Assessment of corneal biomechanical parameters in healthy and keratoconic eyes using dynamic bidirectional applanation device and dynamic Scheimpflug analyzer

The Increase in Corneal Stiffness After Accelerated Corneal Cross-Linking in Progressive Keratoconus Using Different Methods of Epithelial Debridement

Purpose

The purpose of this study was to investigate corneal stiffening after epithelium-off accelerated corneal cross-linking (CXL; 9 mW/cm²) in progressive keratoconus (KC) with different methods of epithelial debridement.

Methods

This was a retrospective, interventional, and non-randomized study. In group 1, the epithelium was removed using a hockey knife (N = 45). In group 2 (N = 39) and group 3 (N = 22), the epithelial thickness was measured by optical coherence tomography (OCT) and the epithelium was ablated by excimer laser, but, in group 3, stromal ablation was performed additionally to correct high order aberrations (HOAs). Corneal biomechanics (integrated invers radius [IIR], stress-strain index [SSI]) and corneal tomography (thinnest corneal thickness [TCT]) were assessed with Corvis ST and Pentacam prior to and 1 month after CXL.

Results

Corneal tomography did not differ among the groups preoperatively (P > 0.05). TCT decreased significantly in all groups after surgery (all P < 0.05). Nonetheless, corneal biomechanical stiffening was found in all three groups indicated by a decreased IIR and an increased SSI (all P < 0.05). For group 3, the HOA improved significantly (P < 0.001). Among the groups, there were no significant differences in changes of biomechanical parameters, but TCT was significantly reduced after laser ablation.

Conclusions

Corneal stiffening after CXL is independent from epithelial removal. In particular, despite the removal of stromal tissue to correct HOA, a stiffening effect was achieved in keratoconic corneas, even it was less pronounced compared to mechanical epithelial removal. The reduction in HOA indicates a possible improvement in visual acuity.

Translation Relevance

Cross-linking stiffens the keratoconus independent of epithelial debridement technique and may compensate minor stromal laser ablation.

Crosslinking with UV-A and riboflavin in progressive keratoconus: From laboratory to clinical practice - Developments over 25 years

Changes in the biomechanical and biochemical properties of the human cornea play an important role in the pathogenesis of ectatic diseases. A number of conditions in primarily acquired (keratoconus or pellucid marginal degeneration) or secondarily induced (iatrogenic keratectasia after refractive laser surgeries) ectatic disorders lead to decreased biomechanical stability. Corneal collagen cross-linking (CXL) represents a technique to slow or even halt the progression of ectatic pathologies. In this procedure, riboflavin is applied in combination with ultraviolet A radiation. This interaction induces the production of reactive oxygen species, which leads to the formation of additional covalent bonds between collagen molecules and subsequent biomechanical corneal strengthening. This procedure is so far the only method that partially interferes etiopathogenetically in the treatment of ectatic diseases that slows or stops the process of corneal destabilization, otherwise leading to the need for corneal transplantation. Besides, CXL process increases markedly resistance of collagenous matrix against digesting enzymes supporting its use in the treatment of corneal ulcers. Since the discovery of this therapeutic procedure and the first laboratory experiments, which confirmed the validity of this method, and the first clinical studies that proved the effectiveness and safety of the technique, it has been spread and adopted worldwide, even with further modifications. Making use of the Bunsen-Roscoe photochemical law it was possible to shorten the duration of this procedure in accelerated CXL and thus improve the clinical workflow and patient compliance while maintaining the efficacy and safety of the procedure. The indication spectrum of CXL can be further expanded by combining it with other vision-enhancing procedures such as individualized topographically-guided excimer ablation. Complementing both techniques will allow a patient with a biomechanically stable cornea to regularize it and improve visual acuity without the need for tissue transplantation, leading to a long-term improvement in quality of life.

Artificial intelligence in corneal diseases: A narrative review

Corneal diseases represent a growing public health burden, especially in resource-limited settings lacking access to specialized eye care. Artificial intelligence (AI) offers promising solutions for automating the diagnosis and management of corneal conditions. This narrative review examines the application of AI in corneal diseases, focusing on keratoconus, infectious keratitis, pterygium, dry eye disease, Fuchs endothelial corneal dystrophy, and corneal transplantation. AI models integrating diverse imaging modalities (e.g., corneal topography, slit-lamp, and anterior segment OCT images) and clinical data have demonstrated high diagnostic accuracy, often outperforming human experts. Emerging trends include the incorporation of biomechanical data to enhance keratoconus detection, leveraging in vivo confocal microscopy for diagnosing infectious keratitis, and employing multimodal approaches for comprehensive disease analysis. Additionally, AI has shown potential in predicting disease progression, treatment outcomes, and postoperative complications in corneal transplantation. While challenges remain such as population heterogeneity, limited external validation, and the "black box" nature of some models, ongoing advancement in explainable AI, data augmentation, and improved regulatory frameworks can serve to address these limitations.

Research progress on measurement methods and clinical applications of corneal elastic modulus

Various corneal diseases are strongly associated with corneal biomechanical characteristics, and early measurement of patients' corneal biomechanics can be utilized in their diagnosis and treatment. Measurement methods for corneal biomechanical characteristics are classified into ex vivo and in vivo. Some of these methods can directly measure certain corneal biomechanical parameters, while others require indirect calculation through alternative methods. However, due to diversities in measurement techniques and environmental conditions, significant differences may exist in the corneal mechanical properties measured by these two methods. Therefore, comprehensive research on current measurement methods and the exploration of novel measurement techniques may have great clinical significance. The corneal elastic modulus, a critical indicator in corneal biomechanics, reflects the cornea's ability to return to its initial shape after undergoing stress. This review aims to provide a comprehensive summary of the corneal elastic modulus, which is a critical biomechanical parameter, and discuss its direct, indirect, and potential measurement methods and clinical applications.

CorNet: Autonomous feature learning in raw Corvis ST data for keratoconus diagnosis via residual CNN approach

PURPOSE

To ascertain whether the integration of raw Corvis ST data with an end-to-end CNN can enhance the diagnosis of keratoconus (KC).

METHOD

The Corvis ST is a non-contact device for in vivo measurement of corneal biomechanics. The CorNet was trained and validated on a dataset consisting of 1786 Corvis ST raw data from 1112 normal eyes and 674 KC eyes. Each raw data consists of the anterior and posterior corneal surface elevation during air-puff induced dynamic deformation. The architecture of CorNet utilizes four ResNet-inspired convolutional structures that employ 1 × 1 convolution in identity mapping. Gradient-weighted Class Activation Mapping (Grad-CAM) was adopted to visualize the attention allocation to diagnostic areas. Discriminative performance was assessed using metrics including the AUC of ROC curve, sensitivity, specificity, precision, accuracy, and F1 score.

RESULTS

CorNet demonstrated outstanding performance in distinguishing KC from normal eyes, achieving an AUC of 0.971 (sensitivity: 92.49%, specificity: 91.54%) in the validation set, outperforming the best existing Corvis ST parameters, namely the Corvis Biomechanical Index (CBI) with an AUC of 0.947, and its updated version for Chinese populations (cCBI) with an AUC of 0.963. Though the ROC curve analysis showed no significant difference between CorNet and cCBI (p = 0.295), it indicated a notable difference between CorNet and CBI (p = 0.011). The Grad-CAM visualizations highlighted the significance of corneal deformation data during the loading phase rather than the unloading phase for KC diagnosis.

CONCLUSION

This study proposed an end-to-end CNN approach utilizing raw biomechanical data by Corvis ST for KC detection, showing effectiveness comparable to or surpassing existing parameters provided by Corvis ST. The CorNet, autonomously learning comprehensive temporal and spatial features, demonstrated a promising performance for advancing KC diagnosis in ophthalmology.

Keratoconus Detection-based on Dynamic Corneal Deformation Videos Using Deep Learning

Objective

To assess the performance of convolutional neural networks (CNNs) for automated detection of keratoconus (KC) in standalone Scheimpflug-based dynamic corneal deformation videos.

Design

Retrospective cohort study.

Participants

We retrospectively analyzed datasets with records of 734 nonconsecutive, refractive surgery candidates, and patients with unilateral or bilateral KC.

Methods

We first developed a video preprocessing pipeline to translate dynamic corneal deformation videos into 3-dimensional pseudoimage representations and then trained a CNN to directly identify KC from pseudoimages. We calculated the model's KC probability score cut-off and evaluated the performance by subjective and objective accuracy metrics using 2 independent datasets.

Main Outcome Measures

Area under the receiver operating characteristics curve (AUC), accuracy, specificity, sensitivity, and KC probability score.

Results

The model accuracy on the test subset was 0.89 with AUC of 0.94. Based on the external validation dataset, the AUC and accuracy of the CNN model for detecting KC were 0.93 and 0.88, respectively.

