Search

Current glaucoma treatments aim to lower intraocular pressure, often with topical ocular hypotensive medications. Unfortunately, the effectiveness of these medications depends on sustained patient adherence to regimens which may involve instilling multiple medications several times daily. Patient adherence to glaucoma medications is often low. Recent innovations in digital sensor technologies have been leveraged to confirm eyedrop medication usage in real-time and relay this information back to providers. Some sensors have also been designed to deliver medication reminders and notifications as well as assist with correct eyedrop administration technique. Here, we review recent innovations targeted at improving glaucoma medication adherence and discuss their limitations.

AimTo compare the repeatability of peripapillary perfusion density and flux index measurements on referenced and non-referenced optical microangiography (OMAG) scans in normal, glaucoma suspect and glaucoma eyes.MethodsIn a cross-sectional study, 48 eyes (33 subjects) underwent three repeat, non-referenced peripapillary OMAG scans in the same session and 43 eyes (25 subjects) underwent three referenced peripapillary OMAG scans. In the referenced scan group, repeat scans (second and the third scan) were acquired exactly on the baseline (first) scan using the 'track to prior scan' option on the device. Repeatability estimates of the mean and four-sector (temporal, superior, nasal and inferior) OMAG measurements on the non-referenced and referenced scans were assessed using within-subject coefficient of repeatability (CRw) and variation (CVw).ResultsCRw (%) of peripapillary perfusion density measurements (range: 2.0-4.1) on non-referenced scans were significantly higher than that on referenced scans (range: 1.4-2.7). CVw (%) on non-referenced and referenced scans ranged from 1.7 to 3.1 and from 1.2 to 2.1, respectively . CRw of flux index on non-referenced and referenced scans ranged from 4.4 to 5.8 and from 3.6 to 4.8, respectively. CVw on non-referenced and referenced scans ranged from 4.1 to 5.2 and from 3.3 to 4.5, respectively.ConclusionsRepeatability estimates of OMAG measurements were better on referenced scans compared with non-referenced scans. Perfusion density measurements had lower variability than flux index. OCTA-measured perfusion density of referenced scans is preferable for monitoring vascular change in glaucoma.

AimsWe compared feasibility, quality and outcomes of visual field (VF) testing in children with neuro-ophthalmic disease between the discontinued 'gold-standard' Goldmann and Octopus perimeters.MethodsChildren with neuro-ophthalmic disease, attending Great Ormond Street Hospital, London, were assessed using standardised protocols by one examiner in a single sitting, using Goldmann and Octopus kinetic perimetry. Outputs were classified to compare severity of loss and defect type. Test quality was assessed using both qualitative and quantitative methods.ResultsThirty children (40% female) aged 5-15 years participated. Goldmann perimetry was completed in full by 90.0% vs 72.4% for Octopus. Inability to plot the blind spot was the most common reason for not completing testing. Over 75% completed a test in ≤20 min. Duration was similar between perimeters (paired t-test, mean difference: 0.48min (-1.2, 2.2), p=0.559). The lowest quality tests were for Octopus perimetry in children

Background/aims: To determine if patients with vitreoretinal lymphoma (VRL) and concomitant central nervous system lymphoma (CNSL) may present without brain MRI findings, but possess cerebrospinal fluid (CSF) suspicious for lymphoma., Methods: This was a retrospective, single-centre, observational study evaluating patients with a diagnosis or suspicion of VRL seen at Memorial Sloan Kettering Cancer Center between 2006 and 2024. Patients were included if the final diagnosis was biopsy-proven CNSL and both MRI brain with and without contrast±CSF evaluation (obligatory for inclusion if MRI negative) were performed at the initial diagnostic workup. Patients were excluded if CNS disease treatment (brain, spine or CSF) preceded ocular disease. Patients with prior extra-CNS disease were included. Clinical records and radiographic imaging were retrospectively reviewed and relevant data were recorded for each patient. We evaluated the proportion of patients with MRI negative and CSF suspicious for lymphoma. Subgroup analysis included imaging features, pathology, treatment and disease course., Results: We identified 65 patients. Of the 65 patients at the presentation of VRL, 30 had negative MRI brain and CSF, 16 had positive brain MRI and negative CSF and 8 had both positive MRI brain and CSF. 11 (16.9%) had CSF suspicious for lymphoma without positive findings on MRI of the brain. In this subgroup, the median age was 66 years (range 49-82) and 36% were female. 86% of these patients were asymptomatic neurologically. 73% underwent systemic treatment. At a mean 3 years follow-up, 91% of patients were living., Conclusion: In patients with suspected VRL, it is possible to have CSF test positive for lymphoma in the context of negative brain MRI. This suggests, when evaluating VRL patients for concomitant CNS disease, CSF evaluation leads to earlier detection and systemic treatment, even when MRI brain findings are negative. In our cohort, an absence of CSF evaluation in the context of negative brain MRI could have missed 16.9% of patients with CNS lymphoma., Competing Interests: Competing interests: RFS: none, DHA: none, JC: none, CG: Consultation for: Roche, BTG, Curis, Ono, JHF: none., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Background/aims: We evaluated longitudinal autoantibody changes after intravenous methylprednisolone (IVMP), compared them with those in untreated patients and identified prognostic factors for treatment response., Methods: In this single-centre, retrospective, observational study, a total of 163 individuals diagnosed with moderate-to-severe thyroid eye disease were enrolled and followed for 12 months. Depending on whether IVMP was administered, we divided the patients into treatment and control groups. Based on the effect of IVMP on TSH receptor (TSH Rc) antibody level, we divided the patients into Ab declined and Ab not declined groups.We evaluated the time, group and interaction associations with the longitudinal autoantibody titres over 12 months using generalised estimating equations. Using multivariable logistic regression, we investigated the prognostic factors for a poor response to IVMP., Results: In the IVMP group, the TSH Rc antibody (Ab) titre decreased rapidly for 6 months and then decreased slowly until 12 months, becoming similar to the control group at 12 months. This suggests a difference in the decreasing pattern over time between the IVMP and control groups (group and time interaction p=0.029). Total cholesterol (OR 1.0217 (95% CI 1.0068 to 1.0370), p=0.0043) was a significant prognostic factor for the steroid response. The threshold total cholesterol value to distinguish between Ab declined and Ab not declined was 186 mg/dL., Conclusion: IVMP significantly decreased the TSH Rc Ab level for the 3 months after treatment, compared with the no-treatment group, but the groups did not differ significantly after 12 months. Patients with high total cholesterol levels generally showed a poor response to IVMP., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Objective: To determine the agreement between measurements of accommodative amplitude (AoA) in children using a specialised accommodative rule and measurments without it., Methods: A total of 502 children underwent optometric examinations, including the measurement of visual acuity, objective and subjective refraction. AoA measurements were done with and without the Berens accommodative rule. The measurements of AoA were conducted monocularly using a -4 D lens. A fixation stick containing English letters equivalent to 20/30 visual acuity and a long millimetre ruler was used to measure AoA without the accommodative rule. This measurement was performed by the two trained examiners. The agreement between these methods was reported by 95% limits of agreement (LoA) and interclass correlation coefficient (ICC)., Results: The mean age of the participants was 11.7±1.3 years (range: 9-15 years) and 52.4% were male. The mean AoA with and without the accommodative rule was 20.02±6.02 D and 22.46±6.32 D, respectively. The 95% LoA between the two methods was -12.5 to 7.5 D, and the ICC was 0.67 (95% CI 0.63 to 0.70). The 95% LoA was narrower in higher age groups and males compared with females (18.92 vs 20.87). The 95% LoA was narrower in hyperopes (16.83 D) compared with emmetropes (18.37 D) and myopes (18.27 D). The agreement was not constant and decreased in higher values of AoA., Conclusion: There is a poor and non-constant agreement between the measurements of the AoA with and without the accommodative rule. The mean AoA was 2.5 D lower with using the accommodative rule., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Purpose: To develop and validate a deep learning model for the segmentation of five retinal biomarkers associated with neovascular age-related macular degeneration (nAMD)., Methods: 300 optical coherence tomography volumes from subject eyes with nAMD were collected. Images were manually segmented for the presence of five crucial nAMD features: intraretinal fluid, subretinal fluid, subretinal hyperreflective material, drusen/drusenoid pigment epithelium detachment (PED) and neovascular PED. A deep learning architecture based on a U-Net was trained to perform automatic segmentation of these retinal biomarkers and evaluated on the sequestered data. The main outcome measures were receiver operating characteristic curves for detection, summarised using the area under the curves (AUCs) both on a per slice and per volume basis, correlation score, enface topography overlap (reported as two-dimensional (2D) correlation score) and Dice coefficients., Results: The model obtained a mean (±SD) AUC of 0.93 (±0.04) per slice and 0.88 (±0.07) per volume for fluid detection. The correlation score (R 2 ) between automatic and manual segmentation obtained by the model resulted in a mean (±SD) of 0.89 (±0.05). The mean (±SD) 2D correlation score was 0.69 (±0.04). The mean (±SD) Dice score resulted in 0.61 (±0.10)., Conclusions: We present a fully automated segmentation model for five features related to nAMD that performs at the level of experienced graders. The application of this model will open opportunities for the study of morphological changes and treatment efficacy in real-world settings. Furthermore, it can facilitate structured reporting in the clinic and reduce subjectivity in clinicians' assessments., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

