Vahid Mohammadzadeh, Erica Su, Sajad Besharati, Abraham Liu, Simon K Law, Anne L Coleman, Joseph Caprioli, Robert E Weiss, Kouros Nouri-Mahdavi
{"title":"A Bayesian Hierarchical Longitudinal Model for Estimation of Central Visual Field Rates of Change in Glaucoma.","authors":"Vahid Mohammadzadeh, Erica Su, Sajad Besharati, Abraham Liu, Simon K Law, Anne L Coleman, Joseph Caprioli, Robert E Weiss, Kouros Nouri-Mahdavi","doi":"10.1167/tvst.14.9.24","DOIUrl":"10.1167/tvst.14.9.24","url":null,"abstract":"<p><strong>Purpose: </strong>Individual visual field (VF) sensitivities become unreliable at threshold sensitivities of 19 dB or less, limiting glaucoma monitoring. We evaluated longitudinal variability of central 10° VF measurements based on baseline sensitivity using a Bayesian hierarchical model.</p><p><strong>Methods: </strong>We included 124 glaucoma patients (124 eyes) with central or moderate-to-advanced VF damage, more than 2 years follow-up, and more than 4 central 10-2 VF tests. A Bayesian linear model estimated pointwise change rates, compared with simple linear regression (SLR). Simulations modeled average (-0.21 dB/year) and benchmark (-0.5 dB/year) slopes with residual standard deviations (SD) of 2, 4, 7, or 10 dB. Outcomes included pointwise residual SDs and proportions of significant slopes in cohort and simulations.</p><p><strong>Results: </strong>The average baseline 10-2 VF mean deviation, follow-up time, and median VF tests were 8.4 ± 5.4 dB, 4.6 ± 0.8 years, and 9 VF tests (range, 4-12 VF tests), respectively. The mean global slopes for Bayesian and SLR models were -0.21 and -0.36 dB/year. Residual SDs were markedly higher when baseline threshold sensitivities was 5 to 20 dB compared with 25 dB or greater. The Bayesian model identified more significant negative slopes, particularly at points with residual SD of less than 4 dB, relative to SLR.</p><p><strong>Conclusions: </strong>When baseline pointwise sensitivity is 5 to 20 dB, residual variability is very large, substantially reducing the ability to detect glaucoma progression.</p><p><strong>Translational relevance: </strong>Visual field locations with sensitivity near or less than 20 dB demonstrate markedly greater variability over time; thus, excluding these points from visual field algorithms or analytical models could improve efficiency in detecting perimetric progression.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 9","pages":"24"},"PeriodicalIF":2.6,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12448127/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145070562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Seema Banerjee, Aleksandra Mihailovic, Ayodeji E Sotimehin, Moneesha Rani Mukherjee, Louay Almidani, Xindi Chen, Mariah Diaz, Jennifer A Schrack, Kerry J Stewart, Pradeep Y Ramulu
{"title":"Outcomes of a Physical Activity Intervention in Patients With Glaucoma: A Feasibility Study.","authors":"Seema Banerjee, Aleksandra Mihailovic, Ayodeji E Sotimehin, Moneesha Rani Mukherjee, Louay Almidani, Xindi Chen, Mariah Diaz, Jennifer A Schrack, Kerry J Stewart, Pradeep Y Ramulu","doi":"10.1167/tvst.14.9.25","DOIUrl":"10.1167/tvst.14.9.25","url":null,"abstract":"<p><strong>Purpose: </strong>To determine the feasibility of increasing physical activity (PA) levels in patients with glaucoma and its correlation with improved intraocular pressure (IOP) and visual function.</p><p><strong>Methods: </strong>Thirty-five participants aged 18 years or older, diagnosed with glaucoma and visual field (VF) damage between -5 and -25 decibels (dB) in at least one eye, were included. Participants were given a Fitbit Charge 2 accelerometer and, through personalized text messages, were asked to increase their PA by 500 steps weekly until reaching 10,000 steps/day. VF testing and IOP were performed for each eye during baseline, 3-month, and 6-month study visits.</p><p><strong>Results: </strong>Average baseline steps/week were 4772 ± 2978. By week 7, participants on average increased their activity to 7656 ± 2985 steps/week. Average steps/week between weeks 8 and 26 were 7793 ± 3460. On average, PA increased by 409 steps/week in the first 7 weeks (95% confidence interval [CI], 314-504; P < 0.001), but no further change (β = -15 steps/week; 95% CI, -35 to 5; P = 0.15) was observed. Lower end-of-study IOPs were noted for the better eye (P < 0.01) but not the worse-seeing eye. Each 10% improvement in PA was associated with 0.05 mm Hg (β = -0.05 mm Hg; 95% CI, -0.09 to -0.001; P = 0.05) lowering in IOP of the worse-seeing eye.