Conclusions

Our deep learning-based approach was highly sensitive and specific in separating normal from keratoconic eyes using dynamic corneal deformation videos at levels that may prove useful in clinical practice.

Financial Disclosures

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Advanced Corneal Imaging in Keratoconus: A Report by the American Academy of Ophthalmology

PURPOSE

To review the published literature on the diagnostic capabilities of the newest generation of corneal imaging devices for the identification of keratoconus.

METHODS

Corneal imaging devices studied included tomographic platforms (Scheimpflug photography, OCT) and functional biomechanical devices (imaging an air impulse on the cornea). A literature search in the PubMed database for English language studies was last conducted in February 2023. The search yielded 469 citations, which were reviewed in abstract form. Of these, 147 were relevant to the assessment objectives and underwent full-text review. Forty-five articles met the criteria for inclusion and were assigned a level of evidence rating by the panel methodologist. Twenty-six articles were rated level II, and 19 articles were rated level III. There were no level I evidence studies of corneal imaging for the diagnosis of keratoconus found in the literature. To provide a common cross-study outcome measure, diagnostic sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were extracted. (A perfect diagnostic test that identifies all cases properly has an AUC of 1.0.) RESULTS: For the detection of keratoconus, sensitivities for all devices and parameters (e.g., anterior or posterior corneal curvature, corneal thickness) ranged from 65% to 100%. The majority of studies and parameters had sensitivities greater than 90%. The AUCs ranged from 0.82 to 1.00, with the majority greater than 0.90. Combined indices that integrated multiple parameters had an AUC in the mid-0.90 range. Keratoconus suspect detection performance was lower with AUCs ranging from 0.66 to 0.99, but most devices and parameters had sensitivities less than 90%.

CONCLUSIONS

Modern corneal imaging devices provide improved characterization of the cornea and are accurate in detecting keratoconus with high AUCs ranging from 0.82 to 1.00. The detection of keratoconus suspects is less accurate with AUCs ranging from 0.66 to 0.99. Parameters based on single anatomic locations had a wide range of AUCs. Studies with combined indices using more data and parameters consistently reported high AUCs.

FINANCIAL DISCLOSURE(S)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Evaluation of Biomechanical Changes After Accelerated Cross-Linking in Progressive Keratoconus: A Prospective Follow-Up Study

PURPOSE

The aim of this study was to analyze the biomechanical effect of accelerated corneal cross-linking (9*10) in progressive keratoconus (KC) in comparison to untreated fellow eyes using Scheimpflug-based tonometry (Corvis ST, CVS).

METHODS

Forty-three eyes of 43 patients with KC showed progressive KC and were treated using accelerated corneal cross-linking. Twenty-five untreated fellow eyes were used as the control group. All eyes were examined biomechanically (CVS) and tomographically (Pentacam) at baseline, after 1-month, 6-month, and 12-month follow-up. Statistical analysis was performed using a linear mixed model. A logistic regression was performed to attribute the effects of changes in each parameter to treatment status (treated or untreated).

RESULTS

Maximum keratometry values decreased statistically significantly at 12 months by -1.1 D (95 confidence interval: -2.0 to -0.1, P = 0.025) compared with baseline. Thinnest corneal thickness decreased significantly after 1 month ( P < 0.001) and recovered to baseline after 12 months ( P = 0.752). In the corneal cross-linking (CXL) group, biomechanical changes were observed by an increased bIOP, a shorter A2 time, and a lower integrated radius after 1 month (all P < 0.05). No biomechanical and tomographical changes were observed in the control group (all P > 0.05). Logistic regression pointed out that treated eyes can be separated from untreated eyes by differences in bIOP, corneal thickness, A1 velocity, integrated radius, and Kc mean at 1, 6, and 12 months.

CONCLUSIONS

The alterations in biomechanical parameters indicated a corneal stiffening effect after CXL treatment, which was mostly detectable 1 month after treatment, although corneal thickness was reduced. The logistic regression model showed an adequate separation between CXL-treated and untreated eyes.

Biomechanical Analysis of Tomographically Regular Keratoconus Fellow Eyes Using Corvis ST

BACKGROUND

Keratoconus is a bilateral, yet asymmetric disease. In rare cases, the second eye may show no signs of tomographic changes. The purpose of this study was to analyze the biomechanical characteristics in tomographically regular keratoconus fellow eyes.

MATERIALS AND METHODS

This retrospective, consecutive case series analyzed 916 eyes of 458 patients who presented to our keratoconus clinic between November 2020 and October 2022. Primary outcome measures included best-corrected visual acuity (BCVA), tomographic Scheimpflug analysis using Pentacam AXL (Oculus, Wetzlar, Germany), and biomechanical assessment using Corvis ST (Oculus, Wetzlar, Germany). Tomographic changes were assessed via analysis of the anterior and posterior curvature, K-max, thinnest corneal thickness (TCT), the Belin/Ambrosio Deviation Display (BAD-D), and the ABCD-Grading. Biomechanical changes were analyzed using Corvis Biomechanical Index (CBI) and Tomographic Biomechanical Index (TBI).

RESULTS

Of 916 eyes, 34 tomographically regular fellow eyes (7.4%) were identified and included in the analysis. Overall, the mean BCVA was - 0.02 ± 0.13 logMAR. Tomographic analysis showed mean K-max of 43.87 ± 1.21 D, mean TCT of 532 ± 23 µm, and mean BAD-D of 1.02 ± 0.43. Biomechanical analysis demonstrated mean CBI of 0.28 ± 0.26 and mean TBI of 0.34 ± 0.30. While normal CBI-values were observed in 16 (47%) of 34 eyes, only 13 eyes (38%) showed a regular TBI and only 7 eyes (21%) showed regular TBI and CBI. The sensitivity of CBI and TBI to detect a tomographically normal keratoconus fellow eye was 53% and 62%, respectively.

CONCLUSION

A highly asymmetric corneal ectasia with regular tomographic finding in a fellow eye is rare among keratoconus patients. In such cases, a biomechanical analysis may be useful in detecting early signs of corneal ectasia. In our analysis, the TBI showed high sensitivity for detecting a biomechanical abnormality in tomographically regular fellow eyes.

Iatrogenic Keratectasia after Refractive Surgery - Causes, Prophylaxis, Therapy

Iatrogenic keratectasia is induced thinning and protrusion of the cornea after laser refractive surgery. Known risk factors include an excessively thin postoperative residual stromal bed, a thicker flap, or preoperatively undetected evidence of preexisting subclinical keratoconus. The rate of post-refractive ectasia in eyes without identifiable preoperative risk factors is 20 per 100 000 eyes for photorefractive keratectomy, 90 per 100 000 eyes for laser in situ keratomileusis, and 11 per 100 000 eyes for small incision lenticule extraction. Traditional screening tools for preoperative risk include the ectasia risk score system and percentage of tissue alteration. More recent methods include corneal elastography and epithelial mapping, in addition to Artificial Intelligence methods for data analysis. Therapy includes contact lenses, cross-linking, implantation of intracorneal ring segments, penetrating or lamellar keratoplasty, and, in early studies, implantation of corneal lenticules.

Comparison of bilateral differential characteristics of corneal biomechanics between keratoconus and normal eyes

Purpose: To compare bilateral differences in corneal biomechanics between keratoconus and normal eyes. Methods: In this case-control study, 346 eyes of 173 patients (aged 22.1 ± 6.1 years) with keratoconus (KC group) and 378 eyes of 189 patients (aged 26.7 ± 5.6 years) with ametropia (control group) were enrolled. Corneal tomography and biomechanical properties were examined using Pentacam HR and Corvis ST, respectively. The corneal biomechanical parameters were compared between eyes with forme fruste keratoconus (FFKC) and normal eyes. Bilateral differences in corneal biomechanical parameters were compared between the KC and control groups. Receiver operating characteristic (ROC) analysis was used to assess discriminative efficacies. Results: The areas under the ROC curves (AUROCs) of stiffness parameter at the first applanation (SP-A1) and Tomographic and Biomechanical Index (TBI) for identifying FFKC were 0.641 and 0.694, respectively. The bilateral differential values of major corneal biomechanical parameters were significantly increased in the KC group (all p < 0.05), except for the Corvis Biomechanical Index (CBI). The AUROCs of the bilateral differential values of the deformation amplitude ratio at 2 mm (ΔDAR2), Integrated Radius (ΔIR), SP-A1 (ΔSP-A1), and the maximum inverse concave radius (ΔMax ICR) for discriminating keratoconus were 0.889, 0.884, 0.826, and 0.805, respectively. The Logistic Regression Model-1 (comprising of ΔDAR2, ΔIR, and age) and the Logistic Regression Model-2 (comprising of ΔIR, ΔARTh, ΔBAD-D, and age) had AUROCs of 0.922 and 0.998, respectively, for discriminating keratoconus. Conclusion: The bilateral asymmetry of corneal biomechanics was significantly increased in keratoconus compared with normal eyes, which may be helpful for the early detection of keratoconus.

