Ophthalmology science最新文献

{"title":"Implementing a Common Data Model in Ophthalmology: Mapping Structured Electronic Health Record Ophthalmic Examination Data to Standard Vocabularies","authors":"Justin C. Quon MD ,&nbsp;Christopher P. Long MD ,&nbsp;William Halfpenny MBBS, MEng ,&nbsp;Amy Chuang MS ,&nbsp;Cindy X. Cai MD, MS ,&nbsp;Sally L. Baxter MD, MSc ,&nbsp;Vamsi Daketi MS ,&nbsp;Amanda Schmitz BS ,&nbsp;Neil Bahroos MS ,&nbsp;Benjamin Y. Xu MD, PhD ,&nbsp;Brian C. Toy MD","doi":"10.1016/j.xops.2024.100666","DOIUrl":"10.1016/j.xops.2024.100666","url":null,"abstract":"<div><h3>Objective</h3><div>To identify and characterize concept coverage gaps of ophthalmology examination data elements within the Cerner Millennium electronic health record (EHR) implementations by the Observational Health Data Sciences and Informatics Observational Medical Outcomes Partnership (OMOP) common data model (CDM).</div></div><div><h3>Design</h3><div>Analysis of data elements in EHRs.</div></div><div><h3>Subjects</h3><div>Not applicable.</div></div><div><h3>Methods</h3><div>Source eye examination data elements from the default Cerner Model Experience EHR and a local implementation of the Cerner Millennium EHR were extracted, classified into one of 8 subject categories, and mapped to the semantically closest standard concept in the OMOP CDM. Mappings were categorized as exact, if the data element and OMOP concept represented equivalent information, wider, if the OMOP concept was missing conceptual granularity, narrower, if the OMOP concept introduced excess information, and unmatched, if no standard concept adequately represented the data element. Descriptive statistics and qualitative analysis were used to describe the concept coverage for each subject category.</div></div><div><h3>Main Outcome Measures</h3><div>Concept coverage gaps in 8 ophthalmology subject categories of data elements by the OMOP CDM.</div></div><div><h3>Results</h3><div>There were 409 and 947 ophthalmology data elements in the default and local Cerner modules, respectively. Of the 409 mappings in the default Cerner module, 25% (n = 102) were exact, 53% (n = 217) were wider, 3% (n = 11) were narrower, and 19% (n = 79) were unmatched. In the local Cerner module, 18% (n = 173) of mappings were exact, 54% (n = 514) were wider, 1% (n = 10) were narrower, and 26% (n = 250) were <em>unmatched</em>. The largest coverage gaps were seen in the local Cerner module under the visual acuity, sensorimotor testing, and refraction categories, with 95%, 95%, and 81% of data elements in each respective category having mappings that were not exact. Concept coverage gaps spanned all 8 categories in both EHR implementations.</div></div><div><h3>Conclusions</h3><div>Considerable coverage gaps by the OMOP CDM exist in all areas of the ophthalmology examination, which should be addressed to improve the OMOP CDM’s effectiveness in ophthalmic research. We identify specific subject categories that may benefit from increased granularity in the OMOP CDM and provide suggestions for facilitating consistency of standard concepts, with the goal of improving data standards in ophthalmology.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100666"},"PeriodicalIF":3.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11783105/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143082438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Topographic Relationships and Geographic Distribution of Prevascular Vitreous Fissures and Cisterns Assessed by Ultrawidefield En Face Vitreous Images","authors":"Fei Deng MD,&nbsp;Mengying Tao MD,&nbsp;Yanjie Zhu MD,&nbsp;Xiaoyu Xu MD,&nbsp;Yue Wu MD,&nbsp;Lisha Li,&nbsp;Ying Lin MD, PhD,&nbsp;Yan Luo MD, PhD","doi":"10.1016/j.xops.2024.100660","DOIUrl":"10.1016/j.xops.2024.100660","url":null,"abstract":"<div><h3>Purpose</h3><div>To determine the topographic relationships and geographic distribution of prevascular vitreous fissures (PVFs) and cisterns across the entire posterior vitreous membrane in healthy subjects, using ultrawidefield en face and cross-sectional swept-source OCT (SS-OCT) images.</div></div><div><h3>Design</h3><div>Observational cross-sectional study.</div></div><div><h3>Participants</h3><div>Ninety-six eyes of 96 healthy participants (age range, 20–49 years) without posterior vitreous detachment.