The detection of keratoconus using a three-dimensional corneal model derived from anterior segment optical coherence tomography

Introduction

Keratoconus (KCN) is a progressive corneal degeneration that affects both eyes, leading to corneal ectasia characterized by steepening and thinning of the cornea (1). These changes can lead to significant astigmatism and myopia, leading to mild to severe impairment in vision (2). Prevalence varies globally, ranging from 0.054% to 0.15% in the USA, 0.0003% in Russia, and 2.3% in Central India (3,4). Both environmental exposures, such as sunlight and hot temperatures, and behavioral components, such as eye rubbing due to atopy and ocular allergies, have been associated with a higher incidence of KCN (3).

Within the refractive surgery population, the incidence of KCN patients may be higher than in the general population, with one study showing an incidence rate of 5.7% in 106 eyes and another study that showing an incidence rate of 18.7% in 2,931 patients (5,6). Due to the increased risk of complications, KCN is the most common contraindication for performing refractive surgery, accounting for 24.0% of declined cases (6). KCN patients who undergo laser in situ keratomileusis (LASIK), photorefractive keratectomy (PRK), or small incision lenticule extraction (SMILE) have an increased risk of developing iatrogenic corneal ectasia, leading to decreased uncorrected visual acuity (7,8). Residual stromal bed thickness, age, and preoperative corneal thickness are significant risk factors (7). Using these metrics, a risk factor stratification model was created to predict this complication with a specificity of 91% and a sensitivity of 96% (9). Adding more metrics to this model will allow for increased specificity and sensitivity, allowing for the earlier detection and treatment of KCN. Additionally, improvement of this model will allow for better refractive surgery screening, decreasing the prevalence of iatrogenic corneal ectasia.

Traditional methods for detecting KCN are the use of ultrasound central pachymetry and Placido disk-based topography, but each technique has significant limitations (10-13). Pachymetry only measures a single point of thickness on the cornea (14), while Placido disk-based topography is only able to measure the anterior and central cornea accurately (15). These limitations are particularly important as the posterior corneal surface may be more sensitive to early ectatic changes (16). In contrast, Scheimpflug 3-D tomography, Orbscan, and various modalities of anterior segment optical coherence tomography (AS-OCT) are capable of assessing the anterior corneal surface along with posterior corneal curvature and corneal thickness (17), but each has weaknesses that limit their diagnostic accuracy. Additionally, in vivo corneal biomechanical assessments using modalities such as Ocular Response Analyzer and CorVis ST have been used for the detection of KCN, but with varying degrees of success (13,18).

Most current AS-OCT technologies available in the USA are unable to directly measure the volume of the cornea, while none can directly measure the surface area (10). However, anterior and posterior corneal surface areas can be derived using AS-OCT, and were used to detect KCN (19,20).

The metrics of posterior and anterior corneal surface areas have not been well studied in the detection of KCN. One study suggests that the corneal surface area remains conserved in various corneal shapes, including KCN (21). In addition, our previous study has shown that corneal arc length, cross-sectional area, and segmental arc length and area changes between adjacent segments differ between keratoconic and healthy control corneas (22). In this small population feasibility study, we plan to derive the anterior corneal surface area, posterior corneal surface area, and corneal volume using AS-OCT images rendered into three-dimensional (3D) models of corneas with KCN and healthy corneas. This pilot study aims to determine if these metrics can be used to distinguish healthy corneas from those with KCN. We present this article in accordance with the STROBE reporting checklist (available at https://aes.amegroups.com/article/view/10.21037/aes-25-24/rc).

Methods

Patients with KCN, along with healthy participants between the ages of 20 and 79 years old, were examined at the slit lamp and then imaged using a Heidelberg Spectralis AS-OCT (Heidelberg Engineering, Heidelberg, Germany). A total of 41 patients were selected from an outpatient clinical practice between 2016 and 2020. The sample size of this study was based on the number of patients who volunteered for this research study. Patients with KCN were initially diagnosed with the use of Placido disk-based topography (inferior corneal steepening, central corneal power >47 diopters, and asymmetric bowtie pattern with skewed radial axes) in conjunction with clinical exam and slit lamp biomicroscopy. Pregnant, nursing, and patients with other anterior segment pathology were excluded. Image sets with poor quality and/or insufficient images were also excluded from the study. The AS-OCT image scan through which the fixation light beam appeared was identified, along with the four inferior and four superior scans.