Repeatability of corneal deformation response parameters by dynamic ultra-high-speed Scheimpflug imaging before and after corneal crosslinking

PURPOSE

To evaluate the repeatability of deformation corneal response (DCR) parameters before and after corneal crosslinking (CXL) compared with their untreated fellow eyes (uFEs).

SETTING

University Hospital Carl Gustav Carus, Dresden, Germany; IRCCS Humanitas Research Hospital, Rozzano, Milan, Italy.

DESIGN

Multicenter, interventional reliability analysis.

METHODS

53 eyes of 53 patients with keratoconus who received CXL treatment after the disease progression (CXL group) were included. Patients were measured 3 times using a dynamic Scheimpflug analyzer to determine repeatability before and 1 month after CXL treatment. The uFEs were measured in the same way (uFE group). Reliability of DCR parameters was assessed by a coefficient of repeatability, coefficient of variation, and intraclass correlation coefficient (ICC).

RESULTS

The repeatability of DCR parameters did not change after CXL compared with the preoperative values for all investigated DCR parameters ( P > .05). In the uFE group, no statistically significant shift was observed regarding the repeatability ( P > .05). An ICC greater than 0.75 was achieved in both groups for almost all parameters. Concerning the biomechanical stiffening induced by CXL, integrated inverse radius and stress-strain index were found to be statistically significantly decreased and increased ( P < .001), respectively, both indicating stiffening. No changes were observed for the uFE group.

CONCLUSIONS

The study demonstrated highly repeatable measurements of the dynamic Scheimpflug analyzer before and after CXL. The improvement of certain DCR parameters after CXL confirmed the capability of the device to detect the stiffening effect.

Corneal Biomechanics After SMILE, Femtosecond-Assisted LASIK, and Photorefractive Keratectomy: A Matched Comparison Study

Purpose

To evaluate the change in corneal stiffness after small incision lenticule extraction (SMILE), femtosecond laser-assisted in situ keratomileusis (FS-LASIK), and photorefractive keratectomy (PRK).

Methods

Age, gender, spherical equivalent, and central corneal thickness (CCT)-matched cases undergoing SMILE with a 120-µ cap, FS-LASIK with a 110-µ flap, and PRK were enrolled. One-year change in the stress-strain index, stiffness parameter at first applanation, integrated inverse radius, deformation amplitude ratio at 2 mm, and deformation amplitude ratio at 1 mm were compared between the surgical groups by linear mixed-effect models.

Results

Within each surgical group, 120 eyes completed 1 year of follow-up. The residual stromal bed (RSB) thickness and (RSB/CCTpostop) were 348.1 ± 35.0 (0.74), 375.4 ± 31.0 (0.77) and 426.7 ± 2 µm (0.88) after SMILE, FS-LASIK, and PRK, respectively. The 1-year change in all biomechanical indices was significant, except the stress-strain index with PRK (P = 0.884). The change in all indices with SMILE were significantly greater than with FS-LASIK and with PRK (all P < 0.01), except the deformation amplitude ratio at 1 mm change between SMILE and FS-LASIK (P = 0.075). The changes in all indices with FS-LASIK were significantly greater than with PRK (all P < 0.05).

Conclusions

Although SMILE preserves the greatest amount of anterior cornea with a cap thickness of 120 µ, this also produces the smallest RSB and the greatest decrease in stiffness. Thus, the RSB is shown to be the predominant determinant of stiffness decreases, rather than the preserved anterior cornea. We recommend using a thinner cap to achieve a thicker RSB and a lesser decrease in the corneal stiffness in the SMILE procedure.

Translational Relevance

After refractive surgery, RSB is predominant determinant of stiffness decreases, rather than the preserved anterior cornea.

Corneal Biomechanics Losses Caused by Refractive Surgery

Recent advances, specifically in the understanding of the biomechanical properties of the cornea and its response to diseases and surgical interventions, have significantly improved the safety and surgical outcomes of corneal refractive surgery, whose popularity and demand continue to grow worldwide. However, iatrogenic keratectasia resulting from the deterioration in corneal biomechanics caused by surgical interventions, although rare, remains a global concern. On one hand, in vivo biomechanical evaluation, enabled by clinical imaging systems such as the ORA and the Corvis ST, has significantly improved the risk profiling of patients for iatrogenic keratectasia. That is despite the fact the biomechanical metrics provided by these systems are considered indicators of the cornea's overall stiffness rather than its intrinsic material properties. On the other hand, new surgical modalities including SMILE were introduced to offer superior biomechanical performance to LASIK, but this superiority could not be proven clinically, creating more myths than answers. The literature also includes sound evidence that tPRK provided the highest preservation of corneal biomechanics when compared to both LASIK and SMILE. The aim of this review is twofold; to discuss the importance of corneal biomechanical evaluation prior to refractive surgery, and to assess the current understanding of cornea's biomechanical deterioration caused by mainstream corneal refractive surgeries. The review has led to an observation that new imaging techniques, parameters and evaluation systems may be needed to reflect the true advantages of specific refractive techniques and when these advantages are significant enough to offer better protection against post-surgery complications.

Evaluation of corneal biomechanical properties using the ocular response analyzer and the dynamic Scheimpflug-Analyzer Corvis ST in high pressure and normal pressure open-angle glaucoma patients

PURPOSE

To characterize differences in corneal biomechanics in high (HPG) and normal pressure (NPG) primary open-angle glaucoma, and its association to disease severity.

METHODS

Corneal biomechanical properties were measured using the Ocular Response Analyzer (ORA) and the dynamic Scheimpflug-Analyzer Corvis ST (CST). Disease severity was functionally assessed by automated perimetry (Humphrey field analyzer) and structurally with the Heidelberg Retina Tomograph. To avoid a possible falsification by intraocular pressure, central corneal thickness and age, which strongly influence ORA and CST measurements, group matching was performed. Linear mixed models and generalized estimating equations were used to consider inter-eye correlation.

RESULTS

Following group matching, 60 eyes of 38 HPG and 103 eyes of 60 NPG patients were included. ORA measurement revealed a higher CRF in HPG than in NPG (P < 0.001). Additionally, the CST parameter integrated radius (P < 0.001) was significantly different between HPG and NPG. The parameter SSI (P < 0.001) representing corneal stiffness was higher in HPG than in NPG. Furthermore, regression analysis revealed associations between biomechanical parameters and indicators of disease severity. In HPG, SSI correlated to RNFL thickness. In NPG, dependencies between biomechanical readings and rim area, MD, and PSD were shown.

CONCLUSION

Significant differences in corneal biomechanical properties were detectable between HPG and NPG patients which might indicate different pathophysiological mechanisms underlying in both entities. Moreover, biomechanical parameters correlated to functional and structural indices of diseases severity. A reduced corneal deformation measured by dynamic methods was associated to advanced glaucomatous damage.

Biomechanical properties analysis of forme fruste keratoconus and subclinical keratoconus

PURPOSE

To analyze the biomechanical properties of the eye in patients with unilateral keratoconus with normal (forme fruste keratoconus [FFKC]) or abnormal topography (subclinical keratoconus [SKC]).

METHODS

This study included 153 eyes of 153 participants, including 95 eyes of patients with unilateral keratoconus, and 58 eyes of 58 healthy controls. Contralateral eyes with unilateral keratoconus were divided into two groups according to clinical manifestations and global consensus: FFKC (n = 30) and SKC (n = 65). The biomechanical characteristics were analyzed using non-parametric tests; further analysis thereof was performed after adjusting for confounding factors (i.e., intraocular pressure, age, and corneal thickness). Receiver operating characteristic curve (ROC) was used to analyze the ability of the biomechanical parameters to distinguish FFKC from SKC.

RESULTS

Statistically significant differences between the FFKC and SKC groups were found in 9 of the 18 corneal biomechanical parameters analyzed using non-parametric tests. After adjusting for confounding factors, the multivariate analysis still revealed significant statistical differences in A1-time (P = 0.017), integrated radius (IR) (P = 0.024), and tomographic and biomechanical index (TBI, P < 0.001) between the FFKC and SKC groups. Stiffness parameter at first applanation (SP-A1) (Area under ROC [AUROC] = 0.765) demonstrated the strongest distinguishing ability, except for TBI (AUROC = 0.858) and Corvis Biomechanical Index (AUROC = 0.849), however, there was no statistically significant difference in SP-A1 (P = 0.366) between FFKC and SKC.