The rapid advancements in generative artificial intelligence are set to significantly influence the medical sector, particularly ophthalmology. Generative adversarial networks and diffusion models enable the creation of synthetic images, aiding the development of deep learning models tailored for specific imaging tasks. Additionally, the advent of multimodal foundational models, capable of generating images, text and videos, presents a broad spectrum of applications within ophthalmology. These range from enhancing diagnostic accuracy to improving patient education and training healthcare professionals. Despite the promising potential, this area of technology is still in its infancy, and there are several challenges to be addressed, including data bias, safety concerns and the practical implementation of these technologies in clinical settings., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Background: Artificial intelligence (AI) in medical imaging diagnostics has huge potential, but human judgement is still indispensable. We propose an AI-aided teaching method that leverages generative AI to train students on many images while preserving patient privacy., Methods: A web-based course was designed using 600 synthetic ultra-widefield (UWF) retinal images to teach students to detect disease in these images. The images were generated by stable diffusion, a large generative foundation model, which we fine-tuned with 6285 real UWF images from six categories: five retinal diseases (age-related macular degeneration, glaucoma, diabetic retinopathy, retinal detachment and retinal vein occlusion) and normal. 161 trainee orthoptists took the course. They were evaluated with two tests: one consisting of UWF images and another of standard field (SF) images, which the students had not encountered in the course. Both tests contained 120 real patient images, 20 per category. The students took both tests once before and after training, with a cool-off period in between., Results: On average, students completed the course in 53 min, significantly improving their diagnostic accuracy. For UWF images, student accuracy increased from 43.6% to 74.1% (p<0.0001 by paired t-test), nearly matching the previously published state-of-the-art AI model's accuracy of 73.3%. For SF images, student accuracy rose from 42.7% to 68.7% (p<0.0001), surpassing the state-of-the-art AI model's 40%., Conclusion: Synthetic images can be used effectively in medical education. We also found that humans are more robust to novel situations than AI models, thus showcasing human judgement's essential role in medical diagnosis., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Objectives: Train an automatic retinal image analysis (ARIA) method to screen glaucomatous optic neuropathy (GON) on non-mydriatic retinal images labelled with the additional results of optical coherence tomography (OCT) and assess different models for the GON classification., Methods: All the images were obtained from the hospital for training and 10-fold cross-validation. Two methods were used to improve the classification performance: (1) using images labelled with the additional results of OCT as the reference standard and (2) generating models using retinal features from the entire images, the region of interest (ROI) of the optic disc, and the ROI of the macula, and the combination of all the features., Results: Overall, we collected 1338 images with paired OCT scans. In 10-fold validation, ARIA achieved sensitivities of 92.2 %, 92.7% and 85.7%, specificities of 88.8%, 86.7% and 80.2% and accuracies of 90.6%, 89.9% and 83.1% using the retinal features from the entire images, the ROI of the optic disc and the ROI of the macula, respectively. We found the model combining all the features has the best classification performance and obtained a sensitivity of 92.5%, a specificity of 92.1% and an accuracy of 92.4%, which is significantly different from other models (p<0.001)., Conclusion: We used two methods to improve the classification performance and found the best model to detect glaucoma on colour fundus retinal images. It can become a cost-effective and relatively more accurate glaucoma screening tool than conventional methods., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Aims: To explore the sensitive components of full-field electroretinography (ERG) as indicators of retina function at the onset of acute ischaemic central retinal vein occlusion (CRVO)., Methods: 11 patients (11 eyes) with ischaemic CRVO and 32 patients (32 eyes) with non-ischaemic CRVO who presented with first-episode unilateral CRVO within 1 month of symptom onset and with no previous intervention were examined by the International Society for Clinical Electrophysiology of Vision standard ERG., Results: A significant amplitude decline and peak time delay in light-adapted (LA) 3 ERG and LA 30 Hz flicker ERG (p<0.05 for all) was found in the ischaemic CRVO eyes, compared with the non-ischaemic CRVO eyes. The b/a amplitude ratio of dark-adapted (DA) 3 ERG, DA 10 ERG and LA 3 ERG was significantly different between the ischaemic and non-ischaemic groups (p<0.05 for all). Regarding oscillatory potentials (OPs), the amplitudes of OP1, OP2 and OP3 as well as the sum of DA 3 OP1-4 amplitudes (∑OPs) showed significant changes (p<0.01 for all) between two groups. No peak time delay of OPs was found between the ischaemic and non-ischaemic CRVO eyes., Conclusion: The amplitude of DA 0.01 ERG, components of LA 3 ERG and LA 30 Hz flicker ERG, and the b/a amplitude ratio could be among the most sensitive indicators in patients with acute ischaemic CRVO. The amplitudes of OP1, OP2, OP3 and ∑OPs in the CRVO eyes were reduced to 40% of the control values, showing that this quantitative method is reliable for detecting ischaemic retinal diseases, even in early stage., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Purpose: To compare the diagnostic utility of metagenomic deep sequencing (MDS) to cytology, flow cytometry and gene rearrangement by PCR in ocular samples of patients with suspected vitreoretinal lymphoma (VRL)., Methods: Patients with suspected VRL underwent ocular sampling of one or both eyes at the Emory Eye Center from September 2017 to June 2022. Ocular samples were evaluated with MDS and conventional diagnostics. MDS was performed at the Ralph and Sophie Heintz Laboratory at the F.I. Proctor Foundation. Relevant demographic and clinical data were retrospectively collected from medical records. Patients were diagnosed with VRL based on clinical assessment and conventional diagnostic testing., Results: This study included 13 patients with suspected VRL who underwent diagnostic vitrectomy, including 1 patient who had an additional subretinal biopsy. Six patients (46.2%) were diagnosed with VRL. Among patients diagnosed with VRL, MDS detected pathogenic mutations in 5 out of 6 patients (83.3%) while cytology was positive for VRL in 4 out of 6 patients (66.7%), flow cytometry in 4 out of 4 patients (100.0%) and PCR in 4 out of 4 patients (100.0%). MDS detected mutations in MYD88 in 2 out of 6 patients diagnosed with VRL. In 7 patients (53.8%) not diagnosed with VRL, MDS detected pathogenic lymphoma mutations in 2 patients (28.6%)., Discussion: MDS detected pathogenic mutations in five out of six patients diagnosed with VRL, including in two patients with negative cytology, demonstrating its potential to improve diagnostic rates of VRL as an adjunctive test., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Background/aims: Female choroideremia carriers present with a spectrum of disease severity. Unlike in men, the rate of disease progression has not been well characterised in carriers. This longitudinal study aimed to determine the rate of retinal degeneration in choroideremia carriers, using multimodal imaging and microperimetry., Methods: Choroideremia carriers previously seen at Oxford Eye Hospital (United Kingdom) between 2012 and 2017 returned for testing between 2015 and 2023, providing up to 11 years' follow-up data. Participants had optical coherence tomography, fundus-tracked microperimetry and fundus autofluorescence (FAF) imaging