Sankara Nethralaya

Importance: Deep learning is a family of computational methods that allow an algorithm to program itself by learning from a large set of examples that demonstrate the desired behavior, removing the need to specify rules explicitly. Application of these methods to medical imaging requires further assessment and validation. Objective: To apply deep learning to create an algorithm for automated detection of diabetic retinopathy and diabetic macular edema in retinal fundus photographs. Design and Setting: A specific type of neural network optimized for image classification called a deep convolutional neural network was trained using a retrospective development data set of 128 175 retinal images, which were graded 3 to 7 times for diabetic retinopathy, diabetic macular edema, and image gradability by a panel of 54 US licensed ophthalmologists and ophthalmology senior residents between May and December 2015. The resultant algorithm was validated in January and February 2016 using 2 separate data sets, both graded by at least 7 US board-certified ophthalmologists with high intragrader consistency. Exposure: Deep learning-trained algorithm. Main Outcomes and Measures: The sensitivity and specificity of the algorithm for detecting referable diabetic retinopathy (RDR), defined as moderate and worse diabetic retinopathy, referable diabetic macular edema, or both, were generated based on the reference standard of the majority decision of the ophthalmologist panel. The algorithm was evaluated at 2 operating points selected from the development set, one selected for high specificity and another for high sensitivity. Results: The EyePACS-1 data set consisted of 9963 images from 4997 patients (mean age, 54.4 years; 62.2% women; prevalence of RDR, 683/8878 fully gradable images [7.8%]); the Messidor-2 data set had 1748 images from 874 patients (mean age, 57.6 years; 42.6% women; prevalence of RDR, 254/1745 fully gradable images [14.6%]). For detecting RDR, the algorithm had an area under the receiver operating curve of 0.991 (95% CI, 0.988-0.993) for EyePACS-1 and 0.990 (95% CI, 0.986-0.995) for Messidor-2. Using the first operating cut point with high specificity, for EyePACS-1, the sensitivity was 90.3% (95% CI, 87.5%-92.7%) and the specificity was 98.1% (95% CI, 97.8%-98.5%). For Messidor-2, the sensitivity was 87.0% (95% CI, 81.1%-91.0%) and the specificity was 98.5% (95% CI, 97.7%-99.1%). Using a second operating point with high sensitivity in the development set, for EyePACS-1 the sensitivity was 97.5% and specificity was 93.4% and for Messidor-2 the sensitivity was 96.1% and specificity was 93.9%. Conclusions and Relevance: In this evaluation of retinal fundus photographs from adults with diabetes, an algorithm based on deep machine learning had high sensitivity and specificity for detecting referable diabetic retinopathy. Further research is necessary to determine the feasibility of applying this algorithm in the clinical setting and to determine whether use of the algorithm could lead to improved care and outcomes compared with current ophthalmologic assessment.

OBJECTIVE: To examine the global prevalence and major risk factors for diabetic retinopathy (DR) and vision-threatening diabetic retinopathy (VTDR) among people with diabetes. RESEARCH DESIGN AND METHODS: A pooled analysis using individual participant data from population-based studies around the world was performed. A systematic literature review was conducted to identify all population-based studies in general populations or individuals with diabetes who had ascertained DR from retinal photographs. Studies provided data for DR end points, including any DR, proliferative DR, diabetic macular edema, and VTDR, and also major systemic risk factors. Pooled prevalence estimates were directly age-standardized to the 2010 World Diabetes Population aged 20-79 years. RESULTS: A total of 35 studies (1980-2008) provided data from 22,896 individuals with diabetes. The overall prevalence was 34.6% (95% CI 34.5-34.8) for any DR, 6.96% (6.87-7.04) for proliferative DR, 6.81% (6.74-6.89) for diabetic macular edema, and 10.2% (10.1-10.3) for VTDR. All DR prevalence end points increased with diabetes duration, hemoglobin A(1c), and blood pressure levels and were higher in people with type 1 compared with type 2 diabetes. CONCLUSIONS: There are approximately 93 million people with DR, 17 million with proliferative DR, 21 million with diabetic macular edema, and 28 million with VTDR worldwide. Longer diabetes duration and poorer glycemic and blood pressure control are strongly associated with DR. These data highlight the substantial worldwide public health burden of DR and the importance of modifiable risk factors in its occurrence. This study is limited by data pooled from studies at different time points, with different methodologies and population characteristics.