CONCLUSIONS

Biomechanical parameters A1-time and IR have a high diversity between FFKC and SKC, besides TBI, and may reflect more subtle changes in corneal biomechanical properties (BPs) preceding SP-A1. The BPs of SKC are weaker than FFKC, which might be a basic and clue for the classification and diagnosis of the severity of early keratoconus in terms of biomechanics.

The short-term effects of wearing swimming goggles on corneal biomechanics

PURPOSE

This study aimed to assess the impact of wearing swimming goggles (SG) on corneal biomechanics.

METHODS

Corneal deformation response, central corneal thickness (CCT), intraocular pressure (IOP) and biomechanically corrected intraocular pressure (bIOP) were measured with the Corvis system (Oculus Optikgeräte GmbH, Wetzlar, Germany) in thirty-one healthy young adults while wearing a drilled SG. All measurements were obtained before, at 30 s, 2 min, 3.5 min and 5 min of wearing SG, just after SG removal and after 2 min of SG removal.

RESULTS

The corneal biomechanics is sensitive to SG wear, observing lower corneal deformability during SG use. Specifically, wearing SG caused an increase in the time and length of the first applanation and radius curvature at the highest concavity, as well as a decrease and in the velocity of the first applanation and time and deformation amplitude of the second applanation (p < 0.001 in all cases). After SG removal, corneal biomechanical parameters showed a rebound-effect, obtaining a higher corneal deformability in comparison with baseline reading (p-corrected < 0.05 in all cases). Additionally, IOP and bIOP significantly increased while wearing SG (p < 0.001 in both cases), whereas CCT remained stable (p = 0.850).

CONCLUSIONS

Wearing SG modifies the biomechanical properties of the cornea, with reduced corneal deformability during SG wear. The outcomes of this study should be taken into consideration when making clinical decisions in subjects at high risk of developing corneal ectasias or glaucoma, as well as in the post-surgical management of these ocular conditions.

Artificial Intelligence-Based Diagnostic Model for Detecting Keratoconus Using Videos of Corneal Force Deformation

Purpose

To develop a novel method based on biomechanical parameters calculated from raw corneal dynamic deformation videos to quickly and accurately diagnose keratoconus using machine learning.

Methods

The keratoconus group was included according to Rabinowitz's criteria, and the normal group included corneal refractive surgery candidates. Independent biomechanical parameters were calculated from dynamic corneal deformation videos. A novel neural network model was trained to diagnose keratoconus. Tenfold cross-validation was performed, and the sample set was divided into a training set for training, a validation set for parameter validation, and a testing set for performance evaluation. External validation was performed to evaluate the model's generalizability.

Results

A novel intelligent diagnostic model for keratoconus based on a five-layer feedforward network was constructed by calculating four biomechanical characteristics, including time of the first applanation, deformation amplitude at the highest concavity, central corneal thickness, and radius at the highest concavity. The model was able to diagnose keratoconus with 99.6% accuracy, 99.3% sensitivity, 100% specificity, and 100% precision in the sample set (n = 276), and it achieved an accuracy of 98.7%, sensitivity of 97.4%, specificity of 100%, and precision of 100% in the external validation set (n = 78).

Conclusions

In the absence of corneal topographic examination, rapid and accurate diagnosis of keratoconus is possible with the aid of machine learning. Our study provides a new potential approach and sheds light on the diagnosis of keratoconus from a purely corneal biomechanical perspective.

Translational Relevance

Our findings could help improve the diagnosis of keratoconus based on corneal biomechanical properties.

Evaluation of Corneal Biomechanical Indices in Distinguishing Between Normal, Very Asymmetric, and Bilateral Keratoconic Eyes

PURPOSE

To evaluate the ability of biomechanical indices provided by the Ocular Response Analyzer (ORA; Reichert Ophthalmic Instruments) and dynamic Scheimpflug analyzer (Corvis ST; Oculus Optikgeräte GmbH) to distinguish between normal eyes and eyes with very asymmetric ectasia (VAE) and mild and moderate keratoconus.

METHODS

This prospective, observational, and monocentric study included normal eyes (defined as keratoconus percentage index < 60, Belin/Ambrósio total deviation value [BAD-D] < 1.6, inferior-superior keratometry [I-S value] < 1.45 and maximum keratometry [Kmax] < 47) and eyes with clinical bilateral keratoconus (mild and moderate) and VAE (defined as unilateral keratoconus, where one eye showed a clinical keratoconus and the fellow eye was topographically normal [VAE-NT] or topographically and tomographically normal [VAE-NTT]). All eyes were measured by the Pentacam (Oculus Optikgeräte GmbH), ORA, and Corvis ST. Receiver operating characteristic curve analysis was performed to test the diagnostic ability.

RESULTS

Fifty-eight normal eyes and 238 ectatic eyes were included. The highest area under the curve (AUC) was provided by the Corvis Biomechanical Index (CBI) with an AUC of 0.979, followed by ORA corneal resistance factor (0.865), and corneal hysteresis (0.824) separating normal eyes from all ectatic subgroups. The AUC of the CBI was statistically significantly higher than all other parameters (DeLong test, P < .001). A sensitivity of 100% and 70.9%, respectively, and a specificity of 93.1% was found to distinguish normal eyes from VAE-NT and VAE-NTT using a cut-off value of 0.2.

CONCLUSIONS

The assessment of biomechanical properties is an additional important method to evaluate corneal ectasia independent of its stage. The CBI provides further information for ectasia screening in cases where corneal topography and tomography are clinically not suspicious by using a cutoff of 0.2. [J Refract Surg. 2022;38(6):364-372.].

Determine Corneal Biomechanical Parameters by Finite Element Simulation and Parametric Analysis Based on ORA Measurements

Purpose: The Ocular Response Analyzer (ORA) is one of the most commonly used devices to measure corneal biomechanics in vivo. Until now, the relationship between the output parameters and corneal typical biomechanical parameters was not clear. Hence, we defined the output parameters of ORA as ORA output parameters. This study aims to propose a method to determine corneal biomechanical parameters based on ORA measurements by finite element simulation and parametric analysis.

Methods: Finite element analysis was used to simulate the mechanics process of ORA measurements with different intraocular pressure (IOP), corneal geometrical parameters and corneal biomechanical parameters. A simplified geometrical optics model was built to simulate the optical process of the measurements to extract ORA output parameters. After that, 70% of the simulated data was used to establish the quantitative relationship between corneal biomechanical parameters and ORA output parameters by parametric analysis and 30% of the simulated data was used to validate the established model. Besides, ten normal subjects were included to evaluate the normal range of corneal biomechanical parameters calculated from ORA.

Results: The quantitative relationship between corneal biomechanical parameters and ORA output parameters is established by combining parametric analysis with finite element simulation. The elastic modulus (E) and relaxation limit (G_∞) of the ten normal subjects were 0.65 ± 0.07 MPa and 0.26 ± 0.15, respectively.

Conclusions: A method was proposed to determine corneal biomechanical parameters based on the results of ORA measurements. The magnitude of the corneal biomechanical parameters calculated according to our method was reasonable.

In Vivo Corneal Biomechanical Properties in a Selected Chinese Population, Measured Using the Corneal Visualization Scheimpflug Technology

Purpose: To evaluate the repeatability and reproducibility of recalculated dynamic corneal response (DCR) parameters and the biomechanical-compensated intraocular pressure (bIOP) derived from the Corneal Visualization Scheimpflug Technology (Corvis ST), as well as to study the variations of DCR parameters and their relationship with demographic, and ocular characteristics.

Methods: A total of 544 healthy subjects were recruited in this study and a series of ophthalmological examinations were performed on their right eyes. Three repeated measurements were obtained at 3-min intervals for 291 of the participants to ensure repeatability. A sum of 100 participants was examined twice within 2-h intervals using two different Corvis ST in the reproducibility study. The repeatability and reproducibility of 37 parameters, including 36 DCR parameters and bIOP, were assessed by the coefficient of repeatability (CR), coefficient of variation (CV), intraclass correlation coefficient (ICC), and within-subject standard deviation (sw). Pearson’s correlation coefficients and stepwise multivariate linear regression models were performed to investigate whether the DCR parameters were related to demographic and ocular characteristics.

Results: Of all the 37 parameters, 34 showed excellent (ICC ≥0.90) or good (ICC ≥0.75) repeatability while 27 of the 37 parameters showed excellent (ICC ≥0.90) or good (ICC ≥0.75) reproducibility. In particular, a CV of less than 20% was found for all DCR parameters and bIOP. A fraction of 14 out of 36 DCR parameters was selected for correlation analysis, based on measurement reliability and clinical relevance in referring to previous literature. Age was negatively associated with the Highest concavity delta arc length (HCdArcL) and peak distance (PD) but it positively correlated with the Whole Eye Movement Max Length (WEMML). Intraocular pressure (IOP) and central corneal thickness (CCT) were negatively associated with the deformation amplitude ratio (DARM) [1 mm], A1 Velocity (A1V), and PD, while positively related to the stiffness parameter at applanation 1 (SP-A1). The bIOP was negatively associated with A1V but positively associated with A2 Velocity (A2V). The anterior chamber volume (ACV) was negatively associated with the pachy slope (PS), WEMML, and SP-A1.

