Comparison of manual and artificial intelligence-automated choroidal thickness segmentation of optical coherence tomography imaging in myopic adults.

BACKGROUND: Myopia affects 1.4 billion individuals worldwide. Notably, there is increasing evidence that choroidal thickness plays an important role in myopia and risk of developing myopia-related conditions. With the advancements in artificial intelligence (AI), choroidal thickness segmentation can now be automated, offering inherent advantages such as better repeatability, reduced grader variability, and less reliance for manpower. Hence, we aimed to evaluate the agreement between AI-automated and manual segmented measurements of subfoveal choroidal thickness (SFCT) using two swept-source optical coherence tomography (OCT) systems. METHODS: Subjects aged ≥ 16 years, with myopia of ≥ 0.50 diopters in both eyes, were recruited from the Prospective Myopia Cohort Study in Singapore (PROMYSE). OCT scans were acquired using Triton DRI-OCT and PLEX Elite 9000. OCT images were segmented both automatically with an established SA-Net architecture and manually using a standard technique with adjudication by two independent graders. SFCT was subsequently determined based on the segmentation. The Bland-Altman plot and intraclass correlation coefficient (ICC) were used to evaluate the agreement. RESULTS: A total of 229 subjects (456 eyes) with mean [± standard deviation (SD)] age of 34.1 (10.4) years were included. The overall SFCT (mean ± SD) based on manual segmentation was 216.9 ± 82.7 µm with Triton DRI-OCT and 239.3 ± 84.3 µm with PLEX Elite 9000. ICC values demonstrated excellent agreement between AI-automated and manual segmented SFCT measurements (PLEX Elite 9000: ICC = 0.937, 95% CI: 0.922 to 0.949, P < 0.001; Triton DRI-OCT: ICC = 0.887, 95% CI: 0.608 to 0.950, P < 0.001). For PLEX Elite 9000, manual segmented measurements were generally thicker when compared to AI-automated segmented measurements, with a fixed bias of 6.3 µm (95% CI: 3.8 to 8.9, P < 0.001) and proportional bias of 0.120 (P < 0.001). On the other hand, manual segmented measurements were comparatively thinner than AI-automated segmented measurements for Triton DRI-OCT, with a fixed bias of - 26.7 µm (95% CI: - 29.7 to - 23.7, P < 0.001) and proportional bias of - 0.090 (P < 0.001). CONCLUSION: We observed an excellent agreement in choroidal segmentation measurements when comparing manual with AI-automated techniques, using images from two SS-OCT systems. Given its edge over manual segmentation, automated segmentation may potentially emerge as the primary method of choroidal thickness measurement in the future.

Myopia control: Seeing beyond efficacy.

SIGNIFICANCE: The availability of a range of effective myopia control modalities enables the clinician to exercise judgment when discussing the treatment plan with the patient and their parents. This article outlines important considerations beyond efficacy.Clinically meaningful myopia control may be attained with some spectacle lenses, select soft contact lenses, some concentrations of atropine, and overnight orthokeratology. Given that satisfactory efficacy can be achieved with a range of modalities, other factors should be considered when deciding upon the best intervention for a given child. Four key factors-compliance, quality of vision, quality of life, and safety-are discussed in this review. Compliance directly impacts efficacy regardless of the modality and is the most important consideration, as it is influenced by quality of vision and comfort. Daily disposal myopia control contact lenses and overnight orthokeratology are generally associated with high compliance, provide better vision-related quality of life than spectacles, and carry a very low risk when used appropriately. A further benefit of overnight orthokeratology is the elimination of a need for optical correction during the day.

Juvenile-onset myopia-who to treat and how to evaluate success.

The risk of eye diseases such as myopic macular degeneration increases with the level of myopia, but there is no safe level of myopia and the burden of lower degrees of myopia remains considerable. Effective treatments are available that slow progression and thus limit the final degree of myopia. In this review, the rationale for slowing progression is summarized, and a case made for treating all myopic children. Measurement of refractive error and axial length is reviewed, stressing the precision of optical biometry, but also the need for cycloplegic autorefraction. The factors influencing progression are considered and the available tools for interpretation of progression rate are discussed. Finally, the need to set attainable treatment goals is emphasized.