BACKGROUND: Many causes of vision impairment can be prevented or treated. With an ageing global population, the demands for eye health services are increasing. We estimated the prevalence and relative contribution of avoidable causes of blindness and vision impairment globally from 1990 to 2020. We aimed to compare the results with the World Health Assembly Global Action Plan (WHA GAP) target of a 25% global reduction from 2010 to 2019 in avoidable vision impairment, defined as cataract and undercorrected refractive error. METHODS: We did a systematic review and meta-analysis of population-based surveys of eye disease from January, 1980, to October, 2018. We fitted hierarchical models to estimate prevalence (with 95% uncertainty intervals [UIs]) of moderate and severe vision impairment (MSVI; presenting visual acuity from <6/18 to 3/60) and blindness (<3/60 or less than 10° visual field around central fixation) by cause, age, region, and year. Because of data sparsity at younger ages, our analysis focused on adults aged 50 years and older. FINDINGS: Global crude prevalence of avoidable vision impairment and blindness in adults aged 50 years and older did not change between 2010 and 2019 (percentage change -0·2% [95% UI -1·5 to 1·0]; 2019 prevalence 9·58 cases per 1000 people [95% IU 8·51 to 10·8], 2010 prevalence 96·0 cases per 1000 people [86·0 to 107·0]). Age-standardised prevalence of avoidable blindness decreased by -15·4% [-16·8 to -14·3], while avoidable MSVI showed no change (0·5% [-0·8 to 1·6]). However, the number of cases increased for both avoidable blindness (10·8% [8·9 to 12·4]) and MSVI (31·5% [30·0 to 33·1]). The leading global causes of blindness in those aged 50 years and older in 2020 were cataract (15·2 million cases [9% IU 12·7-18·0]), followed by glaucoma (3·6 million cases [2·8-4·4]), undercorrected refractive error (2·3 million cases [1·8-2·8]), age-related macular degeneration (1·8 million cases [1·3-2·4]), and diabetic retinopathy (0·86 million cases [0·59-1·23]). Leading causes of MSVI were undercorrected refractive error (86·1 million cases [74·2-101·0]) and cataract (78·8 million cases [67·2-91·4]). INTERPRETATION: Results suggest eye care services contributed to the observed reduction of age-standardised rates of avoidable blindness but not of MSVI, and that the target in an ageing global population was not reached. FUNDING: Brien Holden Vision Institute, Fondation Théa, The Fred Hollows Foundation, Bill & Melinda Gates Foundation, Lions Clubs International Foundation, Sightsavers International, and University of Heidelberg.

BACKGROUND: To contribute to the WHO initiative, VISION 2020: The Right to Sight, an assessment of global vision impairment in 2020 and temporal change is needed. We aimed to extensively update estimates of global vision loss burden, presenting estimates for 2020, temporal change over three decades between 1990-2020, and forecasts for 2050. METHODS: We did a systematic review and meta-analysis of population-based surveys of eye disease from January, 1980, to October, 2018. Only studies with samples representative of the population and with clearly defined visual acuity testing protocols were included. We fitted hierarchical models to estimate 2020 prevalence (with 95% uncertainty intervals [UIs]) of mild vision impairment (presenting visual acuity ≥6/18 and <6/12), moderate and severe vision impairment (<6/18 to 3/60), and blindness (<3/60 or less than 10° visual field around central fixation); and vision impairment from uncorrected presbyopia (presenting near vision <N6 or <N8 at 40 cm where best-corrected distance visual acuity is ≥6/12). We forecast estimates of vision loss up to 2050. FINDINGS: In 2020, an estimated 43·3 million (95% UI 37·6-48·4) people were blind, of whom 23·9 million (55%; 20·8-26·8) were estimated to be female. We estimated 295 million (267-325) people to have moderate and severe vision impairment, of whom 163 million (55%; 147-179) were female; 258 million (233-285) to have mild vision impairment, of whom 142 million (55%; 128-157) were female; and 510 million (371-667) to have visual impairment from uncorrected presbyopia, of whom 280 million (55%; 205-365) were female. Globally, between 1990 and 2020, among adults aged 50 years or older, age-standardised prevalence of blindness decreased by 28·5% (-29·4 to -27·7) and prevalence of mild vision impairment decreased slightly (-0·3%, -0·8 to -0·2), whereas prevalence of moderate and severe vision impairment increased slightly (2·5%, 1·9 to 3·2; insufficient data were available to calculate this statistic for vision impairment from uncorrected presbyopia). In this period, the number of people who were blind increased by 50·6% (47·8 to 53·4) and the number with moderate and severe vision impairment increased by 91·7% (87·6 to 95·8). By 2050, we predict 61·0 million (52·9 to 69·3) people will be blind, 474 million (428 to 518) will have moderate and severe vision impairment, 360 million (322 to 400) will have mild vision impairment, and 866 million (629 to 1150) will have uncorrected presbyopia. INTERPRETATION: Age-adjusted prevalence of blindness has reduced over the past three decades, yet due to population growth, progress is not keeping pace with needs. We face enormous challenges in avoiding vision impairment as the global population grows and ages. FUNDING: Brien Holden Vision Institute, Fondation Thea, Fred Hollows Foundation, Bill & Melinda Gates Foundation, Lions Clubs International Foundation, Sightsavers International, and University of Heidelberg.