performed., Results: Thirty-four eyes of 17 choroideremia carriers were examined using multimodal imaging. Median age was 44 (range: 15-73) years at baseline and median follow-up duration was 7 (range: 1-11) years. At baseline, phenotype was classified as fine (n=5 eyes), coarse (n=13 eyes), geographic (n=12 eyes) or male pattern (n=4 eyes). Thirteen patients showed no change in phenotype classification, four showed slight changes associated with choroideremia-related retinal degeneration. Despite this, carriers with severe retinal phenotypes had a statistically significant decline in average retinal sensitivity (-0.7 dB and -0.8 dB per year, respectively, p<0.001), area of geographic loss defined by FAF (+2.5 mm 2 and +3.7 mm 2 per year, respectively, p<0.001) and thinning of the photoreceptor complex (up to -2.8 microns and -10.3 microns per year, p<0.001)., Conclusion: Choroideremia carriers, particularly those with severe retinal phenotypes, exhibit progressive retinal degeneration, as evident by multimodal imaging biomarkers and functional testing. Clinicians should not rely on retinal severity classification alone to assess disease progression., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Background/aims: Recent decades have seen significant advances in both structural and functional testing of retinal disease. However, the current clinical value of specific testing modalities, as well as future trends, need to be clearly identified in order to highlight areas for further development in routine care and clinical trials., Methods: We designed a modified two-round Delphi study to obtain the opinion of a multidisciplinary group of 33 international experts involved in the field of retinal disease management/research to determine the level of agreement and consensus regarding the value and performance of specific structural and functional testing methods for retinal disease. On a Likert scale, a median of 1-2 indicated disagreement with the statement, and 5-6 indicated agreement with the statement. An IQR of ≤2 indicated consensus in the responses. Several questions also allowed comments on responses., Results: There was overall agreement that structural testing currently predominates for detection and monitoring. There was moderate agreement that functional testing remains important and will continue to do so in the future because it provides complementary information. Certain respondents considered that properly designed and applied psychophysical tests are as reliable and repeatable as structural observations and that functional changes are the most important in the long run. Respondents considered future care and research to require a combination of structural and functional testing with strong consensus that the relative importance will depend on disease type and stage., Conclusion: The study obtained important insights from a group of international experts regarding current and future needs in the management of retinal disease using a mix of quantitative and qualitative approaches. Responses provide a rich range of opinions that will be of interest to researchers seeking to design tests for future patient care and clinical trials., Competing Interests: Competing interests: The study was funded and facilitated by Alliance Pharmaceuticals Ltd. The authors received financial payment from the funder, but the study objectives, design, analysis and subsequent write-up was entirely undertaken by the authors. Alliance Pharmaceuticals Ltd played no part in the study design, the formulation of the questions, or the analysis and respondents were not paid for their responses. The online questionnaires were administered by the funder and the data curated by Alliance staff to render the authors blind to the identity of respondents until final analysis., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Background: Tuberculosis (TB)-immunoreactivity, measured in vivo (tuberculin skin test (TST)) or in vitro (interferon gamma release assay (IGRA)), can be found in latent, active or even following clearance of TB infection. In this case-control study, we compared the systemic and ocular outcomes between patients with or without TB-immunoreactivity, who received immunomodulatory therapy (IMT) for non-infectious uveitis., Methods: We retrospectively reviewed charts of patients with (cases) or without (controls) TB-immunoreactivity (TST±IGRA), who received conventional IMT for ≥6 months, for the treatment of non-infectious uveitis. Patients who received prior or concomitant anti-TB therapy were excluded. Systemic and ocular outcomes were compared between both groups., Results: 36 cases and 70 controls (gender-matched and age-matched) were included. New-onset pulmonary or extrapulmonary TB developed in one case and none of the controls. Based on this outcome, the absolute risk increase for systemic TB reactivation was noted to be 0.028 (95% CI 0.005 to 0.051) and the number needed to harm was 36. The incidence of persistent or recurrent (worsening ≥2 grades) intraocular inflammation during IMT was comparable between both groups (cases 18/36, controls 35/70, p=1.0). A change in anatomical site of presentation at recurrence was not seen in any case, but in six controls (p=0.15). No new focal chorio-retinal lesions were noted in either group., Conclusions: Conventional IMT has a very low risk of systemic TB reactivation, and no additional detrimental effect on ocular outcomes, in TB-immunoreactive patients with non-infectious uveitis., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Aim: This study used swept-source optical coherence tomography (SS-OCT) to investigate subfoveal choroidal thickness (SFCT) in patients with thyroid-associated ophthalmopathy (TAO) who displayed different levels of disease activity and severity., Methods: Thirty patients with TAO (60 eyes) and 38 healthy controls (67 eyes) in Shanghai, China, were recruited for this study. Disease activity and severity were graded using European Group on Graves' Orbitopathy standardised criteria. SFCT values were determined by SS-OCT., Results: In total, 129 eyes were included in the final analysis. The mean SFCT was significantly thicker among patients with active disease (276.23±84.01 µm) than among patients with inactive disease (224.68±111.61 µm; p=0.049) or healthy controls (223.56±78.69 µm; p=0.01). There were no differences in SFCT among patients with moderate-to-severe disease, patients with severe disease and healthy controls (p>0.05). Changes in SFCT demonstrated strong predictive ability to distinguish active TAO from inactive TAO (area under the curve=0.659, 95% CI 0.496 to 0.822)., Conclusions: SFCT was strongly associated with Clinical Activity Score in patients with TAO. Choroidal thickening was observed during active TAO. SS-OCT offers a non-invasive method for follow-up assessment., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY. Published by BMJ.)

Background/aims: The goal of health research is to improve patients care and outcomes. Thus, it is essential that research addresses questions that are important to patients and clinicians. The aim of this study was to develop a list of priorities for glaucoma research involving stakeholders from different countries in Europe., Methods: We used a three-phase method, including a two-round electronic Delphi survey and a workshop. The clinician and patient electronic surveys were conducted in parallel and independently. For phase I, the survey was distributed to patients from 27 European countries in 6 different languages, and to European Glaucoma Society members, ophthalmologists with expertise in glaucoma care, asking to name up to five research priorities. During phase II, participants were asked to rank the questions identified in phase I using a Likert scale. Phase III was a 1 day workshop with patients and clinicians. The purpose was to make decisions about the 10 most important research priorities using the top 20 priorities identified by patients and clinicians., Results: In phase I, 308 patients and 150 clinicians were involved. In phase II, the highest-ranking priority for both patients and clinicians was 'treatments to restore vision'. In phase III, eight patients and four clinicians were involved. The top three priorities were 'treatments to stop sight loss', 'treatments to restore vision' and 'improved detection of worsening glaucoma'., Conclusion: We have developed a list of priorities for glaucoma research involving clinicians and patients from different European countries that will help guide research efforts and investment., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY. Published by BMJ.)