</div></div><div><h3>Methods</h3><div>For each eye, a 29 × 24-mm SS-OCT volume scan was obtained, along with standardized horizontal and vertical scans through the fovea.</div></div><div><h3>Main Outcome Measures</h3><div>Ultrawidefield en face and cross-sectional images were analyzed to assess the topographic relationships and geographic distribution of PVFs and cisterns in the posterior vitreous.</div></div><div><h3>Results</h3><div>En face imaging readily distinguished various preretinal liquefaction spaces throughout the posterior vitreous, extending to near the equator. Aside from the posterior precortical vitreous pocket (PPVP) and the area of Martegiani, all preretinal liquefied fissures and cisterns were distributed along superficial retinal vessels, suggesting they originated from prevascular vitreous liquefaction. In 96 eyes of healthy young and middle-aged adults, PVFs were identified in all participants, presenting a continuous course. Cisterns were detected in 79 eyes (82.3%) and were distributed as follows: superotemporal (91.1%), infratemporal (63.3%), supranasal (41.8%), and inferonasal (22.8%), respectively. The superotemporal cistern was most frequently observed (<em>P</em> &lt; 0.001), and cisterns were more likely to involve multiple quadrants with age (<em>P</em> = 0.005). Additionally, all preretinal liquefaction spaces, including the PPVP, PVFs, and cisterns, were consistently located overlying the vitreoretinal tightly adhered regions.</div></div><div><h3>Conclusions</h3><div>Ultrawidefield en face vitreous imaging in healthy young and middle-aged adults revealed that (1) PVFs distributed along superficial retinal vessels with continuous course; (2) cisterns may develop from PVFs and are more common in the superotemporal quadrant; (3) cisterns appear early in life and become more widespread with age; (4) preretinal vitreous liquefaction follows a stereotypic pattern, aligning along regions of firm vitreoretinal adhesion.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100660"},"PeriodicalIF":3.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143168917","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated Quantification of Retinopathy of Prematurity Stage via Ultrawidefield OCT","authors":"Spencer S. Burt BA ,&nbsp;Aaron S. Coyner PhD ,&nbsp;Elizabeth V. Roti BS ,&nbsp;Yakub Bayhaqi PhD ,&nbsp;John Jackson MD ,&nbsp;Mani K. Woodward MS ,&nbsp;Shuibin Ni PhD ,&nbsp;Susan R. Ostmo MS ,&nbsp;Guangru Liang BS ,&nbsp;Yali Jia PhD ,&nbsp;David Huang MD ,&nbsp;Michael F. Chiang MD ,&nbsp;Benjamin K. Young MD ,&nbsp;Yifan Jian PhD ,&nbsp;John Peter Campbell MD","doi":"10.1016/j.xops.2024.100663","DOIUrl":"10.1016/j.xops.2024.100663","url":null,"abstract":"<div><h3>Purpose</h3><div>Retinopathy of prematurity (ROP) stage is defined by the visual appearance of the vascular-avascular border, which reflects a spectrum of pathologic neurovascular tissue (NVT). Previous work demonstrated that the thickness of the ridge lesion, measured using OCT, corresponds to higher clinical diagnosis of stage. This study evaluates whether the volume of anomalous NVT (ANVTV), defined as abnormal tissue protruding from the regular contour of the retina, can be measured automatically using deep learning to develop quantitative OCT-based biomarkers in ROP.</div></div><div><h3>Design</h3><div>Single-center retrospective case series.</div></div><div><h3>Participants</h3><div>Thirty-three infants with ROP in the Oregon Health &amp; Science University neonatal intensive care unit.</div></div><div><h3>Methods</h3><div>OCT B-scans were collected using an investigational ultrawidefield OCT. The ANVTV was manually segmented. A set of 3347 B-scans and corresponding manual segmentations from 12 volumes from 6 patients were used to train an automated segmentation tool using a U-Net. An additional held-out test data set of 60 B-scans from 6 infants was used to evaluate model performance. The Dice–Sorensen coefficient (DSC) comparing manual and automated segmentation of ANVTV was calculated. Scans from 21 additional infants were used for clinical evaluation of ANVTV using the visit in which they had developed their peak stage of ROP. Each infant had every B-scan in a volume automatically segmented for ANVTV (total number of segmented voxels within the 60° temporal to the optic disc). The ANVTV was compared between infants with stage 1 to 3 ROP using a Kruskal–Wallis test and tracked over time in all infants with stage 3 ROP.