Artificial intelligence (AI) based on deep learning (DL) has sparked tremendous global interest in recent years. DL has been widely adopted in image recognition, speech recognition and natural language processing, but is only beginning to impact on healthcare. In ophthalmology, DL has been applied to fundus photographs, optical coherence tomography and visual fields, achieving robust classification performance in the detection of diabetic retinopathy and retinopathy of prematurity, the glaucoma-like disc, macular oedema and age-related macular degeneration. DL in ocular imaging may be used in conjunction with telemedicine as a possible solution to screen, diagnose and monitor major eye diseases for patients in primary care and community settings. Nonetheless, there are also potential challenges with DL application in ophthalmology, including clinical and technical challenges, explainability of the algorithm results, medicolegal issues, and physician and patient acceptance of the AI ‘black-box’ algorithms. DL could potentially revolutionise how ophthalmology is practised in the future. This review provides a summary of the state-of-the-art DL systems described for ophthalmic applications, potential challenges in clinical deployment and the path forward.

Purpose: COVID-19-associated rhino-orbital-cerebral mucormycosis (ROCM) has reached epidemic proportion during India's second wave of COVID-19 pandemic, with several risk factors being implicated in its pathogenesis. This study aimed to determine the patient demographics, risk factors including comorbidities, and medications used to treat COVID-19, presenting symptoms and signs, and the outcome of management. Methods: This was a retrospective, observational study of patients with COVID-19-associated ROCM managed or co-managed by ophthalmologists in India from January 1, 2020 to May 26, 2021. Results: Of the 2826 patients, the states of Gujarat (22%) and Maharashtra (21%) reported the highest number of ROCM. The mean age of patients was 51.9 years with a male preponderance (71%). While 57% of the patients needed oxygen support for COVID-19 infection, 87% of the patients were treated with corticosteroids, (21% for > 10 days). Diabetes mellitus (DM) was present in 78% of all patients. Most of the cases showed onset of symptoms of ROCM between day 10 and day 15 from the diagnosis of COVID-19, 56% developed within 14 days after COVID-19 diagnosis, while 44% had delayed onset beyond 14 days. Orbit was involved in 72% of patients, with stage 3c forming the bulk (27%). Overall treatment included intravenous amphotericin B in 73%, functional endoscopic sinus surgery (FESS)/paranasal sinus (PNS) debridement in 56%, orbital exenteration in 15%, and both FESS/PNS debridement and orbital exenteration in 17%. Intraorbital injection of amphotericin B was administered in 22%. At final follow-up, mortality was 14%. Disease stage >3b had poorer prognosis. Paranasal sinus debridement and orbital exenteration reduced the mortality rate from 52% to 39% in patients with stage 4 disease with intracranial extension (p < 0.05). Conclusion: : Corticosteroids and DM are the most important predisposing factors in the development of COVID-19-associated ROCM. COVID-19 patients must be followed up beyond recovery. Awareness of red flag symptoms and signs, high index of clinical suspicion, prompt diagnosis, and early initiation of treatment with amphotericin B, aggressive surgical debridement of the PNS, and orbital exenteration, where indicated, are essential for successful outcome.