The underestimated role of myopia in uncorrectable visual impairment in the United States.

We estimate the US prevalence of uncorrectable visual impairment in 2050 accounting for the changing distribution of both age and myopia. Age projections of the US population (from an estimated total of 379 million in 2050), were taken from the US census website. The distribution of myopia, by severity, was calculated from literature-derived prevalence estimates of 58.4% (≤ - 0.50 D, 2050 projection) and 33.1% (≤ - 1.00 D, 1999-2004 estimate) to provide predicted and conservative estimates, respectively. Uncorrectable visual impairment as a function of age and refractive error was modelled by multiple linear regression. Finally, the likely number of individuals in the US with visual impairment in 2050 was calculated. For a projected myopia prevalence of 58.4%, 222 million are projected to be myopic and 48 million will have high myopia (- 5 D or worse). The projected total number with uncorrectable visual impairment is 11.4 million of which 4.9 million cases (43%) of visual impairment will be directly attributed to increased risk of eye disease associated with myopia. For a projected myopia prevalence of 33.1%, 8.9 million are projected to have uncorrectable visual impairment of which 2.4 million cases (27%) will be directly attributed to myopia. It is predicted that between 27 and 43% of uncorrectable visual impairment in the US population in 2050 will be directly attributable to myopia. Failure to account for the increasing prevalence of myopia among the aging population leads to a substantial underestimate of the prevalence of visual impairment.

Impact of zone geometry on the introduction of myopic defocus in young adult eyes wearing multi-zone lenses.

PURPOSE: Multizone contact lenses control myopia progression by proposed introduction of myopic defocus. This project investigated how much of the pupil area and how many dioptres of myopic defocus are introduced by different lens zone geometries with near- and off-axis viewing. METHODS: Ten young myopic adults (18-25 years) binocularly wore four soft contact lenses including a single vision (SV), concentric-ring dual-focus (DF), centre-distance multifocal (MF) and a RingBoost™ (RB) multi-zone design containing a combination of coaxial and non-coaxial zones. A modified aberrometer captured aberrations and pupil sizes at four target vergences between -0.25 and -4.00 D (on-axis) and across the central ±30° of the horizontal retina (off-axis). Defocus was quantified as the difference between the measured refractive state and the target vergence within each zone of a multi-zone design within the pupil and compared with that of equivalent zone areas of the SV lens. The percentage of the pupil containing myopic defocused light for each lens was calculated. RESULTS: Defocus within the distance correction zones of multi-zone lenses was similar to that of the SV lens. When viewing on-axis at -0.25 D target vergence, on average 11% of the pupil was myopic with SV, whereas 62%, 84% and 50% of the pupil was myopic for the DF, MF and RB designs, respectively. At -4.00 D target vergence, all lenses exhibited a systematic decrease in the percentage of pupil area having myopic defocus (SV: 3%; DF: 18%; MF: 5% and RB: 26%). The off-axis proportions were similar across multi-zone lenses; however, multi-zone lenses retained approximately 1.25-3.0× more myopic defocus than the SV lens. CONCLUSIONS: Subjects accommodated using the distance-correction zones of multi-zone lenses. Multi-zone contact lenses introduced significant myopic defocus on-axis and across the central ±30° retina. However, the magnitude and proportion of defocus were influenced by zone geometry, add power and pupil size.

Applications of Genomics and Transcriptomics in Precision Medicine for Myopia Control or Prevention.

Myopia is a globally emerging concern accompanied by multiple medical and socio-economic burdens with no well-established causal treatment to control thus far. The study of the genomics and transcriptomics of myopia treatment is crucial to delineate disease pathways and provide valuable insights for the design of precise and effective therapeutics. A strong understanding of altered biochemical pathways and underlying pathogenesis leading to myopia may facilitate early diagnosis and treatment of myopia, ultimately leading to the development of more effective preventive and therapeutic measures. In this review, we summarize current data about the genomics and transcriptomics of myopia in human and animal models. We also discuss the potential applicability of these findings to precision medicine for myopia treatment.