</div></div><div><h3>Main Outcome Measurements</h3><div>Cross sectional and longitudinal association between ANVTV and stages 1 to 3 ROP.</div></div><div><h3>Results</h3><div>Comparing automated and manual segmentation of ANVTV achieved a DSC of 0.61 ± 0.13. Using the U-Net, ANVTV was associated with higher disease stage both cross sectionally and longitudinally. Median ANVTV significantly increased as ROP stage worsened from 1 (0, [interquartile range: 0–0] kilovoxels) to 2 (170.1 [interquartile range: 104.2–183.6] kilovoxels) to 3 (421.4 [interquartile range: 312.3–1110.8] kilovoxels; <em>P</em> &lt; 0.001).</div></div><div><h3>Conclusions</h3><div>Automated OCT-based measurement of ANVTV was associated with clinical disease stage in ROP, both cross sectionally and longitudinally. Ultrawidefield-OCT may facilitate more objective screening, diagnosis, and monitoring in the future.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100663"},"PeriodicalIF":3.2,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11760822/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143048627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pivotal Trial Toward Effectiveness of Self-administered OCT in Neovascular Age-related Macular Degeneration. Report 2—Artificial Intelligence Analytics","authors":"Eric W. Schneider MD ,&nbsp;Jeffrey S. Heier MD ,&nbsp;Nancy M. Holekamp MD ,&nbsp;Miguel A. Busquets MD ,&nbsp;Alan L. Wagner MD ,&nbsp;S. Krishna Mukkamala MD ,&nbsp;Christopher D. Riemann MD ,&nbsp;Seong Y. Lee MD ,&nbsp;Brian C. Joondeph MD ,&nbsp;Steven S. Houston MD ,&nbsp;Kester Nahen PhD ,&nbsp;Nishant Mohan PhD ,&nbsp;Gidi Benyamini MBA","doi":"10.1016/j.xops.2024.100662","DOIUrl":"10.1016/j.xops.2024.100662","url":null,"abstract":"<div><h3>Purpose</h3><div>To validate the performance of the Notal OCT Analyzer (NOA) in processing self-administered OCT images from an OCT system designed for home use (home OCT [HOCT]) as part of a pivotal study aimed at achieving de novo United States Food and Drug Admininstration marketing authorization.</div></div><div><h3>Design</h3><div>A prospective quantitative cross-sectional artificial intelligence study.</div></div><div><h3>Participants</h3><div>The study enrolled adults aged ≥55 years diagnosed with neovascular age-related macular degeneration (nAMD) in ≥1 eligible eye with a best-corrected visual acuity of 20/320 or better.</div></div><div><h3>Methods</h3><div>Participants self-imaged 4 times on each of 2 HOCT devices and once with an in-office OCT (IO-OCT) device during a single visit. Scans were segmented by the NOA and human graders at a certified reading center (RC).</div></div><div><h3>Main Outcome Measures</h3><div>Intradevice and interdevice repeatability and reproducibility of total retinal hyporeflective (TRO) volume estimation by the NOA on HOCT-acquired images as compared with that of RC graders on IO-OCT-acquired images. Additionally, to assess the performance of the NOA in segmentation of TRO over multiple B-scans as compared with RC graders.</div></div><div><h3>Results</h3><div>Self-imaging was performed successfully by 387 participants in 2451 of 2616 attempts (93.7%). For repeatability, the coefficient of variance for NOA was 11.1% for eyes with &gt;10 volume units of TRO imaged with HOCT compared with 16.4% for RC graders segmenting IO-OCT images. The median DICE similarity coefficients for segmentation of TRO by NOA vs. Grader 1, Grader 2, and Grader 3; Grader 1 vs. Grader 2 and Grader 3; and Grader 2 vs. Grader 3 were 0.68, 0.68, 0.61, 0.72, 0.63, 0.70, respectively.</div></div><div><h3>Conclusions</h3><div>The performance of NOA supports the intended use of the system as a tool to monitor TRO at home between routine clinical visits in support of the management of nAMD.