Importance: Early diagnosis of retinoblastoma, the most common intraocular cancer, can save both a child's life and vision. However, anecdotal evidence suggests that many children across the world are diagnosed late. To our knowledge, the clinical presentation of retinoblastoma has never been assessed on a global scale. Objectives: To report the retinoblastoma stage at diagnosis in patients across the world during a single year, to investigate associations between clinical variables and national income level, and to investigate risk factors for advanced disease at diagnosis. Design, Setting, and Participants: A total of 278 retinoblastoma treatment centers were recruited from June 2017 through December 2018 to participate in a cross-sectional analysis of treatment-naive patients with retinoblastoma who were diagnosed in 2017. Main Outcomes and Measures: Age at presentation, proportion of familial history of retinoblastoma, and tumor stage and metastasis. Results: The cohort included 4351 new patients from 153 countries; the median age at diagnosis was 30.5 (interquartile range, 18.3-45.9) months, and 1976 patients (45.4%) were female. Most patients (n = 3685 [84.7%]) were from low- and middle-income countries (LMICs). Globally, the most common indication for referral was leukocoria (n = 2638 [62.8%]), followed by strabismus (n = 429 [10.2%]) and proptosis (n = 309 [7.4%]). Patients from high-income countries (HICs) were diagnosed at a median age of 14.1 months, with 656 of 666 (98.5%) patients having intraocular retinoblastoma and 2 (0.3%) having metastasis. Patients from low-income countries were diagnosed at a median age of 30.5 months, with 256 of 521 (49.1%) having extraocular retinoblastoma and 94 of 498 (18.9%) having metastasis. Lower national income level was associated with older presentation age, higher proportion of locally advanced disease and distant metastasis, and smaller proportion of familial history of retinoblastoma. Advanced disease at diagnosis was more common in LMICs even after adjusting for age (odds ratio for low-income countries vs upper-middle-income countries and HICs, 17.92 [95% CI, 12.94-24.80], and for lower-middle-income countries vs upper-middle-income countries and HICs, 5.74 [95% CI, 4.30-7.68]). Conclusions and Relevance: This study is estimated to have included more than half of all new retinoblastoma cases worldwide in 2017. Children from LMICs, where the main global retinoblastoma burden lies, presented at an older age with more advanced disease and demonstrated a smaller proportion of familial history of retinoblastoma, likely because many do not reach a childbearing age. Given that retinoblastoma is curable, these data are concerning and mandate intervention at national and international levels. Further studies are needed to investigate factors, other than age at presentation, that may be associated with advanced disease in LMICs.

BACKGROUND: Nonophthalmologist physicians do not confidently perform direct ophthalmoscopy. The use of artificial intelligence to detect papilledema and other optic-disk abnormalities from fundus photographs has not been well studied. METHODS: We trained, validated, and externally tested a deep-learning system to classify optic disks as being normal or having papilledema or other abnormalities from 15,846 retrospectively collected ocular fundus photographs that had been obtained with pharmacologic pupillary dilation and various digital cameras in persons from multiple ethnic populations. Of these photographs, 14,341 from 19 sites in 11 countries were used for training and validation, and 1505 photographs from 5 other sites were used for external testing. Performance at classifying the optic-disk appearance was evaluated by calculating the area under the receiver-operating-characteristic curve (AUC), sensitivity, and specificity, as compared with a reference standard of clinical diagnoses by neuro-ophthalmologists. RESULTS: The training and validation data sets from 6779 patients included 14,341 photographs: 9156 of normal disks, 2148 of disks with papilledema, and 3037 of disks with other abnormalities. The percentage classified as being normal ranged across sites from 9.8 to 100%; the percentage classified as having papilledema ranged across sites from zero to 59.5%. In the validation set, the system discriminated disks with papilledema from normal disks and disks with nonpapilledema abnormalities with an AUC of 0.99 (95% confidence interval [CI], 0.98 to 0.99) and normal from abnormal disks with an AUC of 0.99 (95% CI, 0.99 to 0.99). In the external-testing data set of 1505 photographs, the system had an AUC for the detection of papilledema of 0.96 (95% CI, 0.95 to 0.97), a sensitivity of 96.4% (95% CI, 93.9 to 98.3), and a specificity of 84.7% (95% CI, 82.3 to 87.1). CONCLUSIONS: A deep-learning system using fundus photographs with pharmacologically dilated pupils differentiated among optic disks with papilledema, normal disks, and disks with nonpapilledema abnormalities. (Funded by the Singapore National Medical Research Council and the SingHealth Duke-NUS Ophthalmology and Visual Sciences Academic Clinical Program.).