Objectives: To evaluate the feasibility and accuracy of a portable, self-imaging optical coherence tomography (OCT) for measuring central subfield thickness (CST) and achieving diagnostic concordance for retinal lesions compared with clinic-based spectral-domain OCT (SD-OCT)., Methods: This comparative, cross-sectional study was conducted between August 2020 and February 2021. Two groups of adult participants were recruited: (1) a selected cohort of 160 participants with confirmed diagnosis and (2) a consecutive cohort of 315 participants recruited randomly. All participants underwent self-imaging OCT examination, as well as standard OCT examination. CST was automatically calculated for comparisons between the two OCT devices. Diagnostic concordance for retinal lesions and the success rate of self-imaging were assessed within the consecutive cohort., Results: In the selected cohort, self-imaging OCT images yielded consistent CST with SD-OCT, with a mean difference of 0.1±7.7 µm for normal eyes, 4.9±10.6 µm for macular oedema, -1.3±9.5 µm for choroidal neovascularisation, 5.0±7.8 µm for epiretinal membrane. The self-imaging OCT also demonstrated good repeatability, with a mean test-retest difference in CST of 0.7±3.9 µm and limits of agreement ranging from -6.9 to 8.3 µm. Additionally, within the consecutive cohort, interdevice κ values ranged for detecting various retinal lesions ranged from 0.8 to 1.0, except in the cases of retinal detachment (κ=0.5). All eyes (100%) in the selected cohort and 242 eyes (76.8%) in the consecutive cohort successfully completed self-imaging. Participants spent less time on self-imaging compared with SD-OCT operated by a technician (66.7±20.1 vs 73.3±32.5, p<0.01). A majority of participants (90%) found the self-imaging process 'easy' and 'comfortable'., Conclusions and Relevance: This study demonstrates that our self-imaging OCT and clinical-used SD-OCT are highly consistent not only in measuring the CST but also in identifying most retinal lesions., Competing Interests: Competing interests: Liu Y and Zheng Y reported that they hold patents on the self-imaging device used in this study, and on the automated imaging analysis system associated with the imaging software used in this study. No other disclosures were reported., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Background/aims: This study aims to evaluate the clinical efficacy of botulinum toxin type A (BTXA) injection and augmented-dosed surgery in the treatment of acute acquired concomitant esotropia (AACE), and explore potential risk factors associated with recurrence., Methods: A total of 104 patients diagnosed with AACE between October 2020 and January 2021 were included and voluntarily chose to undergo augmented surgery or BTXA injection. The follow-up assessments ended in November 2022. Multivariable linear regression analysis was used to identify potential factors that influence the dose-response of bilateral medial rectus recession (MRrec). Kaplan-Meier survival analyses and Cox proportional hazards models were performed to evaluate rate and risk factors for AACE relapse., Results: A total of 31 AACE patients chose augmented-dosed esotropia surgery, and 73 chose BTXA treatment. During the 2-year follow-up, the surgical group achieved more stable postoperative results with no recurrence of diplopia, while only 68.68% (95% CI 55.31% to 78.79%) patients achieved orthophoria in the BTXA group. For patients undergoing BTXA treatment, hours of near work per day were demonstrated to be a significant risk factor for AACE relapse (HR 1.29, 95% CI 1.00 to 1.67). The dose-response of augmented-dosed bilateral MRrec was positively correlated with preoperative deviation angle (R 2 =0.833; β=0.043, 95% CI 0.031 to 0.055; p<0.001)., Conclusion: Our findings provided quantitative evidence that augmented-dosed surgery would achieve more stable and favourable surgical outcomes for AACE patients compared with BTXA injection. However, BTXA treatment is still proposed for patients with small deviation angles due to its advantages of reduced trauma, operational simplicity, low cost and quick recovery., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Background/aims: Recent work has called into question the ability of visual acuity (VA) to accurately represent changes in visual function in infantile nystagmus (IN). This systematic review investigated factors affecting visual performance in IN, to guide development of suitable alternatives to VA., Methods: Included studies used an experimental manipulation to assess changes in visual function in people with IN. Interventional studies, case series and case studies were excluded. Six databases were searched in August 2023. Selection, detection, attrition and measurement bias were assessed. Due to heterogeneous methodologies, narrative synthesis was undertaken., Results: Eighteen relevant papers were identified, 11 of which complied with the review criteria. Articles were grouped according to the factor manipulated to evoke within-participant changes in performance (motion blur, psychological state, gaze angle or visual crowding). Optotype, image, grating and moving stimuli have been employed under varying lighting conditions and exposure duration., Conclusion: Several factors affecting visual performance should be considered when assessing visual function in IN. While maximum VA is a useful metric, its measurement deliberately minimises nystagmus-specific factors such as changes in visual performance with gaze angle and the 'slow to see' phenomenon. Maximum VA can be measured using the null zone, providing unlimited viewing time, reducing stress/mental load and minimising visual crowding. Gaze-dependent functional vision space is a promising measure which quantifies the impact of the null zone but does not consider temporal vision. Although no complete measurement technique has yet been proven, this review provides insights to guide future work towards development of appropriate methods., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Aims: To evaluate the effectiveness of glaucoma screening using glaucoma suspect (GS) referral criteria assessed on colour fundus photographs in Singapore's Integrated Diabetic Retinopathy Programme (SiDRP)., Methods: A case-control study. This study included diabetic subjects who were referred from SiDRP with and without GS between January 2017 and December 2018 and reviewed at Singapore National Eye Centre. The GS referral criteria were based on the presence of a vertical cup-to-disc ratio (VCDR) of ≥0.65 and other GS features. The final glaucoma diagnosis confirmed from electronic medical records was retrospectively matched with GS status. The sensitivity, specificity and positive predictive value (PPV) of the test were evaluated., Results: Of 5023 patients (2625 with GS and 2398 without GS) reviewed for glaucoma, 451 (9.0%, 95% CI 8.2% to 9.8%) were confirmed as glaucoma. The average follow-up time was 21.5±10.2 months. Using our current GS referral criteria, the sensitivity, specificity and PPV were 81.6% (95% CI 77.7% to 85.1%), 50.6% (95% CI 49.2% to 52.1%) and 14.0% (95% CI 13.4% to 14.7%), respectively, resulting in 2257 false positive cases. Increasing the VCDR cut-off for referral to ≥0.80, the specificity increased to 93.9% (95% CI 93.1% to 94.5%) but the sensitivity decreased to 11.3% (95% CI 8.5% to 14.6%), with a PPV of 15.4% (95% CI 12.0% to 19.4%)., Conclusions: Opportunistic screening for glaucoma in a lower VCDR group could result in a high number of unnecessary referrals. If healthcare infrastructures are limited, targeting case findings on a larger VCDR group with high specificity will still be beneficial., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Background: Ultrasound imaging is suitable for detecting and diagnosing ophthalmic abnormalities. However, a shortage of experienced sonographers and ophthalmologists remains a problem. This study aims to develop a multibranch transformer network (MBT-Net) for the automated classification of multiple ophthalmic diseases using B-mode ultrasound images., Methods: Ultrasound images with six clinically confirmed categories, including normal, retinal detachment, vitreous haemorrhage, intraocular tumour, posterior scleral staphyloma and other abnormalities, were used to develop and evaluate the MBT-Net. Images were derived from five different ultrasonic devices operated by different sonographers and divided into training set, validation set, internal testing set and temporal external testing set. Two senior ophthalmologists and two junior ophthalmologists were recruited to compare the model's performance., Results: A total of 10 184 ultrasound images were collected. The MBT-Net got an accuracy of 87.80% (95% CI 86.26% to 89.18%) in the internal testing set, which was significantly higher than junior ophthalmologists (95% CI 67.37% to 79.16%; both p<0.05) and lower than senior ophthalmologists (95% CI 89.45% to 92.61%; both p<0.05). The micro-average area under the curve of the six-category classification was 0.98. With reference to comprehensive clinical diagnosis, the measurements of agreement were almost perfect in the MBT-Net (kappa=0.85, p<0.05). There was no significant difference in the accuracy of the MBT-Net across five ultrasonic devices (p=0.27). The MBT-Net got an accuracy of 82.21% (95% CI 78.45% to 85.44%) in the temporal external testing set., Conclusions: The MBT-Net showed high accuracy for screening and diagnosing multiple ophthalmic diseases using only ultrasound images across mutioperators and mutidevices., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Background: Accurate risk stratification of uveal melanoma (UM) patients is important for determining the interval and frequency of surveillance. Loss of BAP1 expression has been shown to be strongly associated with UM-related death and metastasis., Methods: In this study of 164 enucleated UMs, we assessed the prognostic role of preferentially expressed antigen in melanoma (PRAME) expression and Ki67 proliferation index measured by digital quantitation using QuPath programme in patients with BAP1-positive and BAP1-loss UMs., Results: In univariate analyses with log-rank tests and Kaplan-Meier curves, PRAME further stratified only overall survival (OS) in BAP1-positive and BAP1-loss tumour groups. However, Ki67 further