changes with accommodation and this phenomenon is more pronounced in myopes than emmetropes.

Principles of operation, accuracy and precision of an Eye Surface Profiler.

PURPOSE: To introduce a newly developed instrument for measuring the topography of the anterior eye, provide principles of its operation and to assess its accuracy and precision. METHODS: The Eye Surface Profiler is a new technology based on Fourier transform profilometry for measuring the anterior eye surface encompassing the corneo-scleral area. Details of technical principles of operation are provided for the particular case of sequential double fringe projection. Technical limits of accuracy have been assessed for several key parameters such as the carrier frequency, image quantisation level, sensor size, carrier frequency inaccuracy, and level and type of noise. Further, results from both artificial test surfaces as well as real eyes are used to assess precision and accuracy of the device (here benchmarked against one of popular Placido disk videokeratoscopes). RESULTS: Technically, the Eye Surface Profiler accuracy can reach levels below 1 µm for a range of considered key parameters. For the unit tested and using calibrated artificial surfaces, the accuracy of measurement (in terms of RMS error) was below 10 µm for a central measurement area of 8 mm diameter and below 40 µm for an extended measurement area of 16 mm. In some cases, the error reached levels of up to 200 µm at the very periphery of the measured surface (up to 20 mm). The SimK estimates of the test surfaces from the Eye Surface Profiler were in close agreement with those from a Placido disk videokeratoscope with differences no greater than ±0.1 D. For real eyes, the benchmarked accuracy was within ±0.5D for both the spherical and cylindrical SimK components. CONCLUSIONS: The Eye Surface Profiler can successfully measure the topography of the entire anterior eye including the cornea, limbus and sclera. It has a great potential to become an optometry clinical tool that could substitute the currently used videokeratoscopes and provide a high quality corneo-scleral topography.

In vivo identification of the retinal layer containing photopigments in OCT images through correlation with two-photon psychophysics.

Two-photon vision enables near-infrared light perception in humans. We investigate the possibility to utilize this phenomenon as an indicator of the location of the outer segments of photoreceptor cells in the OCT images. Since two-photon vision is independent on OCT imaging, it could provide external to OCT reference relative to which positions of retinal layers visible in OCT imaging could be measured. We show coincidence between OCT imaging of outer retinal layers and two-photon light perception. The experiment utilizes an intrinsic nonlinear process in the retina, two-photon absorption of light by visual photopigments, which triggers perception of near-infrared light. By shifting the focus of the imaging/stimulus beam, we link the peak efficiency of two-photon vision with the visibility of outer segments of photoreceptor cells, which can be seen as in vivo identification of a retinal layer containing visual photopigments in OCT images. Determination of the in-focus retinal layer is achieved by analysis of en face OCT image contrast. We discuss experimental methods and experimental factors that may influence two-photon light perception and the accuracy of the results. The limits of resolution are discussed in analysis of the one-photon and two-photon point spread functions.

Anterior scleral thickness in Marfan syndrome: A quantitative analysis.

PURPOSE: To investigate the anterior scleral thickness (AST) in patients with Marfan syndrome (MFS). METHODS: A prospective, cross-sectional study was conducted at the Department of Ophthalmology, Ghent University Hospital, Ghent, including patients with a genetically confirmed clinical diagnosis of MFS and age-, gender- and axial length-matched controls. Subjects with known corneal, conjunctival or scleral pathology and a history of ocular surgery, including pars plana vitrectomy, recent contact lens use or high-grade astigmatism were excluded. Subjects underwent non-cycloplegic autorefraction, Scheimpflug-based corneal tomography, axial length measurement and spectral-domain optical coherence tomography (OCT). AST was manually measured at 1 mm (AST1), 2 mm (AST2) and 3 mm (AST3) from the scleral spur, temporally and nasally. RESULTS: A total of 56 subjects (28 subjects in the MFS group and 28 matched subjects in the control group) were included in this study. In patients with MFS, AST was significantly reduced compared to matched controls, both overall and at every analysed measuring point in the nasal and temporal areas (p < 0.001). Central corneal thickness (CCT) and mean keratometry (Kmean) values were significantly lower in patients with MFS (p < 0.05). A positive correlation was found between nasal AST and CCT in patients with MFS. No correlation was found between AST and Kmean or between AST and axial length. In patients with MFS with ectopia lentis, compared to those without, temporal AST3 was significantly lower (p < 0.05). AST was significantly lower in patients with MFS harbouring a variant predicted to cause haploinsufficiency compared to those with a variant expected to lead to a dominant negative effect for both nasal and temporal measurements. CONCLUSION: Based on anterior segment OCT measurements, AST of patients with MFS is significantly lower compared to matched controls.

OCT based corneal densitometry: the confounding effect of epithelial speckle.