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100662"},"PeriodicalIF":3.2,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11731483/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142985681","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated Quantitative Assessment of Retinal Vascular Tortuosity in Patients with Sickle Cell Disease","authors":"Jimmy S. Chen MD ,&nbsp;Fritz Gerald P. Kalaw MD ,&nbsp;Eric D. Nudleman MD, PhD ,&nbsp;Nathan L. Scott MD, MPP","doi":"10.1016/j.xops.2024.100658","DOIUrl":"10.1016/j.xops.2024.100658","url":null,"abstract":"<div><h3>Objective</h3><div>To quantitatively assess the retinal vascular tortuosity of patients with sickle cell disease (SCD) and retinopathy (SCR) using an automated deep learning (DL)-based pipeline.</div></div><div><h3>Design</h3><div>Cross-sectional study.</div></div><div><h3>Subjects</h3><div>Patients diagnosed with SCD and screened for SCR at an academic eye center between January 2015 and November 2022 were identified using electronic health records. Eyes of unaffected matched patients (i.e., no history of SCD, hypertension, diabetes mellitus, or retinal occlusive disorder) served as controls.</div></div><div><h3>Methods</h3><div>For each patient, demographic data, sickle cell diagnosis, types and total number of sickle cell crises, SCD medications used, ocular and systemic comorbidities, and history of intraocular treatment were extracted. A previously published DL algorithm was used to calculate retinal microvascular tortuosity using ultrawidefield pseudocolor fundus imaging among patients with SCD vs. controls.</div></div><div><h3>Main Outcome Measures</h3><div>Cumulative tortuosity index (CTI).</div></div><div><h3>Results</h3><div>Overall, 64 patients (119 eyes) with SCD and 57 age- and race-matched controls (106 eyes) were included. The majority of the patients with SCD were females (65.6%) and of Black or African descent (78.1%), with an average age of 35.1 ± 20.1 years. The mean number of crises per patient was 3.4 ± 5.2, and the patients took 0.7 ± 0.9 medications. The mean CTI for eyes with SCD was higher than controls (1.06 ± vs. 1.03 ± 0.02, <em>P</em> &lt; 0.001). On subgroup analysis, hemoglobin S, hemoglobin C, and HbS/beta-thalassemia variants had significantly higher CTIs compared with controls (1.07 vs. 1.03, <em>P</em> &lt; 0.001), but not with sickle cell trait variant (1.04 vs. 1.03 control, <em>P</em> = .2). Univariable analysis showed a higher CTI in patients diagnosed with proliferative SCR, most significantly among those with sea-fan neovascularization (1.06 ± 0.02 vs. 1.04 ± 0.01, <em>P</em> &lt; 0.001) and those with &gt;3 sickle cell crises (1.07 ± 0.02 vs. 1.05 ± 0.02, <em>P</em> &lt; 0.001).</div></div><div><h3>Conclusions</h3><div>A DL-based metric of cumulative vascular tortuosity associates with and may be a potential biomarker for SCD and SCR disease severity.</div></div><div><h3>Financial Disclosures</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100658"},"PeriodicalIF":3.2,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11780102/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143069705","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Population Pharmacokinetics of Pegcetacoplan in Patients with Geographic Atrophy or Neovascular Age-related Macular Degeneration","authors":"Ryan L. Crass PharmD ,&nbsp;Komal Prem MD ,&nbsp;Francois Gauderault PhD ,&nbsp;Ramiro Ribeiro MD, PhD ,&nbsp;Caroline R. Baumal MD ,&nbsp;Brandon Smith PhD ,&nbsp;Daniel Epling PhD ,&nbsp;Sunny Chapel PhD","doi":"10.1016/j.xops.2024.100657","DOIUrl":"10.1016/j.xops.2024.100657","url":null,"abstract":"<div><h3>Objective</h3><div>To develop a population pharmacokinetic (PK) model to characterize serum pegcetacoplan concentration-time data after intravitreal administration in patients with geographic atrophy (GA) or neovascular age-related macular degeneration (nAMD).</div></div><div><h3>Design</h3><div>Pharmacokinetic modeling.</div></div><div><h3>Participants</h3><div>Two hundred sixty-one patients with GA or nAMD enrolled in 4 clinical studies of pegcetacoplan.