The future of clinical trials of myopia control.

In the field of myopia control, effective optical or pharmaceutical therapies are now available to patients in many markets. This creates challenges for the conduct of placebo-controlled, randomised clinical trials, including ethics, recruitment, retention, selective loss of faster progressors and non-protocol treatments: Ethics: It is valid to question whether withholding treatment in control subjects is ethical. Recruitment: Availability of treatments is making recruitment into clinical trials more difficult. Retention: If masking is not possible, parents may immediately withdraw their child if randomised to no treatment. Selective loss: Withdrawal of fast progressors in the control group leading to a control group biased towards low progression. Non-protocol treatment: Parents may access other myopia treatments in addition to those within the trial. We propose that future trials may adopt one of the following designs: Non-inferiority trials using an approved drug or device as the control. The choice will depend on whether a regulatory agency has approved the drug or device. Short conventional efficacy trials where data are subsequently entered into a model created from previous clinical trials, which allows robust prediction of long-term treatment efficacy from the initial efficacy. Virtual control group trials based on data relating to axial elongation, myopia progression or both, accounting for subject's age and race. Short-term control data from a cohort, for example, 1 year or less, and applying an appropriate, proportional annual reduction in axial elongation to that population and extrapolating to subsequent years. Time-to-treatment-failure trials using survival analysis; once a treated or control subject progresses or elongates by a given amount, they exit the study and can be offered treatment. In summary, the future development of new treatments in myopia control will be hampered if significant changes are not made to the design of clinical trials in this area.

A Duration-Dependent Interaction Between High-Intensity Light and Unrestricted Vision in the Drive for Myopia Control.

Purpose: To evaluate the duration-dependent and synergetic impact of high-intensity light (HL) and unrestricted vision (UnV) on lens-induced myopia (LIM) development in chickens. Methods: Myopia was induced in one eye in chicks (10 groups, n = 126) from day 1 posthatching (D1) until day 8 (D8) using -10 diopter (D) lenses. Fellow eyes remained uncovered as controls. Nine groups were exposed daily to 2, 4, or 6 hours of HL (15,000 lux), UnV (removal of -10 D lens), or both (HL + UnV). One group served as the LIM group without any interventions. Ocular axial length (AL), refractive error, and choroidal thickness were measured on D1, D4, and D8. Outcome measures are expressed as interocular difference (IOD = experimental eye - control eye) ± SEM. Results: By D8, LIM increased AL (0.36 ± 0.04 mm), myopic refraction (-9.02 ± 0.37 D), and choroidal thinning (-90.27 ± 16.44 µm) in the LIM group (all, P < 0.001). Compared to the LIM group, exposure to 2, 4, or 6 hours of HL, UnV, or HL + UnV reduced myopic refraction in a duration-dependent manner, with UnV being more effective than HL (P < 0.05). Only 6 hours of HL + UnV (not 2 or 4 hours) prevented LIM and was more effective than UnV (P = 0.004) or HL (P < 0.001) in reducing myopic refraction and more effective than HL (P < 0.001) in reducing axial elongation. Conclusions: Daily exposure to 2, 4, or 6 hours of HL, UnV, or HL + UnV reduced lens-induced myopic refraction in a duration-dependent manner in chickens. Only 6 hours of HL + UnV completely stopped LIM development. The synergetic effect of HL and UnV is dependent on the duration of the interventions.

Randomized Trial of Soft Contact Lenses with Novel Ring Focus for Controlling Myopia Progression.