stratified both OS and disease-free survival (DFS) in BAP1-positive and BAP1-loss tumour groups. In multivariate analyses, Ki67 percentage and BAP1 were independent survival predictors for both OS and DFS, whereas PRAME was not a significant covariate. In model comparisons, combined Ki67 and BAP1 performed better than combined PRAME and BAP1 in risk-stratifying patients for both OS and DFS. Ki67 was better than PRAME in risk stratification of BAP1-positive UMs. Low Ki67 index correlated with significantly prolonged DFS in BAP1-loss UMs., Conclusion: A panel of Ki67 and BAP1 could be a helpful risk stratification strategy for UM., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Background: Unpreserved single-dose unit (SDU) eye drops are commonly used to avoid benzalkonium chloride-related toxicity. Although intended for single use, many patients report off-label repeated use of SDUs over a prolonged period. We investigated whether repeated use of dexamethasone 0.1% SDUs in the same patient increases the bacterial contamination rate., Methods: We prospectively enrolled patients scheduled for inpatient corneal and glaucoma surgery receiving dexamethasone 0.1% SDU four times per day from the same vial. To assess contamination rates, one drop from the vial was cultured immediately after opening the SDU (t0), 10 hours later after four drop applications (t10) and 24 hours after opening without further drop applications (t24). Conjunctival swabs were taken before and after drop application. Contamination rate was assessed with a standard clinical culturing protocol without introducing a positive control., Results: 110 eyes of 109 patients were evaluated. Drops collected immediately after opening the SDU (t0) were contaminated in 9/110 cultures (8.1%). At t10, 13/110 cultures were contaminated (11.8%; p=0.267) and 11/110 at t24 (10.0%; t24 vs t0; p=1.00). In 5 of 21 cases of contaminated drops at t10 and/or t24, the same isolates were cultured from the initial conjunctival swab and the SDU. In three cases, the same bacterial species was found in consecutive samples., Conclusion: The contamination rate of the SDU did not increase after multiple use within 24 hours. Contamination from fingertip flora was more likely than from ocular surface flora. Reuse of dexamethasone 0.1% SDU in the same patient within 24 hours appears to be safe., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Background/aims: To assess pupillary light responses (PLRs) in eyes with high myopia (HM) and evaluate the ability of handheld chromatic pupillometry (HCP) to identify glaucomatous functional loss in eyes with HM., Methods: This prospective, cross-sectional study included 28 emmetropes (EM), 24 high myopes without glaucoma (HM) and 17 high myopes with confirmed glaucoma (HMG), recruited at the Singapore National Eye Center. Monocular PLRs were evaluated using a custom-built handheld pupillometer that recorded changes in horizontal pupil radius in response to 9 s of exponentially increasing blue (469.1 nm) and red (640.1 nm) lights. Fifteen pupillometric features were compared between groups. A logistic regression model (LRM) was used to distinguish HMG eyes from non-glaucomatous eyes (EM and HM)., Results: All pupillometric features were similar between EM and HM groups. Phasic constriction to blue (p<0.001) and red (p=0.006) lights, and maximum constriction to blue light (p<0.001) were reduced in HMG compared with EM and HM. Pupillometric features of melanopsin function (postillumination pupillary response, PIPR area under the curve (AUC) 0-12 s (p<0.001) and PIPR 6 s (p=0.01) to blue light) were reduced in HMG. Using only three pupillometric features, the LRM could classify glaucomatous from non-glaucomatous eyes with an AUC of 0.89 (95% CI 0.77 to 1.00), sensitivity 94.1% (95% CI 82.4% to 100.0%) and specificity 78.8% (95% CI 67.3% to 90.4%)., Conclusion: PLRs to ramping-up light stimuli are unaltered in highly myopic eyes without other diagnosed ocular conditions. Conversely, HCP can distinguish glaucomatous functional loss in eyes with HM and can be a useful tool to detect/confirm the presence of glaucoma in patients with HM., Competing Interests: Competing interests: RPN, DM and TA have a patent application based on the handheld pupillometer used in this study (PCT/SG2018/050204): Handheld ophthalmic and neurological screening device. The rest of the authors have no conflicts of interest to disclose., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Aims: To explore the possibility of implementing Choosing Wisely on ocular patients in China by investigating the prevalence of abnormalities in routine preoperative blood tests (RPBTs) and its turnaround time (TAT)., Methods: Data from 102 542 ocular patients between January 2016 and December 2018, at Zhongshan Ophthalmic Center, were pooled from the laboratory information system. The test results were divided into normal and abnormal, including critical values. Ocular diseases were stratified into 11 subtypes based on the primary diagnosis. The TAT of 243 350 blood tests from January 2017 to December 2018 was categorised into transportation time and intralaboratory time., Results: RPBT was grouped into complete blood count (CBC), blood biochemistry (BBC), blood coagulation (BCG) and blood-borne pathogens (BBP), completed for 97.22%, 87.66%, 94.41% and 95.35% of the recruited patients (male, 52 549 (51.25%); median(IQR) age, 54 (29-67) years), respectively. Stratified by the test items, 9.19% (95% CI 9.07% to 9.31%) were abnormal results, and 0.020% (95% CI 0.019% to 0.022%) were critical; most abnormalities were on the CBC, while glucose was the most common critical item. Classified by the patients' primary diagnosis, 76.97% (95% CI 76.71% to 77.23%) had at least one abnormal result, and 0.28% (95% CI 0.25% to 0.32%) were critical; abnormal findings were reported in 45.29% (95% CI 44.98% to 45.60%), 54.97% (95% CI 54.65% to 55.30%), 30.29% (95% CI 30.00% to 30.58%) and 11.32% (95% CI 11.12% to 11.52%) for the CBC, BBC, BCG and BBP tests, respectively. The median transportation time and intralaboratory TAT of the samples were 12 min and 78 min respectively., Conclusion: Blood abnormalities are common in ocular patients. With acceptable timelines, RPBT is still indispensable in China for patient safety., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Aim: To perform an independent validation of deep learning (DL) algorithms for automated scleral spur detection and measurement of scleral spur-based biometric parameters in anterior segment optical coherence tomography (AS-OCT) images., Methods: Patients receiving routine eye care underwent AS-OCT imaging using the ANTERION OCT system (Heidelberg Engineering, Heidelberg, Germany). Scleral spur locations were marked by three human graders (reference, expert and novice) and predicted using DL algorithms developed by Heidelberg Engineering that prioritise a false positive rate <4% (FPR4) or true positive rate >95% (TPR95). Performance of human graders and DL algorithms were evaluated based on agreement of scleral spur locations and biometric measurements with the reference grader., Results: 1308 AS-OCT images were obtained from 117 participants. Median differences in scleral spur locations from reference locations were significantly smaller (p<0.001) for the FPR4 (52.6±48.6 µm) and TPR95 (55.5±50.6 µm) algorithms compared with the expert (61.1±65.7 µm) and novice (79.4±74.9 µm) graders. Intergrader reproducibility of biometric measurements was excellent overall for all four (intraclass correlation coefficient range 0.918-0.997). Intergrader reproducibility of the expert grader (0.567-0.965) and DL algorithms (0.746-0.979) exceeded that of the novice grader (0.146-0.929) for images with narrow angles defined by OCT measurement of angle opening distance 500 µm anterior to the scleral spur (AOD500)<150 µm., Conclusions: DL algorithms on the ANTERION approximate expert-level measurement of scleral spur-based biometric parameters in an independent patient population. These algorithms could enhance clinical utility of AS-OCT imaging, especially for evaluating patients with angle closure and performing intraocular lens calculations., Competing Interests: Competing interests: BYX and ASH receive research support from Heidelberg Engineering. MS is employed by Heidelberg Engineering., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Aims: To create and assess the performance of an artificial intelligence-based image analysis tool for the measurement and quantification of the corneal neovascularisation (CoNV) area., Methods: Slit lamp images of patients with CoNV were exported from the electronic medical records and included in the study. An experienced ophthalmologist made manual annotations of the CoNV areas, which were then used to create, train and evaluate an automated image analysis tool that uses deep learning to segment and detect CoNV areas. A pretrained neural network (U-Net) was used and fine-tuned on the annotated images. Sixfold cross-validation was used to evaluate the performance of the algorithm on each subset of 20 images. The main metric for our evaluation was intersection over union (IoU)., Results: The slit lamp images of 120 eyes of 120 patients with CoNV were included in the analysis. Detections of the total corneal area achieved IoU between 90.0% and 95.5% in each fold and those of the non-vascularised area achieved IoU between 76.6% and 82.2%. The specificity for the detection was between 