</div></div><div><h3>Methods</h3><div>Serum concentration data were pooled from 4 clinical studies. Pegcetacoplan dosing included single intravitreal injections of 4, 10, and 20 mg and multiple intravitreal injections of 15 mg monthly or every other month. Considering a high proportion of samples were below the limit of quantification (BLQ) in serum following intravitreal administration, the M3 method of likelihood-based handling of data BLQ was employed in NONMEM (version 7.4). Covariate model development was performed using stepwise forward (α = 0.05) and backward (α = 0.001) selection. Predicted PK parameters and exposure metrics were generated via simulation in serum and vitreous humor.</div></div><div><h3>Main Outcome Measures</h3><div>Pharmacokinetic parameters.</div></div><div><h3>Results</h3><div>Intravitreal pegcetacoplan displayed absorption-limited (i.e., “flip-flop”) kinetics with median empirical Bayes estimated pegcetacoplan absorption and elimination half-lives of 13.1 days and 4.51 days, respectively. Vitreous exposure was predicted to be &gt;1300-fold higher than serum exposure, with maximum concentrations in serum below the threshold required to elicit systemic pharmacodynamic effects. Drug accumulation from first dose to steady state was predicted to be minimal in serum (mean accumulation ratio = 1.50 with monthly dosing, 1.10 with every-other-month dosing) and vitreous humor (mean accumulation ratio = 1.30 with monthly dosing, 1.10 with every-other-month dosing). Age, sex, and baseline C3 level were identified as significant (<em>P</em> &lt; 0.001) predictors of apparent serum pegcetacoplan clearance after intravitreal administration; however, none of the covariate effects appeared to be clinically meaningful given the low absolute maximum serum concentrations achieved (&lt;5 μg/mL). Concomitant anti-VEGF treatment did not significantly influence vitreous disposition of pegcetacoplan as assessed in a dedicated post hoc covariate model.</div></div><div><h3>Conclusions</h3><div>This population PK model adequately described the serum concentration-time profile of pegcetacoplan after intravitreal administration in adults with GA or nAMD.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100657"},"PeriodicalIF":3.2,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11754507/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143030394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep-Reticular Pseudodrusen-Net: A 3-Dimensional Deep Network for Detection of Reticular Pseudodrusen on OCT Scans","authors":"Amr Elsawy PhD ,&nbsp;Tiarnan D.L. Keenan PhD, MD ,&nbsp;Alisa T. Thavikulwat MD ,&nbsp;Amy Lu MD ,&nbsp;Sunil Bellur MD ,&nbsp;Souvick Mukherjee PhD ,&nbsp;Elvira Agron MS ,&nbsp;Qingyu Chen PhD ,&nbsp;Emily Y. Chew MD ,&nbsp;Zhiyong Lu PhD","doi":"10.1016/j.xops.2024.100655","DOIUrl":"10.1016/j.xops.2024.100655","url":null,"abstract":"<div><h3>Objective</h3><div>To propose Deep-RPD-Net, a 3-dimensional deep learning network with semisupervised learning (SSL) for the detection of reticular pseudodrusen (RPD) on spectral-domain OCT scans, explain its decision-making, and compare it with baseline methods.</div></div><div><h3>Design</h3><div>Deep learning model development.</div></div><div><h3>Participants</h3><div>Three hundred fifteen participants from the Age-Related Eye Disease Study 2 Ancillary OCT Study (AREDS2) and 161 participants from the Dark Adaptation in Age-related Macular Degeneration Study (DAAMD).</div></div><div><h3>Methods</h3><div>Two datasets comprising of 1304 (826 labeled) and 1479 (1366 labeled) OCT scans were used to develop and evaluate Deep-RPD-Net and baseline models. The AREDS2 RPD labels were transferred from fundus autofluorescence images, which were captured at the same visit for OCT scans, and DAAMD RPD labels were obtained from the Wisconsin reading center. The datasets were divided into 70%, 10%, and 20% at the participant level for training, validation, and test sets, respectively, for the baseline model. Then, SSL was used with the unlabeled OCT scans to improve the trained model. The performance of Deep-RPD-Net was compared to that of 3 retina specialists on a subset of 50 OCT scans for each dataset. En face and B-scan heatmaps of all networks were visualized and graded on 25 OCT scans with positive labels, using a scale of 1 to 4, to explore the models' decision-making.</div></div><div><h3>Main Outcome Measures</h3><div>Accuracy and area under the receiver-operating characteristic curve (AUROC).