Anterior segment optical coherence tomography scanning protocols and corneal thickness repeatability
Introduction
Anterior segment optical coherence tomography (AS-OCT) is a non-invasive imaging technique with a range of potential clinical applications [1,2], including; the early detection of subtle corneal pathology (e.g. keratoconus) [[3], [4], [5]], monitoring and predicting potential disease progression (e.g. corneal hydrops) [6], pre and post intervention corneal profiling (e.g. morphological or optical changes following corneal collagen cross-linking or surgery) [5] and scleral contact lens fitting [1]. Scleral lenses have been successfully fitted for over a century using trial lenses, patient feedback, and careful slit lamp observation. However in recent years AS-OCT has also been used to further the understanding of scleral lens thickness profiles [7], variations in lens centration [8] and central [[9], [10], [11], [12]], mid-peripheral [13] or limbal corneal clearance over time [14], scleral topography and elevation [15,16], and the interaction between the haptic landing zone with the underlying conjunctival and scleral tissue [17,18]. Consequently, in modern scleral lens practice, many practitioners utilise AS-OCT to aid initial trial lens selection, assess the initial fit, and monitor changes over time.
An OCT creates a 2- or 3- dimensional cross-sectional image by measuring the echo time delay of light backscattered from tissue structures [2]. A B-scan is a lateral beam of light and is composed of multiple axial scans (A-scan) [4,5]. Instruments vary with respect to default settings but can range from 1 to 100 B-scans. Studies examining corneal thickness have used various scan protocols ranging from 4 to 25 radial scans [[19], [20], [21], [22]], thickness maps up to 9 mm in diameter [23,24], corneal cross lines [25,26] and multiple horizontal or volumetric scans [27].
Several recent studies have also used high resolution AS-OCT (typically 3−5 μm axial resolution) to reliably quantify the nature and time course of scleral lens induced corneal oedema in healthy eyes [28,29], and the influence of altering modifiable lens parameters upon corneal swelling (e.g. corneal clearance, material Dk/t) [30,31]. Other studies have utilised Pentacam imaging [32,33] to measure corneal thickness, however this required lens removal prior to image capture and will underestimate the true magnitude of oedema, as deswelling commences immediately following lens removal.
These repeated measure experiments often utilise different imaging protocols to derive corneal thickness, and therefore, corneal oedema values; for example, a single point estimate from a radial line scan centred on the corneal apex [28,31], or a central thickness value averaged across the central 4 mm of the cornea centred on the pupil derived from 12 radial line scans [29]. Despite substantial differences between the imaging protocols outlined in the above examples, these two studies [28,29] yielded remarkably similar results of 1.65 % [28] and 1.18 % [29] total corneal oedema, when averaged across the study participants. However, no study has systematically investigated the influence of AS-OCT imaging protocols (e.g. the optimum number of line scans, B-scans per line scan, or width of the line scan) upon the reliability of corneal thickness measurements. Optimising AS-OCT imaging parameters is critical for repeated measure experimental studies of corneal thickness (since the measurement error or variability directly influences the required sample size) and also in clinical contact lens practice (e.g. to reliably quantify the short-term response to hypoxic stress over a few hours during scleral lens wear, or to monitor longer-term changes in corneal thickness over a number of years).
Averaging multiple B-scan images improves the signal-to-noise ratio in OCT images [34], and in theory, averaging the maximum possible number of B-scans (limited by instrument parameters) should reduce measurement noise and optimise tissue visibility. However, increasing the number of B-scans that are averaged also extends the overall scan duration, and therefore may increase subject fatigue and eye movements during image acquisition. While a small number of studies have examined the influence of B-Scan averaging on OCT images of the posterior eye (retina and choroid) [[35], [36], [37]], currently no studies have investigated B-Scan optimisation for corneal imaging. Therefore, the aim of this study was to determine the influence of OCT B-Scan averaging, line scan averaging, and line scan width (a single central point compared to an average across the central 6 mm) on epithelial, stromal, and total corneal thickness intraobserver and intrasession repeatability.
Section snippets
Participants
This study was approved by the Queensland University of Technology human research ethics committee and followed the tenets of the Declaration of Helsinki. All participants provided informed consent. Fifteen young healthy participants aged between 20–37 years with no ocular pathology and visual acuity of 0.00 logMAR or better in both eyes were recruited. Participants with any history of ocular injury, surgery, regular rigid contact lens wear or current use of topical medications were excluded.
Central corneal thickness
As expected, corneal thickness varied significantly based on the metric used. For the epithelium, the single point metric overestimated thickness by 1.8 ± 0.4 μm (2.9 %) compared to the average metric (p < 0.001). Conversely, the single point metric underestimated the thickness value compared to average metric for both the stromal (23 ± 1 μm, 4.9 %) and total corneal thickness (21 ± 1 μm, 4.0 %) (both p < 0.001). Epithelial and stromal thickness values did not vary significantly with the number
Discussion
The key finding from this study examining AS-OCT images of healthy eyes was that intraobserver and intrasession repeatability can be improved by averaging 20 B-scans per line scan, and by averaging three line scans within a volumetric scan, but any further reduction in measurement error by increasing the number of B-scans or line scans averaged is negligible. For the single line scan analysis (B-scan optimisation), the intraobserver repeatability did not vary significantly with respect to the
Conclusion
For measures of epithelial, stromal, and total corneal thickness, intraobserver or intrasession repeatability did not vary significantly with the number of B-scans averaged per single line scan, or in relation to the metric used to derive the thickness value (single point or average). However, since the intraobserver repeatability turning point (best intraobserver agreement) appears to occur for approximately 20–30 averaged B-scans, at least 20 B-scans should be averaged per line scan to
Declaration of Competing Interest
None.