Conclusion. The Corvis ST showed good precision for the repeatability and reproducibility of 36 DCR parameters and bIOP parameters in healthy eyes. The IOP, CCT, bIOP, Km, and ACV significantly influenced the DCR parameters of the eyes.

Correlation of the Corvis Biomechanical Factor with tomographic parameters in keratoconus

PURPOSE

To investigate the relationship between corneal biomechanics and keratoconus (KC) severity as described by tomographic parameters.

SETTING

University-based German ophthalmology department.

DESIGN

Retrospective cross-sectional study.

METHODS

A total of 448 KC corneas of the Homburg Keratoconus Center and 112 healthy corneas (448+112 patients) were examined by Pentacam high-resolution and Corneal Visualization Scheimpflug Technology (Pentacam HR and Corvis ST). The KC population included a wide spectrum of disease severity based on Belin's ABCD classification. Linear regression analysis was performed between the linear term of the Corvis Biomechanical Index (CBI) (CBI beta) and the tomographic values anterior radius of curvature (ARC), posterior radius of curvature (PRC), and thinnest corneal thickness (TCT). A linear transformation of the CBI beta was performed to provide an intuitive scaling, which was referred to as the Corvis Biomechanical Factor (CBiF = -0.24294226 × CBI beta + 6.02). This scaling adjusted the CBI beta to the same scale as posterior corneal curvature (PRC).

RESULTS

There was a high correlation of the CBI beta and its modification, the CBiF, with TCT (Pearson, r = -0.775), ARC (r = -0.835), and PRC (r = -0.839) in the KC population (P < .001). In the control corneas, the correlation between the CBI beta and ARC was weak (r = -0.216, P = .022), not significant (PRC, r = -0.146, P = .125), or moderate (TCT, r = -0.628, P < .001).

CONCLUSIONS

The linear term of the CBI was highly associated with KC severity as defined by corneal tomography. The CBiF represents a new scale based on biomechanical characteristics in KC, which could serve as a basis for a biomechanical KC classification in the future.

A Potential Screening Index of Corneal Biomechanics in Healthy Subjects, Forme Fruste Keratoconus Patients and Clinical Keratoconus Patients

Purpose: This study aims to evaluate the validity of corneal elastic modulus (E) calculated from corneal visualization Scheimpflug technology (Corvis ST) in diagnosing keratoconus (KC) and forme fruste keratoconus (FFKC).

Methods: Fifty KC patients (50 eyes), 36 FFKC patients (36 eyes, the eyes were without morphological abnormality, while the contralateral eye was diagnosed as clinical keratoconus), and 50 healthy patients (50 eyes) were enrolled and underwent Corvis measurements. We calculated E according to the relation between airpuff force and corneal apical displacement. One-way analysis of variance (ANOVA) and receiver operating characteristic (ROC) curve analysis were used to identify the predictive accuracy of the E and other dynamic corneal response (DCR) parameters. Besides, we used backpropagation (BP) neural network to establish the keratoconus diagnosis model.

Results: 1) There was significant difference between KC and healthy subjects in the following DCR parameters: the first/second applanation time (A1T/A2T), velocity at first/second applanation (A1V/A2V), the highest concavity time (HCT), peak distance (PD), deformation amplitude (DA), Ambrosio relational thickness to the horizontal profile (ARTh). 2) A1T and E were smaller in FFKC and KC compared with healthy subjects. 3) ROC analysis showed that E (AUC = 0.746) was more accurate than other DCR parameters in detecting FFKC (AUC of these DCR parameters was not more than 0.719). 4) Keratoconus diagnosis model by BP neural network showed a more accurate diagnostic efficiency of 92.5%. The ROC analysis showed that the predicted value (AUC = 0.877) of BP neural network model was more sensitive in the detection FFKC than the Corvis built-in parameters CBI (AUC = 0.610, p = 0.041) and TBI (AUC = 0.659, p = 0.034).

Conclusion: Corneal elastic modulus was found to have improved predictability in detecting FFKC patients from healthy subjects and may be used as an additional parameter for the diagnosis of keratoconus.

Distribution of Corneal Geometric Landmarks and Relationship Between Their Distances and Biomechanical Parameters in the Development of Keratoconus

Purpose: To analyze the changes in coordinates and distances among three typical geometric landmarks of the cornea, namely, the thinnest point (TP), maximum curvature (Kmax), and corneal apex (AP) during the development of keratoconus, and explore the potential relationship between these changes and the abnormalities of corneal biomechanics. Methods: Normal eyes (n = 127), clinical keratoconic eyes (CKC, n = 290), and the eyes of forme fruste keratoconus (FFKC, n = 85) were included; among them, the CKC group was classified into four grades based on the Topographic Keratoconus Classification (TKC) provided by Pentacam. A total of 38 Corvis ST output parameters and three distance parameters of three typical landmarks (D_Kmax-AP, D_TP-AP, and D_Kmax-TP) based on Pentacam were included. The differences of parameters among the abovementioned six groups (Normal, FFKC, and CKC stage I to CKC stage IV) were analyzed. Spearman's rank correlation test was performed to choose several dynamic corneal response (DCR) parameters that could best reflect the changes of corneal biomechanical characteristics during the progression of the disease, and the Pearson's or Spearman's correlation test was conducted to determine the association between the three distances and the selected DCR parameters in each grade. In addition, by flipping the X coordinate of the left eye on the vertical axis to reflect the direction of the right eye, the coordinates of TP and Kmax in different developmental grades were highlighted. Results: From CKC stage II, the three geometric landmark distances commenced to correlate with the corneal DCR parameters (CBI, SPA1, IR, DA Ratio 2, ARTh, MIR, Radius, Pachy, and DA Ratio 1), which could better represent the changes of biomechanical properties from normal cornea to keratoconus. From normal cornea to CKC stage IV, the coordinates of Kmax were gradually tended to the inferior temporal region from dispersion, while TP was always concentrated in the inferior temporal region. Although D_Kmax-AP, D_Kmax-TP, and D_TP-AP all showed a gradual decreasing trend with the progress of the disease, the first two did not change significantly, and only D_TP-AP significantly approached AP in the later stage of disease development. In addition, from the FFKC group, the corresponding values of D_Kmax-TP in each disease development group were smaller than D_Kmax-AP. Conclusions: In the later stage of keratoconus, the relationship between the three typical landmark distance parameters and DCR parameters is stronger, and even the weakening of corneal biomechanical properties may be accompanied by the merger of typical landmark positions.

Effect of Mydriasis-Caused Intraocular Pressure Changes on Corneal Biomechanical Metrics

Purpose: To evaluate the dependence of biomechanical metrics on intraocular pressure (IOP).

Methods: 233 refractive surgery patients were included in this study—all were examined 3 times with the Corvis ST before and after dilation, and the differences (∆) in the main device parameters were assessed. The data collected included the biomechanically corrected IOP (bIOP), the central corneal thickness (CCT), and six dynamic corneal response (DCR) parameters, namely DA, DARatio2mm, IIR, SP-A1, CBI, and SSI. Participants were divided into three groups according to the changes in patients’ bIOP after mydriasis.

Results: Intra-operator repeatability was generally high in most of the DCR parameters obtained before and after dilation. The mean changes in bIOP and CCT after dilation were −0.12 ± 1.36 mmHg and 1.95 ± 5.23 μm, respectively. Only ∆DARatio2mm, ∆IIR, and ∆CBI exhibited a statistically significant correlation with ∆CCT (p < 0.05). The changes in all DCR parameters, especially ∆DA and ∆SP-A1 were also correlated with ∆bIOP (p < 0.01)—a 1-mmHg change in bIOP was associated, on average, with 5.612 and −0.037 units of change in SP-A1 and DA, respectively. In contrast, the weakest correlation with ∆bIOP was exhibited by ∆SSI.

Conclusion: Most corneal DCR parameters, provided by the Corvis ST, were correlated with IOP, and more weakly with CCT. Changes experienced in CCT and IOP should therefore be considered in studies on corneal biomechanics and how it is affected by disease progression and surgical or medical procedures.

Comparisons of corneal biomechanical and tomographic parameters among thin normal cornea, forme fruste keratoconus, and mild keratoconus

Background

To compare the dynamic corneal response (DCR) and tomographic parameters of thin normal cornea (TNC) with thinnest corneal thickness (TCT) (≤ 500 µm), forme fruste keratoconus (FFKC) and mild keratoconus (MKC) had their central corneal thickness (CCT) matched by Scheimpflug imaging (Pentacam) and corneal visualization Scheimpflug technology (Corvis ST).