</div></div><div><h3>Results</h3><div>Deep-RPD-Net achieved the highest performance metrics, with accuracy and AUROC of 0.81 (95% confidence interval [CI]: 0.76–0.87) and 0.91 (95% CI: 0.86–0.95) on the AREDS2 dataset and 0.80 (95% CI: 0.75–0.84) and 0.86 (95% CI: 0.79–0.91) on the DAAMD dataset. On the subjective test, it achieved accuracy of 0.84 compared with 0.76 for the most accurate retina specialist on the AREDS2 dataset and 0.82 compared with 0.84 on the DAAMD dataset. It also achieved the highest visualization grades, of 3.26 and 3.32 for en face and B-scan heatmaps, respectively.</div></div><div><h3>Conclusions</h3><div>Deep-RPD-Net was able to detect RPD accurately from OCT scans. The visualizations of Deep-RPD-Net were the most explainable to the retina specialist with the highest accuracy. The code and pretrained models are publicly available at https://github.com/ncbi-nlp/Deep-RPD-Net.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100655"},"PeriodicalIF":3.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11758204/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143048690","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Short-term Assessment of Glaucoma Progression in Clinical Trials Using Trend-based Visual Field Progression Analysis","authors":"Ryo Asaoka MD, PhD ,&nbsp;Makoto Nakamura MD, PhD ,&nbsp;Masaki Tanito MD, PhD ,&nbsp;Yuri Fujino CO ,&nbsp;Akira Obana MD, PhD ,&nbsp;Shiro Mizoue MD, PhD ,&nbsp;Kazuhiko Mori MD, PhD ,&nbsp;Katsuyoshi Suzuki MD, PhD ,&nbsp;Takehiro Yamashita MD, PhD ,&nbsp;Kazunori Hirasawa CO, PhD ,&nbsp;Nobuyuki Shoji MD, PhD ,&nbsp;Hiroshi Murata MD","doi":"10.1016/j.xops.2024.100656","DOIUrl":"10.1016/j.xops.2024.100656","url":null,"abstract":"<div><h3>Objective</h3><div>To evaluate the effect of disease stage, frequency and clustering of visual field (VF) tests, inclusion of 1 or both eyes, and 1 (1 arm; before and after a treatment) or 2 groups (2 arms; treatment and control arm) on sample size calculation in clinical trials.</div></div><div><h3>Design</h3><div>Clinical cohort study.</div></div><div><h3>Participants</h3><div>A series of VFs were simulated based on test-retest VF data in the early, moderate, and advanced stages of glaucoma with 231, 204, and 226 eyes, respectively.</div></div><div><h3>Methods</h3><div>The mean of mean deviation (MD) slope was −0.75 decibels (dB)/year before treatment initiation in the 1-arm trial, and in the control group in the 2-arm trial. Visual field measurements were scheduled as 8 times in 2 years.</div></div><div><h3>Main Outcome Measures</h3><div>Sample size calculation in clinical trials.</div></div><div><h3>Results</h3><div>In the 1-arm trial, when only 1 eye was used in each patient, the 80% probability of significance in the moderate stage was observed with sample size = 70 eyes. Disease in the early stage and inclusion of both eyes decreased this number to 30 eyes; these decreasing effects were significantly larger than performing 1 or 2 additional VFs at the beginning and end of the observation. Conversely, a greater number of eyes was necessary in advanced stage than in moderate stage. In the 2-arm trial (80% probability of significance, and 1 eye per patient), the 80% probability of significance was observed with sample size = 80 eyes in each arm, a tendency that was similar to what observed for the 1-arm trial. Similar tendency was observed in the simulations with much slower VF progression (mean MD slope = −0.25 dB/year).</div></div><div><h3>Conclusions</h3><div>The present study highlights the importance of considering disease stage when planning a clinical trial.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100656"},"PeriodicalIF":3.2,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11780079/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143069787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated Detection of Filamentous Fungal Keratitis on Whole Slide Images of Potassium Hydroxide Smears with Multiple Instance Learning","authors":"Jad F. Assaf MD ,&nbsp;Hady Yazbeck MD ,&nbsp;Prajna N. Venkatesh DO, DNB, FRCOpth ,&nbsp;Lalitha Prajna MD, DNB ,&nbsp;Rameshkumar Gunasekaran MSc ,&nbsp;Karpagam Rajarathinam CMLT ,&nbsp;Thomas M. Lietman MD ,&nbsp;Jeremy D. Keenan MD, MPH ,&nbsp;J. Peter Campbell MD, MPH ,&nbsp;Xubo Song PhD ,&nbsp;Travis