</p><p><strong>Conclusions: </strong>Patients with glaucoma enrolled in a structured PA intervention increased mean PA by at least 50% over 7 weeks, retaining activity for 6 months, with some IOP declines but no VF changes.</p><p><strong>Translational relevance: </strong>These findings demonstrate that a structured program to improve PA is feasible and may aid in investigating possible beneficial effects of PA on glaucoma.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 9","pages":"25"},"PeriodicalIF":2.6,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12449823/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145076198","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Albert Yang, Jennifer Cano, Benjamin Xu, Mahnaz Shahidi
{"title":"Conjunctival Venular Hemodynamic Biomarkers in Glaucoma.","authors":"Albert Yang, Jennifer Cano, Benjamin Xu, Mahnaz Shahidi","doi":"10.1167/tvst.14.9.21","DOIUrl":"10.1167/tvst.14.9.21","url":null,"abstract":"<p><strong>Purpose: </strong>Vascular dysfunction is considered a factor contributing to glaucomatous damage. Imaging the bulbar conjunctiva offers a noninvasive approach for directly assessing systemic microcirculation. The purpose of the current study was to test the hypothesis that there is an association between alterations in conjunctival venular hemodynamic biomarkers and presence of primary open-angle glaucoma (POAG).</p><p><strong>Methods: </strong>This cross-sectional study included 26 normal control (NC) and 29 POAG subjects. Conjunctival microcirculation was assessed non-invasively by combining video microscopy and an automated software to measure the following conjunctival venular hemodynamic biomarkers: vessel diameter (D), blood velocity (V), blood flow (Q), wall shear rate (WSR), and wall shear stress (WSS).</p><p><strong>Results: </strong>Age differed between NC and POAG groups (P < 0.001), while other ocular and systemic factors were not significantly different (P ≥ 0.05). Conjunctival venular V, WSR, and WSS were significantly reduced in POAG compared to NC, both with and without adjusting for covariates and accounting for multiple comparisons (adjusted P ≤ 0.03). In the adjusted models, the differences in conjunctival venular D and Q between NC and POAG groups were not significant (P ≥ 0.16).</p><p><strong>Conclusions: </strong>The finding of reduced conjunctival venular wall shear stress in POAG supports the vascular theory of glaucoma, implicating endothelial dysfunction. Conjunctival wall shear stress may serve as a novel biomarker for glaucomatous vascular damage.</p><p><strong>Translational relevance: </strong>Assessment of microvasculature wall shear stress may serve as a novel biomarker of endothelium function with the potential to improve the understanding, diagnosis and monitoring of glaucoma and other optic neuropathies.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 9","pages":"21"},"PeriodicalIF":2.6,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12442937/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145065740","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Andrew Turpin, William H Morgan, Allison M McKendrick
{"title":"Personalizing the Transition From 24-2 to 10-2 Visual Field Testing Using ARREST - A Simulation Study.","authors":"Andrew Turpin, William H Morgan, Allison M McKendrick","doi":"10.1167/tvst.14.8.1","DOIUrl":"10.1167/tvst.14.8.1","url":null,"abstract":"<p><strong>Purpose: </strong>The decision to switch between peripheral (24-2) and central (10-2) visual field testing in glaucoma currently rests with clinicians and requires a choice between either intensively sampling central vision or including testing of more peripheral locations (not both). We evaluate an automated approach to incorporate 10-2 locations into the 24-2, without increasing test time or reducing the ability to detect glaucomatous progression at the 24-2 locations.