Purpose: To evaluate efficacy and vision with 2 prototype myopia control soft contact lenses with noncoaxial ring-focus designs (for enhancing efficacy [EE] and enhancing vision [EV]) compared with dual-focus (DF) and single-vision (SV) designs. Design: Multicenter, 6-month, randomized, controlled, double-masked clinical trial. Participants: One hundred ninety-nine myopic (-0.75 diopters [D] to -4.50 D) children aged 7 to 12 years. Methods: Participants were randomized with stratification into myopia control (EE, EV, or DF) or SV arms at 9 clinical sites in 3 countries. Postcycloplegia axial length (AL) and spherical equivalent autorefraction (SECAR) were measured at baseline and 26 weeks. Axial length was also measured without cycloplegia at baseline, 1, 4, 13, and 26 weeks. Progression was analyzed using linear mixed models by intention-to-treat population. Visual acuity (VA) and vision quality were monitored. Main Outcome Measures: Axial elongation, change in SECAR. Results: A total of 185 subjects completed the study (n = 44, 49, 45, and 47 for EE, EV, DF, and SV, respectively). There were no serious/significant ocular adverse events. After 26 weeks, EE, EV, and DF all had statistically significantly less axial elongation than SV (unadjusted mean [standard deviation]: EE, 0.079 [0.125]; EV, 0.119 [0.101]; DF, 0.135 [0.117]; SV; 0.189 [0.121] mm). The estimated least-square mean (LSM) differences (adjusted 95% confidence interval) compared with SV were -0.105 (-0.149, -0.062), -0.063 (-0.106, -0.020), and -0.056 (-0.100, -0.013) mm for EE, EV, and DF, respectively. Enhancing efficacy alone had statistically significantly less progression of SECAR than SV (EE: -0.12 [0.27] D vs. SV: -0.35 [0.33] D; LSM difference: 0.22 D [0.09, 0.35]). Enhancing efficacy also had statistically significantly less axial elongation than DF (-0.049 mm [-0.093, -0.004]). Changes in AL and SECAR of EV and DF were not statistically different. All 3 myopia control lenses had mean VA close to 0.00 logarithm of the minimum angle of resolution (logMAR) with estimated 95% upper confidence limits <0.10 logMAR. Enhancing efficacy and DF produced similar reports of halos but more than EV and SV. Conclusions: The prototype contact lenses met the design intent; EE was more efficacious in slowing axial elongation than DF with comparable vision performance, whereas EV produced comparable efficacy to DF with similar vision performance to SV.

Myopia: An ounce of prevention is worth a pound of cure.

PURPOSE: Myopia severity has a profound impact on visual impairment in later life. A patient's final level of myopia may be lowered by myopia control, but also by delaying onset. Here, we evaluate the influence of the age of onset on the final recorded level of myopia. METHODS: Data were extracted from: (1) Three prospective cohort studies of myopia progression in East Asia and the United States where the final recorded level of myopia is presented as a function of the established age of onset. (2) Four retrospective studies of myopia progression in Finland, India, the Netherlands and China and two cross-sectional studies in Argentina and the UK where the age of onset was based on self-report of age at first spectacle prescription. (3) A cohort study of Finnish subjects originally recruited for a clinical trial and followed into adulthood. Subjects were divided into five groups according to age at recruitment that was used as a surrogate for the age of onset. RESULTS: Final recorded level of myopia was plotted as a function of age of onset for all studies. Among the three East Asian studies, the slopes are between 0.68 and 0.97 D/year, meaning that each later year of onset is associated with between 0.68 and 0.97 less myopia at the final recorded refraction. For six of the seven non-East Asian studies, the slopes are substantially flatter, with slopes between 0.23 and 0.50 D/year. By contrast, the slope for the Finnish study was 0.87 D/year. Increasing age of final recorded refraction tended to be associated with higher levels of myopia. CONCLUSION: Among East Asians, delaying the onset of myopia by 1 year has the potential to lower the final myopia level by 0.75 D or more-equivalent to 2-3 years of myopia control with existing modalities. The benefit is lower, but meaningful, among non-East Asians.

Choroidal Thickness in Early Postnatal Guinea Pigs Predicts Subsequent Naturally Occurring and Form-Deprivation Myopia.