K. Redd MD, MPH","doi":"10.1016/j.xops.2024.100653","DOIUrl":"10.1016/j.xops.2024.100653","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;Purpose&lt;/h3&gt;&lt;div&gt;The diagnosis of fungal keratitis using potassium hydroxide (KOH) smears of corneal scrapings enables initiation of the correct antimicrobial therapy at the point-of-care but requires time-consuming manual examination and expertise. This study evaluates the efficacy of a deep learning framework, dual stream multiple instance learning (DSMIL), in automating the analysis of whole slide imaging (WSI) of KOH smears for rapid and accurate detection of fungal infections.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Design&lt;/h3&gt;&lt;div&gt;Retrospective observational study.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Participants&lt;/h3&gt;&lt;div&gt;Corneal scrapings from 568 patients with suspected fungal keratitis; 51% contained filamentous fungi according to human expert interpretation.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Methods&lt;/h3&gt;&lt;div&gt;Dual stream multiple instance learning was employed to analyze WSI of KOH smears. Due to the extensive size of these images, often exceeding 100 000 pixels, conventional computer vision methods (e.g., convolutional neural networks) are not feasible. Dual stream multiple instance learning segments the WSI into patches for analysis, extracting relevant features from each patch and aggregating these to make a comprehensive slide-level diagnosis while generating heat maps to visualize areas contributing most to the prediction. Fivefold cross-validation was used for training and validation, with a hold-out test set comprising 15% of the total samples.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Main Outcome Measures&lt;/h3&gt;&lt;div&gt;Accuracy, sensitivity, specificity, area under the receiver operating characteristic curve (AUC), F1 score, positive predictive value (PPV), and negative predictive value (NPV) in distinguishing fungal from nonfungal slides.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Results&lt;/h3&gt;&lt;div&gt;Dual stream multiple instance learning demonstrated an overall AUC of 0.88 with an accuracy of 79% and an F1 score of 0.79 in distinguishing fungal from nonfungal slides, with sensitivity of 85%, specificity of 71%, PPV of 80%, and NPV of 79%. For “consensus cases,” where 2 human graders agreed on the slide interpretation, the model achieved an accuracy of 85% and an F1 score of 0.85. For “discrepant cases,” the accuracy was 71% with an F1 score of 0.71. The generated heatmaps highlighted regions corresponding to fungal elements. Code and models are open-sourced and available at &lt;span&gt;&lt;span&gt;https://github.com/Redd-Cornea-AI/KOH-Smear-DSMIL&lt;/span&gt;&lt;svg&gt;&lt;path&gt;&lt;/path&gt;&lt;/svg&gt;&lt;/span&gt;.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Conclusions&lt;/h3&gt;&lt;div&gt;The DSMIL framework shows significant promise in automating interpretation of KOH smears. Its capability to handle large, high-resolution WSI data and accurately detect fungal infections, while providing visual explanations through heatmaps, could enhance the scalability of KOH smear interpretation, ultimately reducing the global burden of blindness from infectious keratitis.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Financial Disclosure(s)&lt;/h3&gt;&lt;div&gt;Proprietary or commercial disclosure may be found in the Footnotes and Disclosures ","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100653"},"PeriodicalIF":3.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11731208/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142985664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantification of Fundus Autofluorescence Features in a Molecularly Characterized Cohort of >3500 Patients with Inherited Retinal Disease from the United Kingdom","authors":"William A. Woof PhD ,&nbsp;Thales A.C. de Guimarães MD, PhD ,&nbsp;Saoud Al-Khuzaei MBChB, DPhil ,&nbsp;Malena Daich Varela MD, PhD ,&nbsp;Sagnik Sen MD, MRCSEd ,&nbsp;Pallavi Bagga PhD ,&nbsp;Bernardo Mendes BSci ,&nbsp;Mital Shah DPhil, FRCOphth ,&nbsp;Paula Burke ,&nbsp;David Parry ,&nbsp;Siying Lin MBBS, PhD ,&nbsp;Gunjan Naik PhD ,&nbsp;Biraja Ghoshal PhD ,&nbsp;Bart J. Liefers PhD ,&nbsp;Dun Jack Fu MD, PhD ,&nbsp;Michalis Georgiou MD, PhD ,&nbsp;Quang Nguyen MRes ,&nbsp;Alan Sousa da Silva PhD ,&nbsp;Yichen Liu MSci ,&nbsp;Yu Fujinami-Yokokawa PhD ,&nbsp;Nikolas Pontikos PhD","doi":"10.1016/j.xops.2024.100652","DOIUrl":"10.1016/j.xops.2024.100652","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;Purpose&lt;/h3&gt;&lt;div&gt;To quantify relevant fundus autofluorescence (FAF) features cross-sectionally and longitudinally in a large cohort of patients with inherited retinal diseases (IRDs).