</p><p><strong>Methods: </strong>We applied our previously published ARREST approach, restricting added locations to a 10-2 pattern (A10). Using computer simulation, the sensitivity of A10 for detecting progression was compared to a ZEST procedure on the 24-2 pattern. Input fields were a synthetic series of 10 visits (6 monthly) of 126 eyes, derived from empirical longitudinal 24-2 data from people with glaucoma.</p><p><strong>Results: </strong>Forty-seven of the 126 progressing visual field series had locations added by A10. The procedures used a similar number of presentations (mean ± SD: A10, 220 ± 27; ZEST, 226 ± 29). Area under the curve (AUC) and survival time for detecting progressing series were similar between methods. A10 allowed visualization of the macular visual field with higher fidelity.</p><p><strong>Conclusions: </strong>The A10 approach allows automated incorporation of 10-2 locations into the 24-2 pattern that are customized to individuals without increasing test times.</p><p><strong>Translational relevance: </strong>The A10 enables higher spatial sampling in individually relevant areas of the 10-2, without neglecting testing of the more peripheral visual field.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 8","pages":"1"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12320903/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144761454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Carlos R Cassanello, Neil W Roach, Chris Scholes, Paul V McGraw
{"title":"The Effect of Contrast Reversal on Peripheral Visual Acuity.","authors":"Carlos R Cassanello, Neil W Roach, Chris Scholes, Paul V McGraw","doi":"10.1167/tvst.14.8.23","DOIUrl":"10.1167/tvst.14.8.23","url":null,"abstract":"<p><strong>Purpose: </strong>Patients with central vision loss must rely upon regions of their unaffected peripheral retina to carry out fine-scale visual tasks. We aimed to determine the conditions under which contrast reversal of targets and/or backgrounds may act to improve visual acuity in the peripheral field.</p><p><strong>Methods: </strong>Ten participants with normal or corrected-to-normal vision performed a visual acuity task in which they were asked to identify the orientation of a letter-C target presented at 10° in the right visual field. Minimum letter size thresholds were measured for static (black or white) and contrast-reversing (8.5 or 17 Hz) targets presented on a uniform mean luminance background, a static patterned background, or a dynamic version of the patterned background, which was contrast reversing at the same temporal frequencies used for the targets.</p><p><strong>Results: </strong>When presented on a uniform background, black targets produced lower thresholds than white targets, and there was no consistent benefit of contrast reversal. However, when presented on a patterned background, acuity for contrast-reversing targets consistently exceeded that for either polarity static target. Contrast reversing the background also improved acuity, for both static targets and targets that contrast reversed at a different rate.</p><p><strong>Conclusions: </strong>Peripheral visual acuity is significantly enhanced by dynamic contrast modulation, particularly when the modulation acts to promote segmentation between target and background.</p><p><strong>Translational relevance: </strong>Adding temporal modulation to the target or background could provide a simple way of enhancing peripheral acuity for patients with central vision loss.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 8","pages":"23"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12369918/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144875410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Anterior Chamber Angle Changes and Its Associated Factor After Posterior Chamber Phakic Intraocular Lens Implantation in Eyes With Shallow Anterior Chambers.","authors":"Shengtao Liu, Fang Liu, Mingrui Cheng, Chiwen Cheng, Xiaoying Wang, Xingtao Zhou","doi":"10.1167/tvst.14.8.41","DOIUrl":"https://doi.org/10.1167/tvst.14.8.41","url":null,"abstract":"<p><strong>Purpose: </strong>The purpose of this study was to investigate anterior chamber angle (ACA) changes after implantable collamer lens (ICL) placement in eyes with shallow anterior chamber depth (ACD) and the effect of vault size on ACA changes.