Images were then opened in ImageJ (NIH, Maryland, USA) for analysis. Using a horizontal rectilinear line 570 pixels in length (equivalent to 6 mm), a baseline was created perpendicular to the incident light beam, establishing bounds for further measurements (Figure 1). For further geographic analysis, the 6 mm corneal cross-sectional image was then divided into six 1 mm segments (labeled 1–6, from temporal to nasal).

Figure 1 ImageJ image with demarcated central 6 mm and divided into six 1 mm segments.

Free-D (Institut Jean-Pierre Bourgin, Versailles, France) was then used for further analysis (23). A pixel ratio of 10.53 was used to convert the pixel width and length between the two programs. This value was calculated using the 6mm horizontal rectilinear line established on ImageJ. A slice spacing of 282 micrometers was used, as this was the standard for our 15° × 10° AS-OCT scan pattern size. Polygonal models were created in the program to trace the cross-sectional area within the central 6 mm and the cross-sectional area within each 1 mm segment. Polyline models were created to trace the anterior corneal curve, posterior corneal curve, anterior surface area segments (1-6), and posterior surface area segments (1-6), with each segment labeled, left to right, regardless of the eye. The points within the boundaries of each model were manually selected. 3D images (Figure 2) as well as surface areas and volumes were then generated using these models. All measurements were manually obtained from ImageJ and Free-D. ImageJ measurements took approximately 20 minutes per eye, and the Free-D measurements took approximately one hour per eye. All of the measurements were performed by a single author (S.N.T.) who helped develop the technique and had relevant knowledge of the disease and computer programs.

Figure 2 Free-D 3D model of a cornea after being demarcated for anterior corneal surface area, posterior corneal surface area, and corneal volume. 3D, three-dimensional.

Statistical analysis was performed with GraphPad Prism (GraphPad Software, Inc., San Diego, California). When comparing all healthy corneas to all corneas with KCN, a mixed effects model with a random effect for patients was utilized. This model allowed for different variances for cases and controls in order to account for the inclusion of two eyes from one patient. The eyes with KCN were then subcategorized into either more advanced or mild disease states. Since only OCT images were used in the final dataset for this pilot study, pachymetry was used to assign subgroups. For patients with KCN who had both eyes measured, the thinner of the two was selected as part of the more advanced group. In contrast, the thicker of the two corneas with KCN comprised the less advanced KCN group. Patients with only one eye measured would automatically be placed in the more advanced group. Control patients with only one eye measured were automatically selected for the healthy cohort. Control patients with both eyes measured, right or left, were selected by a random number generator for the healthy cohort. A two-tailed Mann-Whitney test was used to assess statistical significance when comparing these subsets, and a P value of <0.05 was considered significant.

Ethical considerations

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by institutional ethics board of The University of Maryland School of Medicine (protocol HM-HP-00072265-11) and informed consent was taken from all individual participants.

Results

Overall, 33 eyes from 21 patients with KCN were included in this study, along with 33 eyes from 20 healthy controls, ranging from 20 to 79 years old, with an average age of 44 years old. The KCN group comprised 17 right eyes and 16 left eyes. The control group was comprised of 17 right eyes and 16 left eyes. The average age of the KCN and control groups was 38 years and 49 years, respectively.

Comparison of the central 6 mm images between KCN and healthy control eyes revealed statistically significant differences in the anterior corneal surface area (13.991 vs. 13.927, P =0.046), posterior corneal surface area (14.173 vs. 14.045, P<0.001), and corneal volume (7.773 vs. 8.430, P<0.001) within the central 6 mm (Table 1). Further comparison of the more advanced eyes with the KCN subgroup, the difference for all three metrics widened and remained statistically significant (Table 2).

Analysis of each corneal segment between all the eyes with KCN was then compared to all healthy control eyes. This revealed statistically significant differences at all corneal volume segments 1 (P<0.001), 2 (P<0.001), 3 (P<0.001), 4 (P<0.001), 5 (P=0.005), and 6 (P=0.01), anterior corneal surface area at segments 1 (P=0.005), 2 (P=0.01), and 4 (P=0.002), and posterior corneal surface area at segments 1 (P=0.002), 2 (P<0.001), 4 (P=0.003), and 5 (P=0.002) (Table S1, Figure 3). Examining the subgroup of more advanced corneas with KCN compared to healthy control eyes showed similar statistically significant differences at those same corneal segments, but with a wider margin (Table S2, Figure 4).