Methods

CCT were matched in 50 eyes with FFKC, 50 eyes with MKC, and 53 TNC eyes with TCT ≤ 500 µm. The differences in DCR and tomographic parameters among the three groups were compared. The receiver operating characteristic (ROC) curve was used to analyze the diagnostic significance of these parameters. Back propagation (BP) neural network was used to establish the keratoconus diagnosis model.

Results

Fifty CCT-matched FFKC eyes, 50 MKC eyes and 50 TNC eyes were included. The age and biomechanically corrected intraocular pressure (bIOP) did not differ significantly among the three groups (all P > 0.05). The index of height asymmetry (IHA) and height decentration (IHD) differed significantly among the three groups (all P < 0.05). IHD also had sufficient strength (area under the ROC curves (AUC) > 0.80) to differentiate FFKC and MKC from TNC eyes. Partial DCR parameters showed significant differences between the MKC and TNC groups, and the deflection amplitude of the first applanation (A1DA) showed a good potential to differentiate (AUC > 0.70) FFKC and MKC from TNC eyes. Diagnosis model by BP neural network showed an accurate diagnostic efficiency of about 91%.

Conclusions

The majority of the tomographic and DCR parameters differed among the three groups. The IHD and partial DCR parameters assessed by Corvis ST distinguished FFKC and MKC from TNC when controlled for CCT.

Optical Quality in Keratoconus Is Associated With Corneal Biomechanics

PURPOSE

To evaluate the correlations between corneal biomechanical indices from dynamic Scheimpflug assessment and optical quality assessed as higher-order aberrations (HOAs) using a Hartmann-Shack ocular wavefront sensor in patients with keratoconus (KC).

METHODS

In this prospective, observational case series, the eyes with KC or KC suspect (KCS) from Osaka University Hospital, Osaka, Japan, were analyzed. Corneal biomechanical assessment was performed using Corvis ST (Oculus Optikgeräte GmbH, Wetzlar, Germany), and ocular wavefront aberrations were measured using the KR-1W (Topcon Corp, Tokyo, Japan). Correlations between the biomechanical indices and ocular HOAs were assessed. Corneal biomechanical indices included the deformation amplitude ratio within 2 mm, integrated radius, stiffness parameter at the first applanation, and the linear Corvis Biomechanical Index. Wavefront data of the central 4-mm region were expanded up to the sixth order of Zernike polynomials. The magnitudes of trefoil, coma, tetrafoil, secondary astigmatism, and spherical aberration were calculated by Zernike vector analysis and then used as ocular HOA parameters along with total HOAs.

RESULTS

Thirty-four KC eyes and 37 KCS eyes were included. KC eyes showed significant correlations between ocular HOAs and biomechanics, whereas there were few significant correlations in KCS eyes. In KC eyes, deformation amplitude ratio within 2 mm, integrated radius, and Corvis Biomechanical Index beta showed stronger correlations with coma among the wavefront parameters.

CONCLUSIONS

Corneal biomechanical indices correlated with ocular HOAs in patients with KC. In particular, there was a strong association with the increase in coma caused by inferosuperior asymmetry of the shape of the cornea in patients with KC.

Comparison of corneal biological parameters between transepithelial and epithelium-off corneal cross-linking in keratoconus

AIM

To evaluate the differences in corneal biological parameters between transepithelial and epithelium-off corneal cross-linking in keratoconus.

METHODS

In our prospective clinical trial, 40 patients (60 eyes) with progressive keratoconus were randomized to undergo corneal cross-linking with transepithelial (TE group, n=30) or epithelium-off (EO group, n=30) keratoconus. Examinations comprised topography, corneal biomechanical analysis and specular microscopy at 6mo postoperatively.

RESULTS

The keratometer values were not significantly different between the TE and EO corneal cross-linked groups in different periods (each P>0.05). The corneal thickness of the EO group was greater than that of the TE group at 1wk after the operation (each P<0.05). Regarding corneal biomechanical responses, the EO group showed a longer second applanation length than TE group (P=0.003). Regarding the corneal endothelial function, standard deviation of the endothelial cell size, and coefficient of variation in the cell area, the values of EO group were larger than those of TE group at 1wk (P=0.011, 0.026), and the percentage of hexagonal cells in EO group was lower than that in TE group at 1 and 6mo (P=0.018, 0.019).

CONCLUSION

Epithelium-off corneal cross-linking may strengthen corneal biomechanics better than TE procedure can. However, the TE procedure with a lower ultraviolet-A irradiation intensity would be safer for corneal endothelial function.

[Keratoconus detection and classification from parameters of the Corvis®ST : A study based on algorithms of machine learning]

BACKGROUND AND OBJECTIVE

In the last decades increasingly more systems of artificial intelligence have been established in medicine, which identify diseases or pathologies or discriminate them from complimentary diseases. Up to now the Corvis®ST (Corneal Visualization Scheimpflug Technology, Corvis®ST, Oculus, Wetzlar, Germany) yielded a binary index for classifying keratoconus but did not enable staging. The purpose of this study was to develop a prediction model, which mimics the topographic keratoconus classification index (TKC) of the Pentacam high resolution (HR, Oculus) with measurement parameters extracted from the Corvis®ST.

PATIENTS AND METHODS

In this study 60 measurements from normal subjects (TKC 0) and 379 eyes with keratoconus (TKC 1-4) were recruited. After measurement with the Pentacam HR (target parameter TKC) a measurement with the Corvis®ST device was performed. From this device 6 dynamic response parameters were extracted, which were included in the Corvis biomechanical index (CBI) provided by the Corvis®ST (ARTh, SP-A1, DA ratio 1 mm, DA ratio 2 mm, A1 velocity, max. deformation amplitude). In addition to the TKC as the target, the binarized TKC (1: TKC 1-4, 0: TKC 0) was modelled. The performance of the model was validated with accuracy as an indicator for correct classification made by the algorithm. Misclassifications in the modelling were penalized by the number of stages of deviation between the modelled and measured TKC values.

RESULTS

A total of 24 different models of supervised machine learning from 6 different families were tested. For modelling of the TKC stages 0-4, the algorithm based on a support vector machine (SVM) with linear kernel showed the best performance with an accuracy of 65.1% correct classifications. For modelling of binarized TKC, a decision tree with a coarse resolution showed a superior performance with an accuracy of 95.2% correct classifications followed by the SVM with linear or quadratic kernel and a nearest neighborhood classifier with cubic kernel (94.5% each).

CONCLUSION

This study aimed to show the principle of supervised machine learning applied to a set-up for the modelled classification of keratoconus staging. Preprocessed measurement data extracted from the Corvis®ST device were used to mimic the TKC provided by the Pentacam device with a series of different algorithms of machine learning.

Development of a classification system based on corneal biomechanical properties using artificial intelligence predicting keratoconus severity

Background

To investigate machine-learning (ML) algorithms to differentiate corneal biomechanical properties between different topographical stages of keratoconus (KC) by dynamic Scheimpflug tonometry (CST, Corvis ST, Oculus, Wetzlar, Germany). In the following, ML models were used to predict the severity in a training and validation dataset.

Methods

Three hundred and eighteen keratoconic and one hundred sixteen healthy eyes were included in this monocentric and cross-sectional pilot study. Dynamic corneal response (DCR) and corneal thickness related (pachymetric) parameters from CST were chosen by appropriated selection techniques to develop a ML algorithm. The stage of KC was determined by the topographical keratoconus classification system (TKC, Pentacam, Oculus). Patients who were classified as TKC 1, TKC 2 and TKC 3 were assigned to subgroup mild, moderate, and advanced KC. If patients were classified as TKC 1–2, TKC 2–3 or TKC 3–4, they were assigned to subgroups according to the normative range of further corneal indices (index of surface variance, keratoconus index and minimum radius). Patients classified as TKC 4 were not included in this study due to the limited amount of cases. Linear discriminant analysis (LDA) and random forest (RF) algorithms were used to develop the classification models. Data were divided into training (70% of cases) and validation (30% of cases) datasets.

Results

LDA model predicted healthy, mild, moderate, and advanced KC eyes with a sensitivity (S_n)/specificity (S_p) of 82%/97%, 73%/81%, 62%/83% and 68%/95% from a validation dataset, respectively. For the RF model, a S_n/S_p of 91%/94%, 80%/90%, 63%/87%, 72%/95% could be reached for predicting healthy, mild, moderate, and advanced KC eyes, respectively. The overall accuracy of LDA and RF was 71% and 78%, respectively. The accuracy for KC detection including all subgroups of KC severity was 93% in both models.

Conclusion

The RF model showed good accuracy in predicting healthy eyes and various stages of KC. The accuracy was superior with respect to the LDA model. The clinical importance of the models is that the standalone dynamic Scheimpflug tonometry is able to predict the severity of KC without having the keratometric data.