</p><p><strong>Methods: </strong>In this prospective cohort study of 120 patients (120 eyes) undergoing ICL implantation, participants were stratified into shallow (ACD ≤3.0 mm, n = 60) and normal ACD groups (ACD >3.0 mm, n = 60). Trabecular-iris angle 500 (TIA500), angle opening distance 500 (AOD500), trabecular-iris space area (TISA500), angle opening distance circumference area 500 (AODA500), trabecular-iris circumference volume 500 (TICV500), and anterior chamber volume (ACV) measured by swept source optical coherence tomography (SS-OCT) preoperatively and at 3 months postoperatively were compared within and between groups and according to vault size.</p><p><strong>Results: </strong>Relative to the preoperative values, the TIA500, AOD500, and TISA500 values decreased significantly in both groups (all P < 0.01). Similarly, the AODA500, TICV500, and ACV values decreased by 52.1%, 48.2%, and 38.9%, respectively, in the shallow ACD group (all P < 0.001), and by 59.2%, 56.6%, and 37.4%, respectively, in the normal ACD group (all P < 0.001). However, the reduction in all ACA parameters in the shallow ACD group in each meridian was significantly lower than that in the normal ACD group, both for low (all P ≤ 0.005) and high vaults (all P ≤ 0.004). Correlation analyses indicated that vault size was positively correlated with the decrease in ACA parameters in both groups (all P < 0.05).</p><p><strong>Conclusions: </strong>Anterior segment structure was more crowded in patients with a shallow ACD. This necessitates strict ICL size limits (≤12.2 mm for ACD <2.8 mm) and prioritizes long-term SS-OCT surveillance of the superior-nasal quadrant, particularly in high-vault cases.</p><p><strong>Translational relevance: </strong>For patients with a shallow anterior chamber implanted ICL, placement in the superior-nasal quadrant should be considered.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 8","pages":"41"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12400972/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144970392","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Morphological Characteristics of the Meibomian Glands: Impact of Demographics and Contact Lens Wear.","authors":"Andrew D Graham, Chun-Hsiao Yeh, Meng C Lin","doi":"10.1167/tvst.14.8.36","DOIUrl":"https://doi.org/10.1167/tvst.14.8.36","url":null,"abstract":"<p><strong>Purpose: </strong>To determine the impact of age, sex, ethnicity, and contact lens wear on the detailed morphology of the Meibomian glands as quantified by a deep learning segmentation model.</p><p><strong>Methods: </strong>A large dataset of meibography images (n = 2233) from 560 subjects was compiled and input to a supervised machine learning model to quantify gland length, width, tortuosity, contrast, atrophy, density, and number of glands. These morphology outcomes were modeled as functions of age, sex, ethnicity, and contact lens wear parameters.</p><p><strong>Results: </strong>Age was significantly associated with shorter glands, more atrophy, and lower gland density (all p < 0.001). No Meibomian gland morphological characteristics were related to sex. Asian subjects exhibited the longest and densest glands, and black subjects exhibited the most gland atrophy. Although contact lens wearers overall had significantly longer glands (∼4%-5%; p < 0.001) than non-wearers, no other contact lens wear parameter was significantly related to any Meibomian gland morphological feature.</p><p><strong>Conclusions: </strong>What constitutes \"normal-looking\" Meibomian glands in a meibography image depends on the age and ethnicity of the patient. There appear to be no significant female/male Meibomian gland morphological differences. Meibomian gland morphology is robust to contact lens wear based on a large-sample analysis of younger, successful contact lens wearers. Now that the impact of these external factors has been established, work is ongoing to determine exactly what alterations in Meibomian gland morphology contribute to downstream pathology.</p><p><strong>Translational relevance: </strong>This study used artificial intelligence to provide clinicians with novel insights into normal versus abnormal Meibomian gland morphological features in their patients.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 8","pages":"36"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12395806/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144970490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zengshuo Wang, Haohan Zou, Yin Guo, Minghe Sun, Xin Zhao, Yan Wang, Mingzhu Sun