References (56)
- et al.
Anterior segment optical coherence tomography
Prog Retin Eye Res
(2018) - et al.
Keratoconus diagnosis with optical coherence tomography pachymetry mapping
Ophthalmology
(2008) - et al.
Anatomic predictive factors of acute corneal hydrops in keratoconus: an optical coherence tomography study
Ophthalmology
(2015) - et al.
Scleral contact lens thickness profiles: the relationship between average and centre lens thickness
Cont Lens Anterior Eye
(2019) - et al.
The temporal dynamics of miniscleral contact lenses: central corneal clearance and centration
Cont Lens Anterior Eye
(2018) - et al.
Change in vault during scleral lens trials assessed with anterior segment optical coherence tomography
Cont Lens Anterior Eye
(2017) - et al.
Variation of clearance considering viscosity of the solution used in the reservoir and following scleral lens wear over time
Cont Lens Anterior Eye
(2017) - et al.
Scleral topography analysed by optical coherence tomography
Cont Lens Anterior Eye
(2017) - et al.
An analysis of anterior scleral shape and its role in the design and fitting of scleral contact lenses
Cont Lens Anterior Eye
(2018) - et al.
Morphological changes in the conjunctiva, episclera and sclera following short-term miniscleral contact lens wear in rigid lens neophytes
Cont Lens Anterior Eye
(2016)
Repeatability of automatic measurements by a new anterior segment optical coherence tomographer combined with placido topography and agreement with 2 Scheimpflug cameras
J Cataract Refract Surg
Repeatability of topographic corneal thickness in keratoconus comparing Visante OCT and Oculus Pentacam HR topographer
Cont Lens Anterior Eye
The time course and nature of corneal oedema during sealed miniscleral contact lens wear
Cont Lens Anterior Eye
Effects of scleral-lens oxygen transmissibility on corneal thickness: a pilot study
Cont Lens Anterior Eye
Corneal changes following short-term miniscleral contact lens wear
Cont Lens Anterior Eye
Anterior segment biometry with the Pentacam: comprehensive assessment of repeatability of automated measurements
J Cataract Refract Surg
Central and peripheral corneal thickness measured with optical coherence tomography, Scheimpflug imaging, and ultrasound pachymetry in normal, keratoconus-suspect, and post-laser in situ keratomileusis eyes
J Cataract Refract Surg
Bilateral corneal epithelial macrocysts following hyperopic laser in situ keratomileusis
J Cataract Refract Surg
Corneal oedema in a unilateral corneal graft patient induced by high dk mini-scleral contact lens
Cont Lens Anterior Eye
The influence of centre thickness on miniscleral lens flexure
Cont Lens Anterior Eye
Optical considerations for scleral contact lenses: a review
Cont Lens Anterior Eye
Optical coherence tomography and scleral contact lenses: clinical and research applications
Clin Exp Optom
Clinical and research applications of anterior segment optical coherence tomography—a review
Graefes Arch Clin Exp Ophthalmol
Anterior segment optical coherence tomography and its clinical applications
Clin Exp Optom
Miniscleral lens wear influences corneal curvature and optics
Ophthalmic Physiol Opt
Oxygen tension beneath scleral lenses of different clearances
Optom Vis Sci
Change in over-refraction after scleral lens settling on average corneas
Ophthalmic Physiol Opt
Effect of time on scleral lens settling and change in corneal clearance
Optom Vis Sci
Cited by (12)
Patch-based CNN for corneal segmentation of AS-OCT images: Effect of the number of classes and image quality upon performance
2023, Computers in Biology and MedicineCitation Excerpt :The model proposed in this study was trained with different image quality arrangements and compared with state-of-the art image classification networks including VGG16 and ResNet50 [24]. AS-OCT images used in this study were obtained from previously published work [3]. Thirty young healthy participants aged between 20 and 37 years with no ocular pathology and normal visual acuity were recruited.
Structural changes of corneal epithelium in belantamab-associated superficial keratopathy using anterior segment optical coherence tomography
2021, American Journal of Ophthalmology Case ReportsCitation Excerpt :AS-OCT is widely used for diagnostic and follow-up purposes in clinical practice. It enables quantitative measurement of corneal epithelial, stromal, and endothelial thicknesses.11,12 Given the fact that BASK was commonly observed in patients receiving belantamab mafodotin, it is important to evaluate this novel finding objectively.
Segmentation of anterior segment boundaries in swept source OCT images
2021, Biocybernetics and Biomedical EngineeringCitation Excerpt :Anterior segment optical coherence tomography (AS-OCT) imaging is an invaluable tool to image the anterior segment of the eye, due to its high-speed and high-resolution capabilities [1,2]. This imaging method allows estimation of the morphology of the anterior segment tissues including the cornea [3–5], crystalline lens [6], anterior chamber [7,8], conjunctiva and sclera [9–11], and contact lenses [12,13]. Additionally, this imaging modality allows the assessment of dynamic changes within the tissue, such as the microvascular properties of the limbal region [14], trabecular meshwork pulsation [15] or pulsatile change in the anterior segment [16].
Conjunctival prolapse during open eye scleral lens wear
2021, Contact Lens and Anterior EyeCitation Excerpt :A volumetric scanning protocol was used which included three horizontal line scans centred on the pupil. Each line scan consisted of 20 B scans separated vertically by 139 μm [26]. Central OCT images were captured while the participant fixated the instruments internal target, and peripheral imaging of the limbus and landing zone were conducted while the participant fixated external targets positioned 26.5 degrees nasally and temporally.
Corneal oedema during reverse piggyback scleral lens wear
2023, Ophthalmic and Physiological OpticsCurrent Trends in Scleral Lens Prescription, Management, and Evaluation
2023, Eye and Contact Lens