Purpose: To identify choroidal characteristics associated with susceptibility to development of naturally occurring and experimentally induced myopia. Methods: We compared choroidal properties between pigmented and albino guinea pig (GP) strains. Biometry, cycloplegic refractive error (RE), and eye wall sublayer thickness were measured from 171 GPs at postnatal day (P)6, 14, and 28. Forty-three P14 GPs underwent two-week monocular form-deprivation myopia (FDM). En face images of choroidal vasculature were obtained with a customized swept-source optical coherence tomography. Multivariate regression analyses were performed, with P28 RE as the outcome and P14 choroidal thickness (ChT) as the main predictor variable. Proteomic analysis was performed on choroidal tissue from P14 albino and pigmented GPs. Results: At P14, RE was correlated with thickness of the choroid (ß = 0.06), sclera (ß = 0.12), and retina (ß = 0.27; all P < 0.001). P14 ChT was correlated with P28 RE both with (ß = 0.06, P = 0.0007) and without FDM (ß = 0.05, P = 0.008). Multivariate regression analysis, taking into account FDM (versus physiological growth) and strain, revealed that for every 10-µm greater ChT at P14, P28 RE was 0.50D more positive (P = 0.005, n = 70). En face images of choroidal sublayers showed that albino choroids were relatively underdeveloped, with frequent avascular regions. Consistent with this finding, proteomic analysis suggested abnormalities of the nitric oxide system in the albino GP choroid. Conclusions: Current results are consistent with the notion that greater ChT could protect from or delay the onset of myopia, while lower ChT is associated with greater susceptibility to myopia development. The underlying mechanism could be related to dysfunction of the choroidal vascular system.

The Limited Value of Prior Change in Predicting Future Progression of Juvenile-onset Myopia.

SIGNIFICANCE: Identifying children at highest risk for rapid myopia progression and/or rapid axial elongation could help prioritize who should receive clinical treatment or be enrolled in randomized clinical trials. Our models suggest that these goals are difficult to accomplish. PURPOSE: This study aimed to develop models predicting future refractive error and axial length using children's baseline data and history of myopia progression and axial elongation. METHODS: Models predicting refractive error and axial length were created using randomly assigned training and test data sets from 916 myopic participants in the Collaborative Longitudinal Evaluation of Ethnicity and Refractive Error Study. Subjects were 7 to 14 years of age at study entry with three consecutive annual visits that included cycloplegic A-scan ultrasound and autorefraction. The effect of adding prior change in axial length and refractive error was evaluated for each model. RESULTS: Age, ethnicity, and greater myopia were significant predictors of future refractive error and axial length, whereas prior progression or elongation, near work, time outdoors, and parental myopia were not. The 95% limits for the difference between actual and predicted change were ±0.22 D and ±0.14 mm without prior change data compared with ±0.26 D and ±0.16 mm with prior change data. Sensitivity and specificity for identifying fast progressors were between 60.8 and 63.2%, respectively, when the cut points were close to the sample average. Positive predictive value and sample yield were even lower when the cut points were more extreme. CONCLUSIONS: Young, more myopic Asian American children in the Collaborative Longitudinal Evaluation of Ethnicity and Refractive Error Study were the most likely to progress rapidly. Clinical trials should expect average progression rates that reflect sample demographics and may have difficulty recruiting generalizable samples that progress faster than that average. Knowing progression or elongation history does not seem to help the clinical decision regarding initiating myopia control.

Rapid Myopic Progression in Childhood Is Associated With Teenage High Myopia.

Purpose: The purpose of this study was to evaluate the association of childhood progression of spherical equivalent (SE) with high myopia (HM) in teenagers in the Singapore Cohort of Risk factors for Myopia (SCORM). Methods: We included 928 SCORM children followed over a mean follow-up of 6.9 ± 1.0 years from baseline (6-11 years old) until their teenage years (12-19 years old). Cycloplegic autorefraction and axial length (AL) measurements were performed yearly. The outcomes in teenagers were HM (SE ≤ -5 diopter [D)], AL ≥ 25 mm, SE and AL. Three-year SE and AL progression in childhood and baseline SE and AL with outcomes were evaluated using multivariable logistic or linear regression models, with predictive performance of risk factors assessed using the area under the curve (AUC). Results: At the last visit, 9.8% of teenagers developed HM and 22.7% developed AL ≥ 25 mm. In multivariate regression analyses, every -0.3 D/year increase in 3-year SE progression and every 0.2 mm/year increase in 3-year AL progression were associated with a -1.14 D greater teenage SE and 0.52 mm greater teenage AL (P values < 0.001). The AUC (95% confidence interval [CI]) of a combination of 3-year SE progression and baseline SE for teenage HM was 0.97 (95% CI = 0.95 - 0.98). The AUC of 3-year AL progression and baseline AL for teenage AL ≥ 25 mm was 0.91 (95% CI = 0.89 - 0.94). Conclusions: Three-year myopia progression in childhood combined with baseline SE or AL were good predictors of teenage HM. Clinicians may use this combination of factors to guide timing of interventions, potentially reducing the risk of HM later in life.