96.4% and 98.6% for the total corneal area and 96.6% and 98.0% for the non-vascularised area., Conclusion: The proposed algorithm showed a high accuracy compared with the measurement made by an ophthalmologist. The study suggests that an automated tool using artificial intelligence may be used for the calculation of the CoNV area from the slit-lamp images of patients with CoNV., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Background/aims: To evaluate the diagnostic accuracy of spectral-domain optical coherence tomography (SD OCT) combined with OCT angiography (OCTA) for myopic myopic macular neovascularisation (MNV) activity., Methods: Both eyes of patients with myopic MNV diagnosed with fluorescein angiography (FA), SD OCT and OCTA were assessed by unmasked investigators. The images were deidentified and randomised before graded by masked investigators, who determined the presence of active myopic MNV by using SD OCT together with OCTA without FA and by FA alone, respectively. The findings of masked investigators were compared with unmasked investigators., Results: 213 eyes of 110 patients comprising 499 imaging episodes were eligible for grading. For diagnosing new-onset myopic MNV without FA, combined use of SD OCT and OCTA had a sensitivity of 0.94, specificity of 0.84 and area under the curve (AUC) of 0.92. FA had a sensitivity of 0.52 (p<0.01), specificity of 0.80 (p=0.38) and AUC of 0.66 (p<0.01). For recurrent myopic MNV, the combination of SD OCT and OCTA had a sensitivity of 0.98, specificity of 0.78 and AUC of 0.88. FA had a sensitivity of 0.50 (p=0.04), specificity of 0.76 (p=0.85) and AUC of 0.63 (p=0.01). Myopic traction maculopathy was more frequently associated with recurrent myopic MNV (p<0.01)., Conclusion: SD OCT with dense volumetric scan was highly sensitive for diagnosing myopic MNV. The addition of OCTA improved the diagnostic specificity without FA. Monitoring of the longitudinal changes on SD OCT and judicious use of FA is a reliable surveillance strategy for myopic MNV., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Purpose: To compare the use of toluidine blue 1% eye drops with anterior segment optical coherence tomography (OCT) for the determination of tumour margins in patients with ocular surface squamous neoplasia (OSSN)., Methods: The study was conducted from July 2020 to June 2021 at the Ocular Oncology department at the Federal University of São Paulo, Brazil. Slit-lamp photographs after toluidine blue staining and OCT of the anterior segment were taken on the same day from patients with OSSN. Photographs and OCT images were analyzed quantitatively using the software ImageJ and IMAGEnet®, respectively. The agreement between techniques was evaluated qualitatively through the Bland-Altman graph and quantitatively through intraclass correlation (ICC)., Results: A total of 21 participants (71.43% males) with a clinical diagnosis of OSSN were included in the study. The average + SD diameter along the chosen axes was 4.43 ± 2.08 mm with OCT of 4.37 ± 2.03 mm with toluidine blue, a difference not statistically significant (p = 0.2891). The Bland-Altman analysis indicated a good qualitative agreement between the methods, with all cases inserted within the limits of agreement from -0.3217 to 0.4268. The ICC quantitative analysis showed an almost perfect agreement of 99.57% (95%CI: 98.96-99.83%; p < 0.001)., Conclusions: Our findings showed that OCT and toluidine eye drops are equivalent in determining margins for tumour measurements, which is particularly relevant in low-income settings where anterior segment OCT is not available. The use of toluidine blue 1% could be an useful alternative to quantify the size of the tumour, help to monitor tumour growth, and outline margins for surgical planning., Competing Interests: Declaration of conflicting interestsThe authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Objective: The objective of this study is to determine the validity and reliability of the red filter meibography by smartphone compared with infrared in assessing meibomian gland drop-out., Methods and Analysis: An analytical cross-sectional study was done with a total of 35 subjects (68 eyes) with suspected MGD based on symptoms and lid morphological abnormalities. Meibomian glands were photographed using two smartphones (Samsung S9 and iPhone XR) on a slit-lamp with added red filter. Images were assessed subjectively using meiboscore by the two raters and drop-out percentages were assessed by ImageJ., Results: There was no agreement in meiboscore and a minimal level of agreement in drop-out percentages between red filter meibography and infrared. Inter-rater reliability showed no agreement between two raters. Intra-rater reliability demonstrated weak agreement in rater 1 and no agreement in rater 2., Conclusion: Validity of the red filter meibography technique by smartphones is not yet satisfactory in evaluating drop-out. Further improvement on qualities of images must be done and research on subjective assessment was deemed necessary due to poor results of intrarater and inter-rater reliability., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Purpose: To prove the safety and performance of the hypothermic corneal storage medium "Corneal Chamber" and the rinsing solution "PSS-L" in support of the new Conformité Européenne (CE) certification process in accordance with the Medical Device Regulation., Methods: Fifteen (n=15) human donor corneas and 11 (n=11) porcine corneas were evaluated for the following parameters: endothelial cell density (ECD) and mortality, percentage of hexagonal cells (HEX%), coefficient of cellular area variation (CV%) and corneal transparency at Day 0 and after 14±1 days of storage in Corneal Chamber medium at 2-8°C. Then, the same parameters were assessed after rinsing of corneas in PSS-L for 1 min at room temperature. Evaluation of gentamicin sulfate carryover after corneal storage and PSS-L rinsing was performed by ultra-high performance liquid chromatography analysis on human corneas homogenates., Results: Human and porcine corneas stored in Corneal Chamber medium showed a good overall quality of the tissue according to the quality parameters evaluated. In particular, mean ECD, HEX% and CV% did not show statistically significant changes at the end of storage and endothelial mortality increased to 3.1±3.3 and 7.8±3.5% in human and porcine corneas, respectively. Tissue rinsing with PSS-L did not affect the quality parameters evaluated before and gentamicin sulfate residues were absent in human corneas., Conclusions: Corneal preservation in Corneal Chamber medium at 2-8°C for 14 days and the corneal rinse with PSS-L are safe and effective procedures allowing the preservation of the corneal quality parameters as well as the complete elimination of gentamicin sulfate from the tissues before transplantation.Cite Now., Competing Interests: Competing interests: LG, UR, CG, OR and JDT are employees of AL.CHI.MI.A. S.R.L. ER was consulted by AL.CHI.MI.A. S.R.L. for the purpose of conducting statistical analysis., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Background: Glaucoma staging is critical for treatment planning but has rarely been tested in severe/end-stage disease. We compared the performance of the Disc Damage Likelihood Scale (DDLS) and cup:disc ratio (CDR) using a functional glaucoma staging system (GSS) as the reference standard., Methods: Post hoc analysis of a randomised controlled trial at the Eye Department of Kilimanjaro Christian Medical Centre, Tanzania. Eligible participants (aged ≥18 years) with open-angle glaucoma, intraocular pressure (IOP) of >21 mm Hg, were randomised to timolol 0.5% eye drops or selective laser trabeculoplasty. Fundoscopy established vertical and horizontal CDRs and DDLS. Visual acuity and static visual fields were graded (GSS). The study used area under the receiver operating characteristic (AROC) curves and Spearman's rank correlation coefficients to compare staging systems. Logistic regression with generalised estimating equations determined risk factors of functional severe/end-stage glaucoma., Results: 382 eyes (201 participants) were evaluated; 195 (51%) had severe or end-stage glaucoma; mean IOP was 26.7 (SD 6.9) mm Hg. DDLS yielded an AROC of 0.90 (95% CI 0.87 to 0.93), vertical cup:disc ratio (vCDR) of 0.88 (95% CI 0.85 to 0.91, p=0.048) for identifying severe/end-stage disease. Correlation coefficients comparing GSS to DDLS and vCDRs were 0.73 and 0.71, respectively. Advanced structural stages, vision impairment, higher IOP and less financial resources were risk factors of functional severe/end-stage glaucoma., Conclusion: This study indicates that both structural staging systems can differentiate severe/end-stage glaucoma from less severe disease, with a moderate advantage of DDLS over CDR. Clinical examination of the optic disc plays an important role in addition to functional assessment when managing severe/end-stage glaucoma., Competing Interests: Competing interests: GG reports personal fees from Alcon, Allergan, Belkin, Equinox, Genentech–Roche, Glaukos, Ivantis, Reichert, Sight Sciences, and Thea; grants from Belkin, Santen and Thea; and non-financial involvement with the patient advocacy group GlaucomaUK, outside the submitted work; he is also a coinvestigator on three other major SLT trials (LIGHT, COAST and Belkin laser). All other authors declare no competing interests., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY. Published by BMJ.)