Corneal densitometry is a clinically validated method for objectively assessing the transparency of stroma. The technique is currently dominated by Scheimpflug technology. Still, optical coherence tomography (OCT), in which examination of the statistical properties of corneal speckle is undertaken, has also been considered to assess corneal densitometry. In-vivo, the stroma is observed via the epithelium. However, the effect of this external layer on stromal densitometry has not been considered as yet. This study aims to quantify the influence of epithelium integrity on corneal OCT densitometry. OCT images from eleven freshly enucleated porcine eyes before and after epithelial debridement were used. OCT densitometry was investigated at different stromal depths using four metrics of speckle statistics. Results indicate that there exist statistically significant differences in speckle statistics for a given stromal depth depending on the presence or absence of the epithelium. The estimation error in speckle statistics can reach over 20% depending on the stromal depth. The anterior stroma densitometry values are the ones most affected by epithelial integrity. In conclusion, if OCT densitometry stromal parameters are to be considered in absolute terms, it is essential to consider the confounding effect of the epithelial layer in the analysis.

Corneoscleral junction angle in healthy eyes assessed objectively.

PURPOSE: To introduce a fully objective method to measure corneoscleral junction (CSJ) angle and evaluate the CSJ angle in healthy eyes. METHODS: Corneoscleral topography (Eye Surface Profiler, ESP) was acquired from the right eye of 105 healthy Caucasian subjects, ranging from 18 to 59 years old. From the raw anterior eye height data, the topographic limbus was automatically demarcated in 360 semi-meridians. Further, in limbal location, the CSJ angle was automatically calculated from corneoscleral height data using neighbouring auxiliary points for angle calculation. Additionally, CSJ angle was statistically analysed regionally. RESULTS: The mean CSJ angle was 177.5 ± 1.1°. There is a mean difference of 7.7 ± 3.7° between the steepest (smallest) and flattest (largest) CSJ angle within the same eye. There exist statistically significant differences between temporal (178.2 ± 1.4°) and nasal (176.4 ± 1.1°) regions (paired t-test, p < 0.001), and between superior (178.1 ± 1.1°) and inferior (177.9 ± 1.1°) regions (p = 0.038). CSJ angle is correlated with limbus position (r = 0.43, p < 0.001). CONCLUSION: CSJ angle is rotationally asymmetric. CSJ varies regionally, being the smallest (steepest) in the nasal region. Significant rough changes in CSJ angle were observed for some healthy individuals.

Typical localised element-specific finite element anterior eye model.

Purpose: The study presents an averaged anterior eye geometry model combined with a localised material model that is straightforward, appropriate and amenable for implementation in finite element (FE) modelling. Methods: Both right and left eye profile data of 118 subjects (63 females and 55 males) aged 22-67 years (38.5 ± 7.6) were used to build an averaged geometry model. Parametric representation of the averaged geometry model was achieved through two polynomials dividing the eye into three smoothly connected volumes. This study utilised the collagen microstructure x-ray data of 6 ex-vivo healthy human eyes, 3 right eyes and 3 left eyes in pairs from 3 donors, 1 male and 2 females aged between 60 and 80 years, to build a localised element-specific material model for the eye. Results: Fitting the cornea and the posterior sclera sections to a 5th-order Zernike polynomial resulted in 21 coefficients. The averaged anterior eye geometry model recorded a limbus tangent angle of 37° at a radius of 6.6 mm from the corneal apex. In terms of material models, the difference between the stresses generated in the inflation simulation up to 15 mmHg in the ring-segmented material model and localised element-specific material model were significantly different (p < 0.001) with the ring-segmented material model recording average Von-Mises stress 0.0168 ± 0.0046 MPa and the localised element-specific material model recording average Von-Mises stress 0.0144 ± 0.0025 MPa. Conclusions: The study illustrates an averaged geometry model of the anterior human eye that is easy to generate through two parametric equations. This model is combined with a localised material model that can be used either parametrically through a Zernike fitted polynomial or non-parametrically as a function of the azimuth angle and the elevation angle of the eye globe. Both averaged geometry and localised material models were built in a way that makes them easy to implement in FE analysis without additional computation cost compared to the limbal discontinuity so-called idealised eye geometry model or ring-segmented material model.

Suspect glaucoma detection from corneal densitometry supported by machine learning.

PURPOSE: To discriminate suspect glaucomatous from control eyes using corneal densitometry based on the statistical modeling of the pixel intensity distribution of Scheimpflug images. METHODS: Twenty-four participants (10 suspect glaucomatous and 14 control eyes) were included in this retrospective study. Corneal biomechanics was assessed with the commercial Scheimpflug camera Corvis ST (Oculus). Sets of 140 images acquired per measurement were exported for further analysis. After corneal segmentation, pixel intensities corresponding to different corneal depths were statistically modeled for each image, from which corneal densitometry in the form of parameters α (brightness) and ß (homogeneity) was derived. After data pre-processing, parameters α and ß were input to six supervised machine learning algorithms that were trained, tested, and compared. RESULTS: There exists a statistically significant difference in α and ß parameters between suspect glaucomatous and control eyes (both, P < 0.05/N, Bonferroni). From the implemented supervised machine learning algorithms, the K-nearest neighbors (K-NN) algorithm reached 83.93% accuracy to discriminate between groups only using corneal densitometry parameters (α and ß). CONCLUSION: Densitometry of the anterior cornea including epithelium on its own has the potential to serve as a clinical tool for early glaucoma risk assessment.