Figure 3 Segmented anterior corneal surface area, posterior corneal surface area, and corneal volume were compared between all healthy controls and all corneas with KCN. (A) Comparison of posterior corneal surface area segments between healthy eyes versus eyes with KCN. (B) Comparison of anterior corneal surface area segments between healthy eyes versus eyes with KCN. (C) Comparison of corneal volume segments between healthy eyes versus eyes with KCN. *, P<0.05; **, P<0.01; ***, P<0.001. KCN, keratoconus.

Figure 4 Segmented anterior corneal surface area, posterior corneal surface area, and corneal volume were compared between healthy controls and both more and less advanced corneas with KCN. (A) Comparison of posterior corneal surface area segments between healthy eyes, less advanced eyes with KCN, and more advanced eyes with KCN. (B) Comparison of anterior corneal surface area segments between healthy eyes, less advanced eyes with KCN, and more advanced eyes with KCN. (C) Comparison of corneal volume segments between healthy eyes, less advanced eyes with KCN, and more advanced eyes with KCN. *, P<0.05; **, P<0.01; ***, P<0.001. KCN, keratoconus.

When examining the changes of these three metrics between adjacent corneal segments, we continued to observe statistically significant differences between corneas with KCN and healthy control eyes. When comparing all corneas with KCN and all healthy control eyes, there were statistically significant differences between 1 segment pair using anterior corneal surface area, 3 segment pairs using corneal volume, and 3 segment pairs using the posterior corneal surface area (Table S3, Figure S1). Examining the subgroup of more advanced KCN eyes compared to healthy control eyes showed statistically significant differences at similar corneal segments, but with a wider margin. In addition, when comparing changes in posterior surface area between segments, there are statistically significant differences between each segment pair (Table S4, Figure S2).

Lastly, the less advanced eyes (n=15) with KCN were compared to the healthy control corneas (n=21). Given the small sample size in this subset analysis, we did not anticipate finding any significant difference between the two groups. There were no significant differences between the two groups using the metrics of anterior corneal surface area and posterior corneal surface area. However, the metric of volume remained significant (P=0.02) (Table 2). Segmental analysis revealed significant differences between the two groups using the metric of cornea volume, at segments 2 (P=0.04), 3 (P=0.03), and 4 (P=0.02), and using the metric of anterior corneal surface area, at segment 4 (P=0.02) (Table S2, Figure 4). Lastly, analyzing the change in corneal volume and surface areas between adjacent segments, there remained statistically significant differences between 2 pairs of segments using the metric of posterior corneal surface area (Table S4, Figure S2).

Discussion

In this feasibility study, the metrics of corneal surface area and volume were measured using 3D models derived from AS-OCT in patients with KCN and healthy patients. While the Belin ABCD has been used to classify KCN (24), we did not have access to this grading system using OCT images alone. For this study, we subdivided our patients into a less advanced group versus a more advanced group based on corneal thickness. Significant differences between corneas with KCN and healthy corneas were noted when comparing anterior corneal surface area, posterior corneal surface area, and corneal volume within the central 6 mm. The results confirmed that the posterior corneal surface area was significantly larger in corneas with KCN compared to healthy corneas (19). The metric of volume within the central 6 mm can be used to differentiate between healthy corneas and less advanced KCN corneas. Additionally, by segmenting the cornea into six 1 mm segments, we observed differences between healthy corneas and corneas with KCN in half or more of these segments. The majority of these segments were localized in the paracentral or temporal aspect of the cornea. When we compared the changes between segments using these three metrics of anterior corneal surface area, posterior surface area, and volume between all corneas with KCN and all healthy corneas, there were statistically significant differences at multiple segmental pairs. When compared to the more advanced KCN group, these previous pairs remained significant. In addition, all pairs of posterior corneal surface area segments were significant, with two pairs remaining statistically even when compared to the less advanced KCN group. This supports the notion that posterior corneal changes may be a more sensitive indicator of ectasia (16).

Corneal ectasia is described as the thinning and steepening of the cornea (1). These findings can be explained by biomechanical factors that affect the cornea over time. Loss of corneal tissue, stretching of corneal tissues, or a combination of both may play a role in the progression and/or development of ectasia. The loss of corneal tissue has long been assumed and can be measured through the metric of volume. Whereas the stretching of corneal tissue is understudied as a biomechanical factor. If one imagines a balloon that gets stretched, the total volume of rubber in the balloon does not decrease, even though areas of the balloon are thinned by stretching. Similarly, KCN may indicate a thinning of the cornea through stretching, without significant overall volume loss. The only way to then detect this biomechanically would be through measuring the surface area, which would be increased. However, these metrics are not available using the current imaging available in the USA.