{"title":"Myopic-Net: Deep Learning-Based Direct Identification of Myopia Onset and Progression.","authors":"Zengshuo Wang, Haohan Zou, Yin Guo, Minghe Sun, Xin Zhao, Yan Wang, Mingzhu Sun","doi":"10.1167/tvst.14.8.38","DOIUrl":"https://doi.org/10.1167/tvst.14.8.38","url":null,"abstract":"<p><strong>Purpose: </strong>Identifying and monitoring the onset and progression of myopia (myopia onset and progression [MOP]) based on the changes in anatomical structures in fundus retinal images has significant clinical application prospects. For this purpose, we tested the performance of deep neural networks.</p><p><strong>Methods: </strong>We established a deep neural network, called Myopic-Net, to detect anatomical changes owing to the MOP from a pair of retinal images collected during different fundoscopies. Myopic-Net was developed using 3964 fundus image pairs without MOP and 2380 fundus image pairs with MOP. Five indicators-accuracy, precision, recall, specificity, and F1-score-were evaluated on the internal test set and the independent external test set. In addition, we use a deep network visualization method to explore the factors driving Myopic-Net to predict.</p><p><strong>Results: </strong>On the internal test set, Myopic-Net achieved an accuracy of 87.3%; the precision, recall, and specificity were 86.2%, 80.1%, and 91.9% respectively, while the identification accuracy of two ophthalmologists is only 66.1% and 73.5%, respectively. Even on the external test set, Myopic-Net still achieved an accuracy of 84.1%. In addition, we found that the factors driving Myopic-Net to predict are mainly anatomical changes in the optic disc and surrounding areas.</p><p><strong>Conclusions: </strong>Myopic-Net has been shown to be able to identify the MOP from fundus image pairs using anatomical changes in optic disc and surrounding areas. And Myopic-Net has good accuracy, reliability, and generalization ability. These factors show that deep neural networks have strong potential in monitoring and final diagnosing the MOP based on fundus image analysis.</p><p><strong>Translational relevance: </strong>With the development of fundus imaging technology based on intelligent mobile terminals, embedding the program based on Myopic-Net has great potential to achieve convenient and fast personalized monitoring of myopia.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 8","pages":"38"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12395868/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144970535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sebastiano Del Fabbro, Alessandro Arrigo, Lorenzo Bianco, Alessio Antropoli, Annamaria Nunziata, Ugo Introini, Daniele Giusto, Francesco Bandello, Maurizio Battaglia Parodi
{"title":"Dome-Shaped Macula Curvature: A New Quantitative Metric to Assess Dome-Shaped Macula Subtypes.","authors":"Sebastiano Del Fabbro, Alessandro Arrigo, Lorenzo Bianco, Alessio Antropoli, Annamaria Nunziata, Ugo Introini, Daniele Giusto, Francesco Bandello, Maurizio Battaglia Parodi","doi":"10.1167/tvst.14.8.9","DOIUrl":"10.1167/tvst.14.8.9","url":null,"abstract":"<p><strong>Purpose: </strong>To test a new quantitative metric focused on the evaluation of dome-shaped macula curvature (DSMC); to assess its clinical role in evaluating round (R-DSM), vertical (V-DSM), and horizontal (H-DSM) DSM types; and to examine its relationships with other ocular parameters.</p><p><strong>Methods: </strong>Patients affected by DSM secondary to high myopia were included in this cross-sectional, observational case study. DSMC was calculated by measuring the base (chord length) and height (distance from the chord to the apex) of the DSM. These measurements were used to determine the radius (R) of the hypothetical circle representing the curvature of the DSM. Curvature (K) was then calculated as the reciprocal of R. In addition, we measured choroidal thickness (CT), best-corrected visual acuity (BCVA), and spherical equivalent (SE) to assess its potential relationship with DSMC.