IMPORTANCE: More than 60 million people in India have diabetes and are at risk for diabetic retinopathy (DR), a vision-threatening disease. Automated interpretation of retinal fundus photographs can help support and scale a robust screening program to detect DR. OBJECTIVE: To prospectively validate the performance of an automated DR system across 2 sites in India. DESIGN, SETTING, AND PARTICIPANTS: This prospective observational study was conducted at 2 eye care centers in India (Aravind Eye Hospital and Sankara Nethralaya) and included 3049 patients with diabetes. Data collection and patient enrollment took place between April 2016 and July 2016 at Aravind and May 2016 and April 2017 at Sankara Nethralaya. The model was trained and fixed in March 2016. INTERVENTIONS: Automated DR grading system compared with manual grading by 1 trained grader and 1 retina specialist from each site. Adjudication by a panel of 3 retinal specialists served as the reference standard in the cases of disagreement. MAIN OUTCOMES AND MEASURES: Sensitivity and specificity for moderate or worse DR or referable diabetic macula edema. RESULTS: Of 3049 patients, 1091 (35.8%) were women and the mean (SD) age for patients at Aravind and Sankara Nethralaya was 56.6 (9.0) years and 56.0 (10.0) years, respectively. For moderate or worse DR, the sensitivity and specificity for manual grading by individual nonadjudicator graders ranged from 73.4% to 89.8% and from 83.5% to 98.7%, respectively. The automated DR system's performance was equal to or exceeded manual grading, with an 88.9% sensitivity (95% CI, 85.8-91.5), 92.2% specificity (95% CI, 90.3-93.8), and an area under the curve of 0.963 on the data set from Aravind Eye Hospital and 92.1% sensitivity (95% CI, 90.1-93.8), 95.2% specificity (95% CI, 94.2-96.1), and an area under the curve of 0.980 on the data set from Sankara Nethralaya. CONCLUSIONS AND RELEVANCE: This study shows that the automated DR system generalizes to this population of Indian patients in a prospective setting and demonstrates the feasibility of using an automated DR grading system to expand screening programs.

Diabetic retinopathy (DR) is the leading cause of preventable blindness worldwide. The risk of DR progression is highly variable among different individuals, making it difficult to predict risk and personalize screening intervals. We developed and validated a deep learning system (DeepDR Plus) to predict time to DR progression within 5 years solely from fundus images. First, we used 717,308 fundus images from 179,327 participants with diabetes to pretrain the system. Subsequently, we trained and validated the system with a multiethnic dataset comprising 118,868 images from 29,868 participants with diabetes. For predicting time to DR progression, the system achieved concordance indexes of 0.754-0.846 and integrated Brier scores of 0.153-0.241 for all times up to 5 years. Furthermore, we validated the system in real-world cohorts of participants with diabetes. The integration with clinical workflow could potentially extend the mean screening interval from 12 months to 31.97 months, and the percentage of participants recommended to be screened at 1-5 years was 30.62%, 20.00%, 19.63%, 11.85% and 17.89%, respectively, while delayed detection of progression to vision-threatening DR was 0.18%. Altogether, the DeepDR Plus system could predict individualized risk and time to DR progression over 5 years, potentially allowing personalized screening intervals.

BACKGROUND: Polypoidal choroidal vasculopathy (PCV) is an exudative maculopathy affecting vision, with clinical features distinct from neovascular age-related macular degeneration. Currently, no evidence-based guidelines exist for its diagnosis and treatment. METHODS: A panel of experts analyzed a systematic literature search on PCV together with results of the EVEREST trial, the only published randomized controlled clinical trial in PCV. At a subsequent Roundtable meeting, recommendations for the management of PCV were agreed based on this analysis and their own expert opinion. RESULTS: Diagnosis of PCV should be based on early-phase nodular hyperfluorescence from choroidal vasculature visualized using indocyanine green angiography. Recommended initial treatment of juxtafoveal and subfoveal PCV is either indocyanine green angiography-guided verteporfin photodynamic therapy or verteporfin photodynamic therapy plus 3 × 0.5 mg ranibizumab intravitreal injections 1 month apart. If there is incomplete regression of polyps by indocyanine green angiography, eyes should be retreated with verteporfin photodynamic therapy monotherapy or verteporfin photodynamic therapy plus ranibizumab. If there is complete regression of polyps by indocyanine green angiography, but there is leakage on fluorescein angiography and other clinical or anatomical signs of disease activity, eyes should be retreated with ranibizumab. CONCLUSION: Practical guidance on the clinical management of PCV is proposed based on expert evaluation of current evidence.