The metrics of anterior corneal surface area, posterior surface area, and volume should be considered in conjunction with the current Placido disk-based topography to detect KCN. Detection of KCN will allow for more accurate preoperative refractive surgical evaluations, and potentially decrease post-operative ectatic disease (25). Previous retrospective studies have been able to use abnormal topography, residual stromal bed thickness, age, and preoperative corneal thickness to develop a risk factor stratification scale to predict this complication (9). The addition of corneal surface area along with corneal volume to these previous scales and scores may increase the specificity and sensitivity of the diagnostic test and treatment guidelines.

Screening for KCN has evolved and new metrics and imaging modalities have been studied more recently. Keratometry measures the central corneal astigmatism and curvature, ultrasound central pachymetry measures corneal thickness, while Placido-based topography maps corneal curvature alterations. Scheimpflug 3-D tomography and very high-frequency ultrasound provide detailed corneal imaging, and optical coherence tomography offers high-resolution cross-sectional views. Corneal epithelium mapping using optical coherence tomography, is well studied for the KCN and shows promise; however, it may lack diagnostic accuracy in subclinical KCN and forme fruste KCN (26). Additionally, ocular wavefront analysis evaluates optical aberrations, while corneal biomechanical assessment gauges tissue integrity and resilience, collectively ensuring a thorough screening approach for KCN detection The introduction of these new metric and additional imaging modalities has aided in the diagnosis of KCN, however, none has reached the pinnacle of early detection needed to detect KCN with accuracy prior to the appearance of other clinical signs (13,18).

With the advent of newer technology, examining the true anatomic detail of the cornea is closer to reality. However, most current technology in the USA is still unable to directly measure 3D features of the cornea, such as surface area and volume (10). Unfortunately, other instruments, such as Scheimpflug cameras, that do measure corneal volume have difficulty ascertaining limbus to limbus measurements at high enough acquisition speed to be reliable and minimize motion. These limitations lead to poor measurements of key corneal parameters, such as corneal thickness, and poor image quality (27). Whereas AS-OCT can produce higher definition images, with studies showing good repeatability in corneas with KCN, when compared to other imaging modalities such as Scheimpflug camera combined with Placido disk-based topography (27,28). Previously, we have shown that AS-OCT can be used with a high inter-rater reliability to measure differences in anterior and posterior corneal arc length in eyes with and without KCN (29). In this small population feasibility study, we showed that by using current AS-OCT imaging and a third-party modeling software, Free-D, we were able to reliably derive the surface area and volume of the cornea and establish possible clinical utility.

After subcategorizing our dataset into more and less advanced disease states, we observed that these metrics showed a statistically significant difference between healthy control eyes and more advanced corneas with KCN at many of the same corneal segments, despite halving the sample size. When comparing healthy control eyes to the less advanced corneas with KCN, the metric of total volume within the central 6 mm, three volume segments, one anterior surface area segment, and changes between two pairs of adjacent posterior surface area segments showed significant differences. The majority of these segments localize to the paracentral or temporal cornea. These observations suggest that early focal thinning in the cornea with KCN can be detected by measuring corneal volume, anterior surface area, and changes in posterior corneal surface area. Significantly, the total volume of the cornea with the central 6 mm can also be used as an indicator to differentiate between subclinical KCN versus normal eyes.

Interestingly, our data showed that there were greater differences in the paracentral and temporal segments. This was seen in the segmental volume and segmental anterior corneal surface area. The posterior segment surface area showed a significant difference in multiple segments, but was most significant in the temporal segments. This finding supports the idea that KCN preferentially affects the temporal and paracentral cornea to a greater extent. Previous studies using topographic evaluation of corneas with KCN have shown that around 65% of apices are found in the inferotemporal quadrant (30). Central apices, within the central 2 mm, were found to have a thinner central corneal thickness and minimum corneal thickness as compared to non-central apices (31). Our findings agree with these previous works. Thus, the inclusion of the metrics of volume, anterior, and posterior surface may aid in the detection of corneas with KCN when used in conjunction with current Placido disk-based topography.