Background/aims: This study evaluates the performance of the Airdoc retinal artificial intelligence system (ARAS) for detecting multiple fundus diseases in real-world scenarios in primary healthcare settings and investigates the fundus disease spectrum based on ARAS., Methods: This real-world, multicentre, cross-sectional study was conducted in Shanghai and Xinjiang, China. Six primary healthcare settings were included in this study. Colour fundus photographs were taken and graded by ARAS and retinal specialists. The performance of ARAS is described by its accuracy, sensitivity, specificity and positive and negative predictive values. The spectrum of fundus diseases in primary healthcare settings has also been investigated., Results: A total of 4795 participants were included. The median age was 57.0 (IQR 39.0-66.0) years, and 3175 (66.2%) participants were female. The accuracy, specificity and negative predictive value of ARAS for detecting normal fundus and 14 retinal abnormalities were high, whereas the sensitivity and positive predictive value varied in detecting different abnormalities. The proportion of retinal drusen, pathological myopia and glaucomatous optic neuropathy was significantly higher in Shanghai than in Xinjiang. Moreover, the percentages of referable diabetic retinopathy, retinal vein occlusion and macular oedema in middle-aged and elderly people in Xinjiang were significantly higher than in Shanghai., Conclusion: This study demonstrated the dependability of ARAS for detecting multiple retinal diseases in primary healthcare settings. Implementing the AI-assisted fundus disease screening system in primary healthcare settings might be beneficial in reducing regional disparities in medical resources. However, the ARAS algorithm must be improved to achieve better performance., Trial Registration Number: NCT04592068., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Aims: To develop an algorithm to classify multiple retinal pathologies accurately and reliably from fundus photographs and to validate its performance against human experts., Methods: We trained a deep convolutional ensemble (DCE), an ensemble of five convolutional neural networks (CNNs), to classify retinal fundus photographs into diabetic retinopathy (DR), glaucoma, age-related macular degeneration (AMD) and normal eyes. The CNN architecture was based on the InceptionV3 model, and initial weights were pretrained on the ImageNet dataset. We used 43 055 fundus images from 12 public datasets. Five trained ensembles were then tested on an 'unseen' set of 100 images. Seven board-certified ophthalmologists were asked to classify these test images., Results: Board-certified ophthalmologists achieved a mean accuracy of 72.7% over all classes, while the DCE achieved a mean accuracy of 79.2% (p=0.03). The DCE had a statistically significant higher mean F1-score for DR classification compared with the ophthalmologists (76.8% vs 57.5%; p=0.01) and greater but statistically non-significant mean F1-scores for glaucoma (83.9% vs 75.7%; p=0.10), AMD (85.9% vs 85.2%; p=0.69) and normal eyes (73.0% vs 70.5%; p=0.39). The DCE had a greater mean agreement between accuracy and confident of 81.6% vs 70.3% (p<0.001)., Discussion: We developed a deep learning model and found that it could more accurately and reliably classify four categories of fundus images compared with board-certified ophthalmologists. This work provides proof-of-principle that an algorithm is capable of accurate and reliable recognition of multiple retinal diseases using only fundus photographs., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Purpose: To study the fungal microbiome (mycobiome) and the virome of the lacrimal sacs in patients with primary acquired nasolacrimal duct obstruction (PANDO)., Methods: A prospective study was performed on 10 consecutive samples of the lacrimal sac contents obtained from patients with PANDO. The samples were obtained from the lacrimal sacs under endoscopy guidance and immediately transported on ice to the laboratory. Following DNA extraction and library preparation, a whole shotgun metagenome sequencing was performed on the Illumina platform (NOVASEQ 6000). The fungal internal transcript spacer analysis was performed using the PIPITS v2.7 . The viral taxonomy profiling was performed using Kraken2 against the virus database., Results: The taxonomic hit distribution across the lacrimal sac samples showed rich fungal diversity (4 phyla, 12 classed, 21 families and 26 genera). The major phyla were Ascomycota and Basidiomycota, and the key genera identified were Alternaria, Hyphopichia, Malassezia, Aspergillus and Epicoccum . The virome analysis identified 13 phyla, 15 classes and 27 families. The viruses were commonly from the families Poxviridae, Retroviridae, Siphoviridae and Myoviridae , Poxviridae being the most prevalent family. The BeAn 58058 virus, a member of the Poxviridae family, was the most abundant in all the samples., Conclusion: The present study is the first whole metagenome sequencing exclusively of the fungal microbiome and virome from the lacrimal sacs of patients with PANDO. The lacrimal sacs harbour diverse fungal and viral communities with distinct ecosystem dynamics. Further studies of their functions and interactions with the hosts would provide valuable insights., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Background/aims: Birdshot uveitis (BU) is a chronic autoimmune posterior uveitis, mostly affecting middle-aged Caucasians. There is a strong association with HLA-A29 and T-cell activation. Safety and efficacy of abatacept, an inhibitor of T-cell costimulation, is tested in active BU., Methods: Fifteen patients with active BU were treated with monotherapy of weekly subcutaneous injections of abatacept 125 mg/mL. Time-to-treatment failure was evaluated as a primary outcome. The secondary objective was to evaluate the utility of different outcome measures to monitor disease activity. Safety was evaluated by adverse event reporting and serial blood analyses., Results: At the year-1 endpoint, there was significant improvement in vitreous haze grade (p=0.0014), central choroidal thickness (CCT) (p=0.0011), Fluorescein Angiography (FA) Score (p=0.0014), Indocyanine Green Angiography (ICGA) Score (p<0.001) and total dual FA-ICGA Score (p<0.001). Best corrected visual acuity (BCVA) (p=0.8354) and central retinal thickness (CRT) (p=0.3549) did not change significantly. There were no serious adverse events reported. In total, 4 out of 15 patients left the trial during year 1 of whom 2 experienced treatment failure., Conclusions: Abatacept is very efficacious to treat both retinal vasculitis and choroiditis in patients with BU and is well tolerated. BCVA and CRT are inadequate to monitor disease activity. On the other hand, CCT is a promising non-invasive tool to detect treatment response in early active BU and dual FA-ICGA Score is very helpful to evaluate retinal vasculitis and choroiditis quantitatively., Trial Registration Number: NCT03871361., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Background/aims: Deep learning systems (DLSs) for diabetic retinopathy (DR) detection show promising results but can underperform in racial and ethnic minority groups, therefore external validation within these populations is critical for health equity. This study evaluates the performance of a DLS for DR detection among Indigenous Australians, an understudied ethnic group who suffer disproportionately from DR-related blindness., Methods: We performed a retrospective external validation study comparing the performance of a DLS against a retinal specialist for the detection of more-than-mild DR (mtmDR), vision-threatening DR (vtDR) and all-cause referable DR. The validation set consisted of 1682 consecutive, single-field, macula-centred retinal photographs from 864 patients with diabetes (mean age 54.9 years, 52.4% women) at an Indigenous primary care service in Perth, Australia. Three-person adjudication by a panel of specialists served as the reference standard., Results: For mtmDR detection, sensitivity of the DLS was superior to the retina specialist (98.0% (95% CI, 96.5 to 99.4) vs 87.1% (95% CI, 83.6 to 90.6), McNemar's test p<0.001) with a small reduction in specificity (95.1% (95% CI, 93.6 to 96.4) vs 97.0% (95% CI, 95.9 to 98.0), p=0.006). For vtDR, the DLS's sensitivity was again superior to the human grader (96.2% (95% CI, 93.4 to 98.6) vs 84.4% (95% CI, 79.7 to 89.2), p<0.001) with a slight drop in specificity (95.8% (95% CI, 94.6 to 96.9) vs 97.8% (95% CI, 96.9 to 98.6), p=0.002). For all-cause referable DR, there was a substantial increase in sensitivity (93.7% (95% CI, 91.8 to 95.5) vs 74.4% (95% CI, 71.1 to 77.5), p<0.001) and a smaller reduction in specificity (91.7% (95% CI, 90.0 to 93.3) vs 96.3% (95% CI, 95.2 to 97.4), p<0.001)., Conclusion: The DLS showed improved sensitivity and similar specificity compared with a retina specialist for DR detection. This demonstrates its potential to support DR screening among Indigenous Australians, an underserved population with a high burden of diabetic eye disease., Competing Interests: Competing interests: PAK has acted as a consultant for DeepMind, Roche, Novartis and Apellis and is an equity owner in Big Picture Medical. He has received speaker fees from Heidelberg Engineering, Topcon, Allergan and Bayer. FH, RS, PB and RT are employees of Google LLC and own Alphabet stock., (© Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY. Published by BMJ.)

Objective: Several artificial intelligence (AI) systems for diabetic retinopathy screening have been validated but there is limited evidence on their performance in real-world settings. This study aimed to assess the performance of an AI software deployed within the diabetic retinopathy screening programme in Dominica., Methods and Analysis: We conducted a prospective, cross-sectional clinical validation study. Patients with diabetes aged 18 years and above attending the diabetic retinopathy screening in primary care facilities in Dominica from 5 June to 3 July 2021 were enrolled.Grading was done at the point of care by the field grader, followed by counselling and referral to the eye clinic. Images were then graded by an AI system. Sensitivity, specificity with 95% CIs and area under the curve (AUC) were calculated for comparing the AI to field grader as gold standard., Results: A total of 587 participants were screened. The AI had a sensitivity and specificity for detecting referable diabetic retinopathy of 77.5% and 91.5% compared with the grader, for all participants, including ungradable images. The AUC was 0.8455. Excluding 52 participants deemed ungradable by the grader, the AI had a sensitivity and specificity of 81.4% and 91.5%, with an AUC of 0.9648., Conclusion: This study provides evidence that AI has the potential to be deployed to assist a diabetic screening programme in a middle-income real-world setting and perform with reasonable accuracy compared with a specialist grader., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2023. Re-use permitted under CC BY. Published by BMJ.)

Background: Despite, the potential clinical utility of 60-4 visual fields, they are not frequently used in clinical practice partly, due to the purported impact of facial contour on field defects. The purpose of this study was to design and test an artificial intelligence-driven platform to predict facial structure-dependent visual field defects on 60-4 visual field tests., Methods: Subjects with no ocular pathology were included. Participants were subject to optical coherence tomography, 60-4 Swedish interactive thresholding algorithm visual field tests and photography. The predicted visual field was compared with observed 60-4 visual field results in subjects. Average and point-specific sensitivity, specificity, precision, negative predictive value, accuracy, and F1-scores were primary outcome measures., Results: 30 healthy were enrolled. Three-dimensional facial reconstruction using a convolution neural network (CNN) was able to predict facial contour-dependent 60-4 visual field defects in 30 subjects without ocular pathology. Overall model accuracy was 97%±3% and 96%±3% and the F1-score, dependent on precision and sensitivity, was 58%±19% and 55%±15% for the right eye and left eye, respectively. Spatial-dependent model performance was observed with increased sensitivity and precision within the far inferior nasal field reflected by an average F1-score of 76%±20% and 70%±29% for the right eye and left eye, respectively., Conclusions: This pilot study reports the development of a CNN-enhanced platform capable of predicting 60-4 visual field defects in healthy controls based on facial contour. Further study with this platform may enhance understanding of the influence of facial contour on 60-4 visual field testing., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.)