Trial registration

NCT04251143 at Clinicaltrials.gov, registered at 12 March 2018 (Retrospectively registered).

Supplementary Information

The online version contains supplementary material available at 10.1186/s40662-021-00244-4.

Corneal biomechanical changes in allergic conjunctivitis

Background

To explore corneal biomechanical changes, identify related factors and determine early indicators of keratoconus (KC) development risk in allergic conjunctivitis (AC) patients.

Methods

A total of 50 patients, including 20 eyes without AC and 30 eyes with AC were enrolled in this study. All patients underwent a complete ocular examination, including evaluations of clinical manifestations of AC, corneal tomography and densitometry by Pentacam, corneal biomechanics by Corvis ST, and corneal and epithelial thickness mapping by RTvue optical coherence tomography (OCT).

Results

The index of surface variance (ISV), index of vertical asymmetry (IVA), keratoconus index (KI), index of height decentration (IHD) and Belin/Ambrosio enhanced ectasia total deviation index (BAD-D) were significantly higher in the AC group than in the non-allergic conjunctivitis (NAC) group (P < 0.05). The tomography and biomechanical index (TBI) was also significantly higher in the AC group (P = 0.04). The average epithelial thickness in the 2–7 mm annulus was significantly thinner in the AC group than in the NAC group (P < 0.05). The average densitometry of the total cornea and the anterior layer were higher in the AC group than in the NAC group (P < 0.001). The ISV, IVA, KI, IHD and BAD-D were significantly correlated with the TBI and changes in corneal epithelial thickness in AC patients (P < 0.05). The changes in epithelial thickness were closely related to the eye rubbing frequency and allergic sign scores (P < 0.05).

Conclusions

AC patients should be advised to routinely undergo corneal tomographic and biomechanical measurements, and the TBI could be used as an indicator of KC development risk in AC patients.

Trial registration

Corneal Biomechanical Changes of Allergic Conjunctivitis, NCT04299399. Registered March 3, 2020 - Retrospectively registered.

Biomechanics of Ophthalmic Crosslinking

Crosslinking involves the formation of bonds between polymer chains, such as proteins. In biological tissues, these bonds tend to stiffen the tissue, making it more resistant to mechanical degradation and deformation. In ophthalmology, the crosslinking phenomenon is being increasingly harnessed and explored as a treatment strategy for treating corneal ectasias, keratitis, degenerative myopia, and glaucoma. This review surveys the multitude of exogenous crosslinking strategies reported in the literature, both "light" (involving light energy) and "dark" (involving non-photic chemical processes), and explores their mechanisms, cytotoxicity, and stage of translational development. The spectrum of ophthalmic applications described in the literature is then discussed, with particular attention to proposed therapeutic mechanisms in the cornea and sclera. The mechanical effects of crosslinking are then discussed in the context of their proposed site and scale of action. Biomechanical characterization of the crosslinking effect is needed to more thoroughly address knowledge gaps in this area, and a review of reported methods for biomechanical characterization is presented with an attempt to assess the sensitivity of each method to crosslinking-mediated changes using data from the experimental and clinical literature. Biomechanical measurement methods differ in spatial resolution, mechanical sensitivity, suitability for detecting crosslinking subtypes, and translational readiness and are central to the effort to understand the mechanistic link between crosslinking methods and clinical outcomes of candidate therapies. Data on differences in the biomechanical effect of different crosslinking protocols and their correspondence to clinical outcomes are reviewed, and strategies for leveraging measurement advances predicting clinical outcomes of crosslinking procedures are discussed. Advancing the understanding of ophthalmic crosslinking, its biomechanical underpinnings, and its applications supports the development of next-generation crosslinking procedures that optimize therapeutic effect while reducing complications.

Evaluation of Intraocular Pressure and Other Biomechanical Parameters to Distinguish between Subclinical Keratoconus and Healthy Corneas

(1) Purpose: To assess the main corneal response differences between normal and subclinical keratoconus (SCKC) with a Corvis® ST device. (2) Material and Methods: We selected 183 eyes of normal patients, of a mean age of 33 ± 9 years and 16 eyes of patients with SCKC of a similar mean age. We measured best corrected visual acuity (BCVA) and corneal topography with a Pentacam HD device to select the SCKC group. Biomechanical measurements were performed using the Corvis® ST device. We carried out a non-parametric analysis of the data with SPSS software (Wilcoxon signed rank-test). (3) Results: We found statistically significant differences between the control and SCKC groups in some corneal biomechanical parameters: first and second applanation time (p = 0.05 and p = 0.02), maximum deformation amplitude (p = 0.016), highest concavity radius (p = 0.007), and second applanation length and corneal velocity ((p = 0.039 and p = 0.016). (4) Conclusions: Our results show that the use of normalised biomechanical parameters provided by noncontact tonometry, combined with a discriminant function theory, is a useful tool for detecting subclinical keratoconus.

Accuracy of new Corvis ST parameters for detecting subclinical and clinical keratoconus eyes in a Chinese population

This study aimed to compare the values of new corneal visualization Scheimpflug technology (Corvis ST) parameters in normal, subclinical keratoconus (SKC) and keratoconus (KC) eyes, and evaluate the diagnostic ability to distinguish SKC and KC eyes from normal eyes. One-hundred normal, 100 SKC and 100 KC eyes were included in the study. Corvis ST parameters containing dynamic corneal response parameters were measured by one ophthalmologist. The receiver operating characteristic curve was used to evaluate the diagnostic ability of new Corvis ST parameters. The new Corvis ST parameters in KC eyes were different from those in the control and SKC eyes after adjusting for IOP and CCT, and stiffness parameter at the first applanation (SP-A1) and Corvis biomechanical index (CBI) were significantly different between the control and SKC eyes (all P < 0.05). The parameter with the highest diagnostic efficiency was SP-A1 (Youden index = 0.40, AUC = 0.753), followed by CBI (Youden index = 0.38, AUC = 0.703), and Integrated Radius (Youden index = 0.33, AUC = 0.668) in diagnosing SKC from control eyes. New Corvis ST parameters in SKC eyes were significantly different from normal control and KC eyes, and could be considered to distinguish SKC and KC eyes from normal eyes.

Corneal biomechanics: Measurement and structural correlations

The characterization of corneal biomechanical properties has important implications for the management of ocular disease and prediction of surgical responses. Corneal refractive surgery outcomes, progression or stabilization of ectatic disease, and intraocular pressure determination are just examples of the many key clinical problems that depend highly upon corneal biomechanical characteristics. However, to date there is no gold standard measurement technique. Since the advent of a 1-dimensional (1D) air-puff based technique for measuring the corneal surface response in 2005, advances in clinical imaging technology have yielded increasingly sophisticated approaches to characterizing the biomechanical properties of the cornea. Novel analyses of 1D responses are expanding the clinical utility of commercially-available air-puff-based instruments, and other imaging modalities-including optical coherence elastography (OCE), Brillouin microscopy and phase-decorrelation ocular coherence tomography (PhD-OCT)-offer new opportunities for probing local biomechanical behavior in 3-dimensional space and drawing new inferences about the relationships between corneal structure, mechanical behavior, and corneal refractive function. These advances are likely to drive greater clinical adoption of in vivo biomechanical analysis and to support more personalized medical and surgical decision-making.

Relationship between corneal biomechanical parameters and corneal sublayer thickness measured by Corvis ST and UHR-OCT in keratoconus and normal eyes

BACKGROUND

To explore the relationship between corneal biomechanical parameters and corneal sublayer thickness using corneal visualization Scheimpflug technology (Corvis ST) and ultrahigh-resolution optical coherence tomography (UHR-OCT) in clinical and suspected keratoconus and normal eyes.

METHODS

Cross-sectional prospective study. A total of 94 eyes of 70 participants were recruited. Twenty five eyes of 19 keratoconus patients, 52 eyes of 34 patients showing high risk of developing keratoconus according to the Belin/Ambrosio Enhanced Ectasia Display, and each eye of 17 normal subjects were enrolled. All participants underwent Corvis ST, Pentacam, and UHR-OCT examinations at the same time. Stiffness parameter A1 (SP-A1), deformation amplitude ratio (DA ratio), and other biomechanical parameters were recorded using Corvis ST. The vertical and horizontal thickness profiles of central 3 mm corneal epithelium, Bowman's layer, and stroma as measured by the perpendicular distance between the neighboring interfaces were generated using UHR-OCT. The flat keratometry and steep keratometry were obtained using Pentacam. Analysis of correlation was applied to explore the association between variables.