Efficacy in myopia control.

There is rapidly expanding interest in interventions to slow myopia progression in children and teenagers, with the intent of reducing risk of myopia-associated complications later in life. Despite many publications dedicated to the topic, little attention has been devoted to understanding 'efficacy' in myopia control and its application. Treatment effect has been expressed in multiple ways, making comparison between therapies and prognosis for an individual patient difficult. Available efficacy data are generally limited to two to three years making long-term treatment effect uncertain. From an evidence-based perspective, efficacy projection should be conservative and not extend beyond that which has been empirically established. Using this principle, review of the literature, data from our own clinical studies, assessment of demonstrated myopia control treatments and allowance for the limitations and context of available data, we arrive at the following important interpretations: (i) axial elongation is the preferred endpoint for assessing myopic progression; (ii) there is insufficient evidence to suggest that faster progressors, or younger myopes, derive greater benefit from treatment; (iii) the initial rate of reduction of axial elongation by myopia control treatments is not sustained; (iv) consequently, using percentage reduction in progression as an index to describe treatment effect can be very misleading and (v) cumulative absolute reduction in axial elongation (CARE) emerges as a preferred efficacy metric; (vi) maximum CARE that has been measured for existing myopia control treatments is 0.44 mm (which equates to about 1 D); (vii) there is no apparent superior method of treatment, although commonly prescribed therapies such as 0.01% atropine and progressive addition spectacles lenses have not consistently provided clinically important effects; (viii) while different treatments have shown divergent efficacy in the first year, they have shown only small differences after this; (ix) rebound should be assumed until proven otherwise; (x) an illusion of inflated efficacy is created by measurement error in refraction, sample bias in only treating 'measured' fast progressors and regression to the mean; (xi) decision to treat should be based on age of onset (or refraction at a given age), not past progression; (xii) the decreased risk of complications later in life provided by even modest reductions in progression suggest treatment is advised for all young myopes and, because of limitations of available interventions, should be aggressive.

Annual Myopia Progression and Subsequent 2-Year Myopia Progression in Singaporean Children.

Purpose: To investigate the association between 1-year myopia progression and subsequent 2-year myopia progression among myopic children in the Singapore Cohort Study of the Risk Factors for Myopia. Methods: This retrospective analysis included 618 myopic children (329 male), 7 to 9 years of age (mean age, 8.0 ± 0.8) at baseline with at least two annual follow-up visits. Cycloplegic autorefraction was performed at every visit. Receiver operating characteristic (ROC) curves from multiple logistic regressions were derived for future fast 2-year myopia progression. Results: Children with slow progression during the first year (slower than -0.50 diopter [D]/y) had the slowest mean subsequent 2-year myopia progression (-0.41 ± 0.33 D/y), whereas children with fast progression (faster than -1.25 D/y) in year 1 had the fastest mean subsequent 2-year myopia progression (-0.82 ± 0.30 D/y) (P for trend < 0.001). Year 1 myopia progression had the highest area under the curve (AUC) for predicting fast subsequent 2-year myopia progression (AUC = 0.77; 95% confidence interval [CI], 0.73-0.80) compared to baseline spherical equivalent (AUC = 0.70; 95% CI, 0.66-0.74) or age of myopia onset (AUC = 0.66; 95% CI, 0.61-0.70) after adjusting for confounders. Age at baseline alone had an AUC of 0.65 (95% CI, 0.61-0.69). Conclusions: One-year myopia progression and age at baseline were associated with subsequent 2-year myopia progression in children 7 to 9 years of age. Translational Relevance: Myopia progression and age at baseline may be considered by eye care practitioners as two of several factors that may be associated with future myopia progression in children.