Digital eye strain (DES) is an entity encompassing visual and ocular symptoms arising due to the prolonged use of digital electronic devices. It is characterized by dry eyes, itching, foreign body sensation, watering, blurring of vision, and headache. Non-ocular symptoms associated with eye strain include stiff neck, general fatigue, headache, and backache. A variable prevalence ranging from 5 to 65% has been reported in the pre-COVID-19 era. With lockdown restrictions during the pandemic, outdoor activities were restricted for all age groups, and digital learning became the norm for almost 2 years. While the DES prevalence amongst children alone rose to 50–60%, the symptoms expanded to include recent onset esotropia and vergence abnormalities as part of the DES spectrum. New-onset myopia and increased progression of existing myopia became one of the most significant ocular health complications. Management options for DES include following correct ergonomics like reducing average daily screen time, frequent blinking, improving lighting, minimizing glare, taking regular breaks from the screen, changing focus to distance object intermittently, and following the 20-20-20 rule to reduce eye strain. Innovations in this field include high-resolution screens, inbuilt antireflective coating, matte-finished glass, edge-to-edge displays, and image smoothening graphic effects. Further explorations should focus on recommendations for digital screen optimization, novel spectacle lens technologies, and inbuilt filters to optimize visual comfort. A paradigm shift is required in our understanding of looking at DES from an etiological perspective, so that customized solutions can be explored accordingly. The aim of this review article is to understand the pathophysiology of varied manifestations, predisposing risk factors, varied management options, along with changing patterns of DES prevalence post COVID-19.

The endothelial (posterior) corneal dystrophies, which result from primary endothelial dysfunction, include Fuchs endothelial corneal dystrophy (FECD), posterior polymorphous corneal dystrophy (PPCD) and congenital hereditary endothelial dystrophy (CHED). Mutations in SLC4A11 gene have been recently identified in patients with recessive CHED (CHED2). In this study, we show that heterozygous mutations in the SLC4A11 gene also cause late-onset FECD. Four heterozygous mutations [three missense mutations (E399K, G709E and T754M) and one deletion mutation (c.99-100delTC)] absent in ethnically matched controls were identified in a screen of 89 FECD patients. Missense mutations involved amino acid residues showing high interspecies conservation, indicating that mutations at these sites would be deleterious. Accordingly, immunoblot analysis, biochemical assay of cell surface localization and confocal immunolocalization showed that missense proteins encoded by the mutants were defective in localization to the cell surface. Our data suggests that SLC4A11 haploinsufficiency and gradual accumulation of the aberrant misfolded protein may play a role in FECD pathology and that reduced levels of SLC4A11 influence the long-term viability of the neural crest derived corneal endothelial cells.

AIM: To compare ocular biometric values in a population based sample of eyes with occludable angles, angle closure glaucoma, and normal subjects. METHOD: 2850 subjects from a population based glaucoma prevalence study underwent complete ocular examination including indentation gonioscopy. Ocular biometry was performed in all subjects classified to have occludable angles (n = 143); angle closure glaucoma (n = 22), and a random subgroup of 419 normal subjects. Ocular biometry readings between the groups were compared and statistically analysed using "t," "z," and Mann-Whitney U tests. RESULTS: The mean age among subjects with occludable angles (54.43 (SD 9.53) years) and angle closure glaucoma (57.45 (8.5) years) was significantly higher (p<0.001) than normal subjects (49.95 (9.95) years). Axial length was shorter (p<0.001) in the occludable angle group (22.07 (0.69) mm) compared to the normal group (22.76 (0.78) mm). Anterior chamber depth (ACD) was shallower (p<0.001) among subjects with occludable angles (2.53 (0.26) mm) than normal subjects (3.00 (0.30) mm). Lens thickness (LT) was greater (p<0.001) in people with occludable angles (4.40 (0.53) mm) compared to normal subjects (4.31 (0.31) mm). No significant difference was noted in axial length, ACD (p = 0.451), and LT (p = 0.302) between angle closure glaucoma and occludable eyes. CONCLUSION: South Indian eyes with angle closure glaucoma and occludable angles seem to have significantly shorter axial lengths, shallower anterior chambers and greater lens thickness compared to the normal group.

Over the last decade the prevalence of glaucoma has been reported by the Vellore Eye Survey, Andhra Pradesh Eye Disease Study, Aravind Comprehensive Eye Survey, Chennai Glaucoma Study, and West Bengal Glaucoma Study. There have been some differences largely because of methodologic variations. We use the reported age and gender stratified prevalence estimates from these studies and the Indian population census estimates to calculate the number of persons with glaucoma or at risk of the disease in the country. On the basis of the available data, we estimate that there are approximately 11.2 million persons aged 40 years and older with glaucoma in India. Primary open angle glaucoma is estimated to affect 6.48 million persons. The estimated number with primary angle-closure glaucoma is 2.54 million. Those with any form of primary angle-closure disease could comprise 27.6 million persons. Most of those with disease are undetected and there exist major challenges in detecting and treating those with disease. In the light of the existing manpower and resource constraints, we evaluate options for improving case detection rates in the country.