</p><p><strong>Results: </strong>A total of 125 eyes from 107 patients (25 males, 23%) were included, with a mean age of 65.67 ± 13.51 years. DSM subtypes-categorized as 57 R-DSM eyes (45.6%), 58 H-DSM eyes (46.4%), and 10 V-DSM eyes (8%)-showed significantly different DSMC (P = 0.039), with R-DSM showing the highest DSMC values. Conversely, V-DSM and H-DSM were characterized by similar DSMC. Subfoveal CT showed a strong inverse correlation with DSMC (P < 0.0001; r = -0.60), with a consistent reduction in CT as curvature increased. DSMC was also significantly associated with SE, with higher DSMC values corresponding to greater myopic refractive errors (P = 0.009; r = -0.21).</p><p><strong>Conclusions: </strong>DSMC may be considered a reliable and reproducible quantitative metric to analyze DSM subtypes. The association of DSMC with CT highlights the influence of scleral curvature on choroidal morphology and trophism, paving the way for future new outcome measures.</p><p><strong>Translational relevance: </strong>DSMC provides an objective and reproducible method to evaluate DSM morphology across subtypes (round, vertical, and horizontal). By quantifying curvature through OCT-based parameters, this metric enables standardized comparisons and enhances the clinical characterization of DSM. Furthermore, the study provides insights into the disease mechanisms by uncovering significant associations between DSMC and choroidal thickness.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 8","pages":"9"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12327540/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144776263","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhiqiang Xu, Anran Liu, Binbin Su, Minhui Wu, Bin Zhang, Guanyan Chen, Fan Lu, Liang Hu, Xinjie Mao
{"title":"Standardized Corneal Topography-Driven AI for Orthokeratology Fitting: A Hybrid Deep/Machine Learning Approach With Enhanced Generalizability.","authors":"Zhiqiang Xu, Anran Liu, Binbin Su, Minhui Wu, Bin Zhang, Guanyan Chen, Fan Lu, Liang Hu, Xinjie Mao","doi":"10.1167/tvst.14.8.16","DOIUrl":"10.1167/tvst.14.8.16","url":null,"abstract":"<p><strong>Purpose: </strong>The purpose of this study was to develop a standardized artificial intelligence (AI) system integrating corneal topography images and numerical parameters for optimizing orthokeratology (OK) lens fitting.</p><p><strong>Methods: </strong>Retrospective data from 1153 patients (2341 eyes) with Euclid OK lenses were analyzed. Five hundred nineteen eyes (393 patients) with treatment zone decentration ≤1 mm were included for model training. A device-agnostic corneal topography reconstruction pipeline generated standardized tangential curvature maps. A hybrid model combined deep learning (ResNet for image features) and machine learning (using numerical parameters) to predict alignment curve (AC) and cylinder power (CP), with numerical regression for AC and classification regression for CP. Multitask learning addressed AC-CP biomechanical coupling.</p><p><strong>Results: </strong>Numerical parameter-based models achieved optimal axial AC prediction (mean absolute error [MAE] = 0.290, R2 = 0.917), and CP prediction (accuracy [ACC] = 0.798, area under the curve [AUC] = 0.791). The image-based deep learning model using baseline corneal topography alone attained acceptable AC prediction (MAE = 0.248, R2 = 0.850), yet demonstrated suboptimal CP classification accuracy (ACC = 0.674, AUC = 0.621). Hybrid modeling achieved breakthrough performance in AC prediction (MAE = 0.136, R2 = 0.973) and superior CP classification (ACC = 0.898, AUC = 0.896).</p><p><strong>Conclusions: </strong>This system standardizes corneal topography across devices, addressing a critical barrier to generalizability in existing AI models, significantly enhancing fitting precision and generalizability for myopia control applications.</p><p><strong>Translational relevance: </strong>The device-agnostic design in the present study allows seamless integration into diverse clinical settings. The hybrid AI framework achieves near-expert accuracy, offering a scalable solution to access to high-quality OK lens fitting.</p>","PeriodicalId":23322,"journal":{"name":"Translational Vision Science & Technology","volume":"14 8","pages":"16"},"PeriodicalIF":2.6,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12366867/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144837834","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}