The study was limited by several factors. First, the lack of significant quantitative data from anterior segment OCT imaging, particularly in the USA, inspired this study but required the use of additional software. This, in turn, meant our analysis was manual and time-intensive, making it more prone to human error. While this method may not be practical in a clinical practice, we hope further developments in the quantitative component of anterior segment OCT commercial software packages will alleviate this issue. The sample size was small, which limited statistical power in subset analysis. This pilot study demonstrated the feasibility of our approach and still showed statistically significant findings, while establishing the need for further study to validate. Additionally, we intentionally cropped our images to the central 6 mm of the cornea due to the effect of shadowing and poor image quality of the peripheral cornea. While we lost some valuable peripheral data, we avoided more significant noise and data loss issues. Some parameters only show small numeric differences between the two groups. This is a testament to the excellent resolution that can be achieved with AS-OCT imaging, allowing for the detection of minimal corneal changes that may be missed with other imaging modalities. Finally, there is a notable difference in the average age of the KCN and control group, 38 and 49 years old, respectively. Studies have shown a 5–6 micron decrease in central corneal thickness with each decade of life (32). However, we would not expect this age-related thinning to significantly impact the value of corneal surface area or volume. Despite these limitations, this study establishes with statistical significance that corneal surface area and volume can reliably be measured from 3D corneal models derived from AS-OCT in patients with KCN and healthy patients and could be considered as quantitative software advances continue to emerge.

Conclusions

Our small population feasibility study establishes that corneal surface area and volume may be important metrics in detecting KCN. Importantly, the volume and the changes in the posterior surface area showed the most significant differences and remained pronounced even in cases of less advanced corneas with KCN. Despite limitations, these novel corneal metrics derived from AS-OCT 3D models are worth further study and automation.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://aes.amegroups.com/article/view/10.21037/aes-25-24/rc

Data Sharing Statement: Available at https://aes.amegroups.com/article/view/10.21037/aes-25-24/dss