Abstract Deep learning algorithms have been used to detect diabetic retinopathy (DR) with specialist-level accuracy. This study aims to validate one such algorithm on a large-scale clinical population, and compare the algorithm performance with that of human graders. A total of 25,326 gradable retinal images of patients with diabetes from the community-based, nationwide screening program of DR in Thailand were analyzed for DR severity and referable diabetic macular edema (DME). Grades adjudicated by a panel of international retinal specialists served as the reference standard. Relative to human graders, for detecting referable DR (moderate NPDR or worse), the deep learning algorithm had significantly higher sensitivity (0.97 vs. 0.74, p &lt; 0.001), and a slightly lower specificity (0.96 vs. 0.98, p &lt; 0.001). Higher sensitivity of the algorithm was also observed for each of the categories of severe or worse NPDR, PDR, and DME ( p &lt; 0.001 for all comparisons). The quadratic-weighted kappa for determination of DR severity levels by the algorithm and human graders was 0.85 and 0.78 respectively ( p &lt; 0.001 for the difference). Across different severity levels of DR for determining referable disease, deep learning significantly reduced the false negative rate (by 23%) at the cost of slightly higher false positive rates (2%). Deep learning algorithms may serve as a valuable tool for DR screening.

BACKGROUND: The survival of retinoblastoma in less-developed countries (LDCs) and the impact of socioeconomic variables on survival are not widely available in the literature. METHODS: A systematic review of publications from LDCs was performed. Articles were from multiple databases and written in seven languages. Results were correlated with socioeconomic indicators. Lower-income countries (LICs) and middle-income countries (MICs) were included in our analyses. RESULTS: An analysis of 164 publications including 14,800 patients from 48 LDCs was performed. Twenty-six per cent of the papers were written in languages other than English. Estimated survival in LICs was 40% (range, 23-70%); in lower MICs, 77% (range, 60-92%) and in upper MICs, 79% (range, 54-93%; p = 0.001).Significant differences were also found in the occurrence of metastasis: in LICs, 32% (range, 12-45); in lower MICs, 12% (range, 3-31) and in upper MICs, 9.5% (range, 3-24; p = 0.04). On multivariate analysis, physician density and human development index were significantly associated with survival and metastasis. Maternal mortality rate and per capita health expenditure were significantly associated with treatment refusal. CONCLUSIONS: Important information from LDCs is not always available in English or in major databases. Indicators of socioeconomic development and maternal and infant health were related with outcome.

BACKGROUND: Diabetic retinopathy is a leading cause of preventable blindness, especially in low-income and middle-income countries (LMICs). Deep-learning systems have the potential to enhance diabetic retinopathy screenings in these settings, yet prospective studies assessing their usability and performance are scarce. METHODS: We did a prospective interventional cohort study to evaluate the real-world performance and feasibility of deploying a deep-learning system into the health-care system of Thailand. Patients with diabetes and listed on the national diabetes registry, aged 18 years or older, able to have their fundus photograph taken for at least one eye, and due for screening as per the Thai Ministry of Public Health guidelines were eligible for inclusion. Eligible patients were screened with the deep-learning system at nine primary care sites under Thailand's national diabetic retinopathy screening programme. Patients with a previous diagnosis of diabetic macular oedema, severe non-proliferative diabetic retinopathy, or proliferative diabetic retinopathy; previous laser treatment of the retina or retinal surgery; other non-diabetic retinopathy eye disease requiring referral to an ophthalmologist; or inability to have fundus photograph taken of both eyes for any reason were excluded. Deep-learning system-based interpretations of patient fundus images and referral recommendations were provided in real time. As a safety mechanism, regional retina specialists over-read each image. Performance of the deep-learning system (accuracy, sensitivity, specificity, positive predictive value [PPV], and negative predictive value [NPV]) were measured against an adjudicated reference standard, provided by fellowship-trained retina specialists. This study is registered with the Thai national clinical trials registry, TCRT20190902002. FINDINGS: Between Dec 12, 2018, and March 29, 2020, 7940 patients were screened for inclusion. 7651 (96·3%) patients were eligible for study analysis, and 2412 (31·5%) patients were referred for diabetic retinopathy, diabetic macular oedema, ungradable images, or low visual acuity. For vision-threatening diabetic retinopathy, the deep-learning system had an accuracy of 94·7% (95% CI 93·0-96·2), sensitivity of 91·4% (87·1-95·0), and specificity of 95·4% (94·1-96·7). The retina specialist over-readers had an accuracy of 93·5 (91·7-95·0; p=0·17), a sensitivity of 84·8% (79·4-90·0; p=0·024), and specificity of 95·5% (94·1-96·7; p=0·98). The PPV for the deep-learning system was 79·2 (95% CI 73·8-84·3) compared with 75·6 (69·8-81·1) for the over-readers. The NPV for the deep-learning system was 95·5 (92·8-97·9) compared with 92·4 (89·3-95·5) for the over-readers. INTERPRETATION: A deep-learning system can deliver real-time diabetic retinopathy detection capability similar to retina specialists in community-based screening settings. Socioenvironmental factors and workflows must be taken into consideration when implementing a deep-learning system within a large-scale screening programme in LMICs. FUNDING: Google and Rajavithi Hospital, Bangkok, Thailand. TRANSLATION: For the Thai translation of the abstract see Supplementary Materials section.