&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Design&lt;/h3&gt;&lt;div&gt;Retrospective study of imaging data.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Participants&lt;/h3&gt;&lt;div&gt;Patients with a clinical and molecularly confirmed diagnosis of IRD who have undergone 55° FAF imaging at Moorfields Eye Hospital (MEH) and the Royal Liverpool Hospital between 2004 and 2019.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Methods&lt;/h3&gt;&lt;div&gt;Five FAF features of interest were defined: vessels, optic disc, perimacular ring of increased signal (ring), relative hypo-autofluorescence (hypo-AF), and hyper-autofluorescence (hyper-AF). Features were manually annotated by 6 graders in a subset of patients based on a defined grading protocol to produce segmentation masks to train an artificial intelligence model, AIRDetect, which was then applied to the entire imaging data set.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Main Outcome Measures&lt;/h3&gt;&lt;div&gt;Quantitative FAF features, including area and vessel metrics, were analyzed cross-sectionally by gene and age, and longitudinally. AIRDetect feature segmentation and detection were validated with Dice score and precision/recall, respectively.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Results&lt;/h3&gt;&lt;div&gt;A total of 45 749 FAF images from 3606 patients with IRD from MEH covering 170 genes were automatically segmented using AIRDetect. Model-grader Dice scores for the disc, hypo-AF, hyper-AF, ring, and vessels were, respectively, 0.86, 0.72, 0.69, 0.68, and 0.65. Across patients at presentation, the 5 genes with the largest hypo-AF areas were &lt;em&gt;CHM&lt;/em&gt;, &lt;em&gt;ABCC6&lt;/em&gt;, &lt;em&gt;RDH12&lt;/em&gt;, &lt;em&gt;ABCA4&lt;/em&gt;, and &lt;em&gt;RPE65&lt;/em&gt;, with mean per-patient areas of 43.72, 29.57, 20.07, 19.65, and 16.92 mm&lt;sup&gt;2&lt;/sup&gt;, respectively. The 5 genes with the largest hyper-AF areas were &lt;em&gt;BEST1&lt;/em&gt;, &lt;em&gt;CDH23&lt;/em&gt;, &lt;em&gt;NR2E3&lt;/em&gt;, &lt;em&gt;MYO7A&lt;/em&gt;, and &lt;em&gt;RDH12&lt;/em&gt;, with mean areas of 0.50, 047, 0.44, 0.38, and 0.33 mm&lt;sup&gt;2&lt;/sup&gt;, respectively. The 5 genes with the largest ring areas were &lt;em&gt;NR2E3, CDH23&lt;/em&gt;, &lt;em&gt;CRX&lt;/em&gt;, &lt;em&gt;EYS&lt;/em&gt;, and &lt;em&gt;PDE6B&lt;/em&gt;, with mean areas of 3.60, 2.90, 2.89, 2.56, and 2.20 mm&lt;sup&gt;2&lt;/sup&gt;, respectively. Vessel density was found to be highest in &lt;em&gt;EFEMP1&lt;/em&gt;, &lt;em&gt;BEST1&lt;/em&gt;, &lt;em&gt;TIMP3&lt;/em&gt;, &lt;em&gt;RS1&lt;/em&gt;, and &lt;em&gt;PRPH2&lt;/em&gt; (11.0%, 10.4%, 10.1%, 10.1%, 9.2%) and was lower in retinitis pigmentosa (RP) and Leber congenital amaurosis genes. Longitudinal analysis of decreasing ring area in 4 RP genes (&lt;em&gt;RPGR&lt;/em&gt;, &lt;em&gt;USH2A&lt;/em&gt;, &lt;em&gt;RHO&lt;/em&gt;, and &lt;em&gt;EYS&lt;/em&gt;) found &lt;em&gt;EYS&lt;/em&gt; to be the fastest progressor at −0.178 mm&lt;sup&gt;2&lt;/sup&gt;/year.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Conclusions&lt;/h3&gt;&lt;div&gt;We have conducted the first large-scale cross-sectional and longitudinal quantitative analysis of FAF features across a diverse range of IRDs using a novel AI approach.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Financial Disclosure(s)&lt;/h3&gt;&lt;div&gt;Proprietary or commercial disclosure may be found in th","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100652"},"PeriodicalIF":3.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11782848/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143082444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}