Peer Review File: Available at https://aes.amegroups.com/article/view/10.21037/aes-25-24/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://aes.amegroups.com/article/view/10.21037/aes-25-24/coif). W.M.M. received payments for Legal consulting and participate on Advisory Board of Sanofi in May 2023. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by institutional ethics board of The University of Maryland School of Medicine (protocol HM-HP-00072265-11) and informed consent was taken from all individual participants.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Ferdi AC, Nguyen V, Gore DM, et al. Keratoconus Natural Progression: A Systematic Review and Meta-analysis of 11 529 Eyes. Ophthalmology 2019;126:935-45. [Crossref] [PubMed]
2. Krachmer JH, Feder RS, Belin MW. Keratoconus and related noninflammatory corneal thinning disorders. Surv Ophthalmol 1984;28:293-322. [Crossref] [PubMed]
3. Gokhale NS. Epidemiology of keratoconus. Indian J Ophthalmol 2013;61:382-3. [Crossref] [PubMed]
4. Munir SZ, Munir WM, Albrecht J. Estimated Prevalence of Keratoconus in the USA From a Large Vision Insurance Database. Eye Contact Lens 2021;47:505-10. [Crossref] [PubMed]
5. Wilson SE, Klyce SD. Screening for corneal topographic abnormalities before refractive surgery. Ophthalmology 1994;101:147-52. [Crossref] [PubMed]
6. Al-Amri AM. Prevalence of Keratoconus in a Refractive Surgery Population. J Ophthalmol 2018;2018:5983530. [Crossref] [PubMed]
7. Wolle MA, Randleman JB, Woodward MA. Complications of Refractive Surgery: Ectasia After Refractive Surgery. Int Ophthalmol Clin 2016;56:127-39. [Crossref] [PubMed]
8. Moshirfar M, Tukan AN, Bundogji N, et al. Ectasia After Corneal Refractive Surgery: A Systematic Review. Ophthalmol Ther 2021;10:753-76. [Crossref] [PubMed]
9. Randleman JB, Woodward M, Lynn MJ, et al. Risk assessment for ectasia after corneal refractive surgery. Ophthalmology 2008;115:37-50. [Crossref] [PubMed]
10. Romero-Jiménez M, Santodomingo-Rubido J, Wolffsohn JS. Keratoconus: a review. Cont Lens Anterior Eye 2010;33:157-66; quiz 205. [Crossref] [PubMed]
11. de Sanctis U, Loiacono C, Richiardi L, et al. Sensitivity and specificity of posterior corneal elevation measured by Pentacam in discriminating keratoconus/subclinical keratoconus. Ophthalmology 2008;115:1534-9. [Crossref] [PubMed]
12. Oliveira CM, Ribeiro C, Franco S. Corneal imaging with slit-scanning and Scheimpflug imaging techniques. Clin Exp Optom 2011;94:33-42. [Crossref] [PubMed]
13. Zhang X, Munir SZ, Sami Karim SA, et al. A review of imaging modalities for detecting early keratoconus. Eye (Lond) 2021;35:173-87. [Crossref] [PubMed]
14. Belin MW, Khachikian SS. New devices and clinical implications for measuring corneal thickness. Clin Exp Ophthalmol 2006;34:729-31. [Crossref] [PubMed]
15. Belin MW, Ratliff CD. Evaluating data acquisition and smoothing functions of currently available videokeratoscopes. J Cataract Refract Surg 1996;22:421-6. [Crossref] [PubMed]
16. Ambrósio R Jr, Caiado AL, Guerra FP, et al. Novel pachymetric parameters based on corneal tomography for diagnosing keratoconus. J Refract Surg 2011;27:753-8. [Crossref] [PubMed]
17. Yazici AT, Bozkurt E, Alagoz C, et al. Central corneal thickness, anterior chamber depth, and pupil diameter measurements using Visante OCT, Orbscan, and Pentacam. J Refract Surg 2010;26:127-33. [Crossref] [PubMed]
18. Kwok S, Pan X, Liu W, et al. High-frequency ultrasound detects biomechanical weakening in keratoconus with lower stiffness at higher grade. PLoS One 2022;17:e0271749. [Crossref] [PubMed]
19. Kitazawa K, Yokota I, Sotozono C, et al. Measurement of Corneal Endothelial Surface Area Using Anterior Segment Optical Coherence Tomography in Normal Subjects. Cornea 2016;35:1229-33. [Crossref] [PubMed]
20. Kitazawa K, Itoi M, Yokota I, et al. Involvement of anterior and posterior corneal surface area imbalance in the pathological change of keratoconus. Sci Rep 2018;8:14993. [Crossref] [PubMed]
21. Smolek MK, Klyce SD. Is keratoconus a true ectasia? An evaluation of corneal surface area. Arch Ophthalmol 2000;118:1179-86. [Crossref] [PubMed]
22. Mohammed ISK, Tran S, Toledo-Espiett LA, et al. The detection of keratoconus using novel metrics derived by anterior segment optical coherence tomography. Int Ophthalmol 2022;42:2117-26. [Crossref] [PubMed]
23. Andrey P, Maurin Y. Free-D: an integrated environment for three-dimensional reconstruction from serial sections. J Neurosci Methods 2005;145:233-44. [Crossref] [PubMed]
24. Belin MW, Kundu G, Shetty N, et al. ABCD: A new classification for keratoconus. Indian J Ophthalmol 2020;68:2831-4. [Crossref] [PubMed]
25. Seiler T, Quurke AW. Iatrogenic keratectasia after LASIK in a case of forme fruste keratoconus. J Cataract Refract Surg 1998;24:1007-9. [Crossref] [PubMed]
26. Abtahi MA, Beheshtnejad AH, Latifi G, et al. Corneal Epithelial Thickness Mapping: A Major Review. J Ophthalmol 2024;2024:6674747. [Crossref] [PubMed]
27. Schiano-Lomoriello D, Bono V, Abicca I, et al. Repeatability of anterior segment measurements by optical coherence tomography combined with Placido disk corneal topography in eyes with keratoconus. Sci Rep 2020;10:1124. [Crossref] [PubMed]
28. Shetty R, Arora V, Jayadev C, et al. Repeatability and agreement of three Scheimpflug-based imaging systems for measuring anterior segment parameters in keratoconus. Invest Ophthalmol Vis Sci 2014;55:5263-8. [Crossref] [PubMed]
29. Lin AN, Mohammed ISK, Munir WM, et al. Inter-rater Reliability and Repeatability of Manual Anterior Segment Optical Coherence Tomography Image Grading in Keratoconus. Eye Contact Lens 2021;47:494-9. [Crossref] [PubMed]
30. Demirbas NH, Pflugfelder SC. Topographic pattern and apex location of keratoconus on elevation topography maps. Cornea 1998;17:476-84. [Crossref] [PubMed]
31. Prakash G, Srivastava D, Choudhuri S, et al. Differences in central and non-central keratoconus, and their effect on the objective screening thresholds for keratoconus. Acta Ophthalmol 2016;94:e118-29. [Crossref] [PubMed]
32. Foster PJ, Baasanhu J, Alsbirk PH, et al. Central corneal thickness and intraocular pressure in a Mongolian population. Ophthalmology 1998;105:969-73. [Crossref] [PubMed]