Primary diabetes care and diabetic retinopathy (DR) screening persist as major public health challenges due to a shortage of trained primary care physicians (PCPs), particularly in low-resource settings. Here, to bridge the gaps, we developed an integrated image-language system (DeepDR-LLM), combining a large language model (LLM module) and image-based deep learning (DeepDR-Transformer), to provide individualized diabetes management recommendations to PCPs. In a retrospective evaluation, the LLM module demonstrated comparable performance to PCPs and endocrinology residents when tested in English and outperformed PCPs and had comparable performance to endocrinology residents in Chinese. For identifying referable DR, the average PCP's accuracy was 81.0% unassisted and 92.3% assisted by DeepDR-Transformer. Furthermore, we performed a single-center real-world prospective study, deploying DeepDR-LLM. We compared diabetes management adherence of patients under the unassisted PCP arm (n = 397) with those under the PCP+DeepDR-LLM arm (n = 372). Patients with newly diagnosed diabetes in the PCP+DeepDR-LLM arm showed better self-management behaviors throughout follow-up (P < 0.05). For patients with referral DR, those in the PCP+DeepDR-LLM arm were more likely to adhere to DR referrals (P < 0.01). Additionally, DeepDR-LLM deployment improved the quality and empathy level of management recommendations. Given its multifaceted performance, DeepDR-LLM holds promise as a digital solution for enhancing primary diabetes care and DR screening.

INTRODUCTION: Diabetes mellitus, particularly type II, is a major public health concern worldwide. While the occurrence of diabetic retinopathy cannot be prevented, with the provision of knowledge to sufferers, sight-threatening complications can be minimized. PURPOSE: To report the results of a KAP (Knowledge, Attitude and Practice) study among a rural population in two areas: diabetes mellitus (DM) and diabetic retinopathy (DR). The level of knowledge was evaluated for both DM and DR; however, the influence of knowledge on practices and attitude was evaluated in only the DR group. METHODS: In rural areas, 145 awareness meetings on DM and DR were conducted attended 28 347 individuals. Using systematic random sampling, the data were collected from every 14th individual. In total, 1938 individuals from a rural population were numbered for gaining their responses to the KAP questionnaire. Univariate and multiple regression analyses were performed to identify independent risk factors related to the knowledge of the disease and influence of this knowledge on attitude and practice. RESULTS: Of 1938 individuals, 966 (49.9%) had knowledge of DM and 718 (37.1%) had knowledge of DR. Knowledge about DM was more in women (OR=1.93; 95% CI: 1.55-2.39), in subjects who followed the Christian faith (OR=1.48; 95% CI: 1.07-2.04) and in those who belonged to the upper socioeconomic strata (OR=2.60; 95% CI: 1.84-3.67). The knowledge of DR was significantly higher among subjects who spoke the Malayalam language (OR=3.80; 95% CI: 2.03-7.13), who followed the Christian faith (OR=1.73; 95% CI: 1.27-2.35), and in those who belonged to the upper socioeconomic strata (OR=1.85; 95% CI: 1.32-2.58). Compared with those who had no knowledge of DR (n = 1220), significant percentages of individuals with knowledge (n = 718) had the right attitude - to go for regular eye examinations - (65.9% vs 93.3%) (p<0.0001) ). Regarding practice patterns, only 36.5% of individuals with knowledge about DR believed that if they controlled their blood sugar, they could avoid a visit to an ophthalmologist, compared with 55.5% with no knowledge (p<0.0001). CONCLUSION: The results suggest that we need to propagate aggressive and comprehensive awareness models to educate rural populations on DM and DR.