Objective: To develop and validate an explainable artificial intelligence (AI) model for detecting geographic atrophy (GA) via colour retinal photographs., Methods and Analysis: We conducted a prospective study where colour fundus images were collected from healthy individuals and patients with retinal diseases using an automated imaging system. All images were categorised into three classes: healthy, GA and other retinal diseases, by two experienced retinologists. Simultaneously, an explainable learning model using class activation mapping techniques categorised each image into one of the three classes. The AI system's performance was then compared with manual evaluations., Results: A total of 540 colour retinal photographs were collected. Data was divided such that 300 images from each class trained the AI model, 120 for validation and 120 for performance testing. In distinguishing between GA and healthy eyes, the model demonstrated a sensitivity of 100%, specificity of 97.5% and an overall diagnostic accuracy of 98.4%. Performance metrics like area under the receiver operating characteristic (AUC-ROC, 0.988) and the precision-recall (AUC-PR, 0.952) curves reinforced the model's robust achievement. When differentiating GA from other retinal conditions, the model preserved a diagnostic accuracy of 96.8%, a precision of 90.9% and a recall of 100%, leading to an F1-score of 0.952. The AUC-ROC and AUC-PR scores were 0.975 and 0.909, respectively., Conclusions: Our explainable AI model exhibits excellent performance in detecting GA using colour retinal images. With its high sensitivity, specificity and overall diagnostic accuracy, the AI model stands as a powerful tool for the automated diagnosis of GA., Competing Interests: Competing interests: VS: consultant for I-Care, outside the submitted work. DV: consultant for Bayer, Novartis, Roche, outside the submitted work. ADN, MM and GF: None. PL: consultant for Aerie, Allergan, Apellis, Bausch & Lomb, Bayer, Biogen, Boehringer Ingelheim, I-Care, Genentech, Novartis, Ocular Therapeutix, Outlook Therapeutics and Roche, outside the submitted work., (© Author(s) (or their employer(s)) 2023. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Objective: To investigate repeatability of refractive state using a smartphone-based assessment tool, the Near Eye Tool for Refractive Assessment (NETRA)., Methods and Analysis: This study included 279 participants, predominantly female (66.7%) of African descent (49.1%). The age range was 9-63 years with mean age ( s ) 22.6 (8.9) years. Two consecutive measurements per eye with the NETRA were measured for both eyes of all participants. However, analyses for the right eyes only are included here. Multivariate statistical analysis included stereo-pair comets and scatterplots with 95% surfaces of constant probability density. Correlation coefficients for repeated samples were determined. Repeatability and agreement for NETRA were assessed with Bland-Altman plots, coefficients of repeatability ([Formula: see text] ; [Formula: see text] is the SD of differences) and intraclass correlation coefficients (ICCs)., Results: Bland-Altman plots, within-subject SD ( s w ), coefficients of repeatability and ICC indicated that repeated measurements were similar for many but not all eyes and there was good agreement (ICC=0.96) for the spherical coefficient ( F I = M ) but less so for antistigmatic coefficients ( F J = J 0 and F K = J 45 ) of power. Although mean differences for repeated samples were almost zero, 95% limits of agreement widths were larger for the stigmatic coefficients. Without cycloplegia, repeatability (2.77 s w ) was 1.63 D, 0.58 D and 0.56 D for the stigmatic and antistigmatic coefficients, respectively., Conclusion: NETRA is a potentially useful and inexpensive portable method in clinical and primary health settings, and especially in less-developed regions of the world. The subjective nature of the self-refraction task can be challenging for younger individuals, and cycloplegia is recommended for NETRA with such patients., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2023. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Purpose: To explore the current research about the role of optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA) in dysthyroid optic neuropathy (DON)., Methods: Studies in the literature that focused on OCT, OCTA and DON were retrieved by searching PubMed, EMBASE, Cochrane databases and Clinical Trial before 20 June 2023. The methodological quality was assessed using the Newcastle-Ottawa scale. The quantitative calculation was performed using Review Manager V.5.3., Results: Twelve studies met the eligibility criteria and were included. DON group presented lower macular ganglion cell complex in the overall, superior and inferior hemifields compared with the non-DON group. Furthermore, the ganglion cell layer and inner plexiform layer in DON group was thinner in contrast to the non-DON group. The optic nerve head vessel density was lower in the DON group than that in the non-DON group. A reduction of radial peripapillary capillary vessel density could be seen in the DON group than the non-DON group in overall, inside disc, peripapillary, superior-hemifield, temporal and nasal. Besides, the macular superficial retinal capillary layer of non-DON and DON is lower than the healthy control group., Conclusions: This study supported the potential value of OCT and OCTA metrics as novel biomarkers of DON. Ophthalmologists should comprehensively consider the retinal structure and microvasculature in dealing with DON., Ethics and Dissemination: This systematic review included data from published literature and was exempt from ethics approval. Results would be disseminated through peer-reviewed publication and presented at academic conferences engaging clinicians., Prospero Registration Number: CRD42023414907., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2023. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Background/aims: Retinal capillary non-perfusion (NP) and neovascularisation (NV) are two of the most important angiographic changes in diabetic retinopathy (DR). This study investigated the feasibility of using deep learning (DL) models to automatically segment NP and NV on ultra-widefield fluorescein angiography (UWFA) images from patients with DR., Methods: Retrospective cross-sectional chart review study. In total, 951 UWFA images were collected from patients with severe non-proliferative DR (NPDR) or proliferative DR (PDR). Each image was segmented and labelled for NP, NV, disc, background and outside areas. Using the labelled images, DL models were trained and validated (80%) using convolutional neural networks (CNNs) for automated segmentation and tested (20%) on test sets. Accuracy of each model and each label were assessed., Results: The best accuracy from CNN models for each label was 0.8208, 0.8338, 0.9801, 0.9253 and 0.9766 for NP, NV, disc, background and outside areas, respectively. The best Intersection over Union for each label was 0.6806, 0.5675, 0.7107, 0.8551 and 0.924 and mean mean boundary F1 score (BF score) was 0.6702, 0.8742, 0.9092, 0.8103 and 0.9006, respectively., Conclusions: DL models can detect NV and NP as well as disc and outer margins on UWFA with good performance. This automated segmentation of important UWFA features will aid physicians in DR clinics and in overcoming grader subjectivity., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2023. No commercial re-use. See rights and permissions. Published by BMJ.)

Purpose: To describe an artificial intelligence platform that detects thyroid eye disease (TED)., Design: Development of a deep learning model., Methods: 1944 photographs from a clinical database were used to train a deep learning model. 344 additional images ('test set') were used to calculate performance metrics. Receiver operating characteristic, precision-recall curves and heatmaps were generated. From the test set, 50 images were randomly selected ('survey set') and used to compare model performance with ophthalmologist performance. 222 images obtained from a separate clinical database were used to assess model recall and to quantitate model performance with respect to disease stage and grade., Results: The model achieved test set accuracy of 89.2%, specificity 86.9%, recall 93.4%, precision 79.7% and an F1 score of 86.0%. Heatmaps demonstrated that the model identified pixels corresponding to clinical features of TED. On the survey set, the ensemble model achieved accuracy, specificity, recall, precision and F1 score of 86%, 84%, 89%, 77% and 82%, respectively. 27 ophthalmologists achieved mean performance of 75%, 82%, 63%, 72% and 66%, respectively. On the second test set, the model achieved recall of 91.9%, with higher recall for moderate to severe (98.2%, n=55) and active disease (98.3%, n=60), as compared with mild (86.8%, n=68) or stable disease (85.7%, n=63)., Conclusions: The deep learning classifier is a novel approach to identify TED and is a first step in the development of tools to improve diagnostic accuracy and lower barriers to specialist evaluation., Competing Interests: Competing interests: JK and DR are on the Tepezza Speakers’ Bureau for Horizon Pharma. JK is a consultant for Triage Inc. The authors otherwise have no conflicts of interests to report., (© Author(s) (or their employer(s)) 2023. No commercial re-use. See rights and permissions. Published by BMJ.)