RESULTS

Most of the biomechanical parameters and corneal sublayer thickness profiles showed statistical differences among three groups. A statistically significant linear relationship was noted between SP-A1 and DA ratio in all three groups. SP-A1 was found to be positively correlated with epithelial and Bowman's layer thickness in the keratoconus (KC) group, and with stromal thickness in all three groups. In the normal and suspected keratoconus (SKC) groups, only stromal thickness was included in the stepwise linear regression to predict SP-A1, whereas in the KC group, steep keratometry and Bowman's layer thickness were included.

CONCLUSIONS

Significant and different correlations were noted between corneal stiffness and corneal sublayer thickness in different groups, indicating that corneal sublayers may play different roles in maintaining corneal biomechanical stability between keratoconus and normal eyes.

Comprehensive evaluation of corneas from normal, forme fruste keratoconus and clinical keratoconus patients using morphological and biomechanical properties

OBJECTIVE

To more comprehensively evaluate the ability of the parameters reflecting the morphological and biomechanical properties of the cornea to distinguish clinical keratoconus (CKC) and forme fruste keratoconus (FFKC) from normal.

METHODS

Normal eyes (n = 50), CKC (n = 45) and FFKC (n = 15) were analyzed using Pentacam, Corvis ST and ORA. Stepwise logistic regression of all parameters was performed to obtain the optimal combination model capable of distinguishing CKC, FFKC from normal, named SLR1 and SLR2, respectively. Receiver operating characteristic (ROC) curves were applied to determine the predictive accuracy of the parameters and the two combination models, as described by the area under the curve (AUC). AUCs were compared using the DeLong method.

RESULTS

The SLR1 model included only the TBI output by Pentacam, while the SLR2 model included the morphological parameter F.Ele.Th and two parameters from the Corvis ST, HC DfA and SP-A1. The majority of the parameters had sufficient strength to differentiate the CKC from normal corneas, even the seven separate parameters and the SLR1 model had a discrimination efficiency of 100%. The predictive accuracy of the parameters was moderate for FFKC, and the SLR2 model (0.965) presented an excellent AUC, followed by TBI, F.Ele.Th and BAD-D.

CONCLUSION

The F.Ele.Th from Pentacam was the most sensitive morphological parameter for FFKC, and the combination of F.Ele.Th, HC DfA and SP-A1 made the diagnosis of FFKC more efficient. The CRF and CH output by ORA did not improve the combined diagnosis, despite the corneal combination of morphological and biomechanical properties that optimized the diagnosis of FFKC.

Association between Corneal Stiffness Parameter at the First Applanation and Keratoconus Severity

Objective

The study aimed to evaluate the character of corneal stiffness parameter at the first applanation (SP-A1) in normal and keratoconus eyes and explore the association between SP-A1 and keratoconus severity indicators.

Methods

A total of 351 normal and 351 keratoconus eyes were included in the current study. Keratoconus was diagnosed according to the corneal topography map and slit-lamp examination. The severity of keratoconus was classified to mild (steep keratometry (Ks) < 48D), moderate (48 ≤ Ks < 55D), and severe (Ks ≥ 55D). The SP-A1 was measured using the Corvis ST software. The correlation analyses and receiver operating characteristic (ROC) curve were performed in the current analysis.

Results

The SP-A1 values of keratoconus were lower than that of normal eyes (72.11 (57.02, 83.08) mmHg/mm vs 110.89 (100.45, 122.47) mmHg/mm, P < 0.001). With the severity of keratoconus increasing, the SP-A1 decreased and the value of SP-A1 was 79.54 (70.30, 90.93) mmHg/mm, 65.11 (53.14, 77.46) mmHg/mm, and 47.59 (37.50, 62.14) mmHg/mm in mild, moderate, and severe keratoconus eyes, respectively (P < 0.001). The negative association between SP-A1 and Ks was found in mild, moderate, and severe keratoconus eyes (r _mild = -0.171, r _moderate = -0.317, r _severe = -0.288, all P < 0.05). A positive association between SP-A1 and the thinnest corneal thickness (TCT) was found in all eyes (r_normal = 0.687, r _mild = 0.519, r _moderate = 0.488, r _severe = 0.382, all P < 0.05). SP-A1 was found to be statistically positively associated with intraocular pressure (IOP), biomechanical corrected IOP (bIOP), time from the initiation of air puff until the first applanation (A1T), corneal velocity at the second applanation (A2V), and negatively associated with deformation amplitude (DA), peak distance (PD), corneal velocity at the first applanation (A1V), time from the initiation of air puff until the second applanation (A2T), and DA Ratio Max [2 mm] both in normal and keratoconus eyes (all P < 0.05). The ROC analysis indicated that the AUC (95% CI) of SP-A1 was 0.952 (0.934-0.967) and 0.930 (0.904-0.951) in detecting keratoconus eyes and mild keratoconus eyes from normal eyes, respectively.

Conclusions

The SP-A1 value decreased while the keratoconus severity increased. It was lower in keratoconus than that in normal eyes and could be helpful in identifying keratoconus eyes from normal eyes. Further researches would be warranted to expand the clinical utility of SP-A1.

Correlation Between Corneal Biomechanical Indices and the Severity of Keratoconus

PURPOSE

To investigate the correlations between the biomechanical indices determined in Scheimpflug-based corneal biomechanical assessments and the severity of keratoconus (KC) based on corneal tomographic assessments in patients with different stages of KC.

METHODS

Fifty-three patients who presented with clinical KC in 1 eye and KC suspect in the fellow eye were included. Corneal tomographic and biomechanical assessments were performed using the Pentacam HR and Corvis ST (Oculus Optikgeräte GmbH, Wetzlar, Germany). Correlations between the tomographic indices and biomechanical indices were assessed, including the anterior radius of curvature (ARC) and posterior radius of curvature (PRC) at a 3.0-mm optical zone and the thinnest pachymetry (Tmin), deformation amplitude ratio max 2 mm (DAR2mm), integrated radius, stiffness parameter at the first applanation, and linear Corvis Biomechanical Index (beta).

RESULTS

DAR2mm correlated negatively with ARC (R = -0.722), PRC (R = -0.677), and Tmin (R= -0.650) (P < 0.001 for all). Integrated radius correlated negatively with ARC (R = -0.700), PRC (R = -0.668), and Tmin (R= -0.648) (P < 0.001 for all). Stiffness parameter at the first applanation correlated positively with ARC (R = 0.622), PRC (R = 0.601), and Tmin (R = 0.703) (P < 0.001 for all). The Corvis Biomechanical Index beta correlated negatively with ARC (R = -0.754), PRC (R = -0.755), and Tmin (R= -0.765) (P < 0.001 for all).

CONCLUSIONS

Corneal biomechanical indices correlated with corneal tomographic parameters in patients with KC. These findings support the possibility of developing biomechanical-based staging classification for KC in combination with topographic or tomographic indices.

Comparison of waveform-derived corneal stiffness and stress-strain extensometry-derived corneal stiffness using different cross-linking irradiances: an experimental study with air-puff applanation of ex vivo porcine eyes

Purpose

To assess corneal stiffening of standard (S-CXL) and accelerated (A-CXL) cross-linking protocols by dynamic corneal response parameters and corneal bending stiffness (Kc[mean/linear]) derived from Corvis (CVS) Scheimpflug-based tonometry. These investigations were validated by corneal tensile stiffness (K[ts]), derived from stress-strain extensometry in ex vivo porcine eyes.

Methods

Seventy-two fresh-enucleated and de-epithelized porcine eyes were soaked in 0.1% riboflavin solution including 10% dextran for 10 min. The eyes were separated into four groups: controls (n = 18), S-CXL (intensity in mW/cm²*time in min; 3*30) (n = 18), A-CXL (9*10) (n = 18), and A-CXL (18*5) (n = 18), respectively. CXL was performed using CCL Vario. CVS measurements were performed on all eyes. Subsequently, corneal strips were extracted by a double-bladed scalpel and used for stress-strain measurements. K[ts] was calculated from a force-displacement curve. Mean corneal stiffness (Kc[mean]) and constant corneal stiffness (Kc[linear]) were calculated from raw CVS data.

Results

In CVS, biomechanical effects of cross-linking were shown to have a significantly decreased deflection amplitude as well as integrated radius, an increased IOP, and SP A1 (P < 0.05). Kc[mean]/Kc[linear] were significantly increased after CXL (P < 0.05). In the range from 2 to 6% strain, K[ts] was significantly higher in S-CXL (3*30) compared to A-CXL (9*10), A-CXL (18*5), and controls (P < 0.05). At 8% to 10% strain, all protocols induced a higher stiffness than controls (P < 0.05).

Conclusion

Several CVS parameters and Kc[mean] as well as Kc[linear] verify corneal stiffening effect after CXL on porcine eyes. S-CXL seems to have a higher tendency of stiffening than A-CXL protocols have, which was demonstrated by Scheimpflug-based tonometry and stress-strain extensometry.