200 eyes of 110 male serving military personnel showing retinal vasculitis in various stages were managed with institution of steroids, ATT, photocoagulation, cryoablation and vitreo retinal surgery as indicated in a span of 78 months. 156 eyes showing evidence of vasoproliferation responded favourably to laser photocoagulation. Over all 80% of the patients showed good functional recovery with combined modalities of management thereby obviating recurrent morbidity and invalidation in trained combatant manpower in wage earning age group. Cases with complicated retinal vasculitis had to be treated in a sophisticated retina centre having facilities for Fluorescein angiography, laser and vitreo-retinal surgery.
[EN] Mutations in the ABCA4 gene are a common cause of Stargardt disease; however, other retinal phenotypes have also been associated with mutations in this gene. We describe an observational case report of an unusual clinical phenotype of Stargardt disease. The ophthalmological examination included best corrected visual acuity, color and autofluorescence photography, fluorescein angiography, optical coherence tomography, and electrophysiology tests. Targeted next-generation sequencing of 99 genes associated with inherited retinal dystrophies was performed in the index patient. A 48-year-old woman presented with a best corrected visual acuity of 20/25 and 20/20. Fundoscopy revealed perifoveal yellow flecked-like lesions. Fluorescein angiography and fundus autofluorescence findings were consistent with pattern dystrophy. Pattern electroretinogram demonstrated bilateral decrease of p50 values. Genetic testing identified two heterozygous missense mutations, c.428C>T, p.(Pro143Leu) and c.3113C>T, p.(Ala.1038Val), in the ABCA4 gene. Based on our results, we believe that these particular mutations in the ABCA4 gene could be associated with a specific disease phenotype characterized by funduscopic appearance similar to pattern dystrophy. A detailed characterization of the retinal phenotype in patients carrying specific mutations in ABCA4 is crucial to understand disease expression and ensure optimal clinical care for patients with inherited retinal dystrophies. ; [PT] Mutações no gene ABCA4 são causa comum da doença de Stargardt, mas outros fenótipos da retina também foram associados a mutações nesse gene. Apresentamos um relato de caso observacional de um fenótipo clínico incomum da doença de Stargardt. O exame oftalmológico incluiu a acuidade visual com melhor correção, fotografia em cores e com autofluorescência, angiofluoresceinografia, tomografia de coerência óptica e testes de eletrofisiologia. Na paciente em questão, realizou-se o sequenciamento de próxima geração de 99 genes associados a distrofias retinais hereditárias. Tratava-se de uma mulher de 48 anos com melhor acuidade visual corrigida de 20/25 e 20/20. A fundoscopia revelou lesões puntiformes amarelas perifoveais. Os resultados da angiofluoresceinografia e da autofluorescência do fundo de olho foram consistentes com distrofia em padrão. A eletrorretinografia por padrões mostrou diminuição bilateral dos valores de p50. Os testes genéticos revelaram duas mutações missense heterozigóticas, c.428C>T, p. (Pro143Leu) e c.3113C>T, p. (Ala.1038Val), no gene ABCA4. Nossos resultados nos fazem pensar que essas mutações específicas em ABCA4 talvez possam estar associadas a um fenótipo específico da doença, caracterizado por uma aparência fundoscópica semelhante à da distrofia em padrão. Uma caracterização detalhada do fenótipo da retina em pacientes portadores de mutações específicas em ABCA4 é crucial para compreender a expressão da doença e para garantir o tratamento clínico ideal para pacientes com distrofias retinais hereditárias. ; This work was supported by Instituto de Salud Carlos III (ISCIII), Spanish Ministry of Economy and Competitiveness and co-funded by European Union (ERDF, "A way to make Europe") [PI15-01648] and [PI1800612]; regional Ministry of Economy, Innovation, Science and Employment [CTS-1664] and regional Ministry of Health and Families [PEER-0501-2019] of the Autonomous Government of Andalusia; Foundation Isabel Gemio/Foundation Cajasol [FGEMIO-2019-01].
Financiado para publicación en acceso aberto: Universidade da Coruña/CISUG ; [Abstract] The detection of retinal microaneurysms is crucial for the early detection of important diseases such as diabetic retinopathy. However, the detection of these lesions in retinography, the most widely available retinal imaging modality, remains a very challenging task. This is mainly due to the tiny size and low contrast of the microaneurysms in the images. Consequently, the automated detection of microaneurysms usually relies on extensive ad-hoc processing. In this regard, although microaneurysms can be more easily detected using fluorescein angiography, this alternative imaging modality is invasive and not adequate for regular preventive screening. In this work, we propose a novel deep learning methodology that takes advantage of unlabeled multimodal image pairs for improving the detection of microaneurysms in retinography. In particular, we propose a novel adversarial multimodal pre-training consisting in the prediction of fluorescein angiography from retinography using generative adversarial networks. This pre-training allows learning about the retina and the microaneurysms without any manually annotated data. Additionally, we also propose to approach the microaneurysms detection as a heatmap regression, which allows an efficient detection and precise localization of multiple microaneurysms. To validate and analyze the proposed methodology, we perform an exhaustive experimentation on different public datasets. Additionally, we provide relevant comparisons against different state-of-the-art approaches. The results show a satisfactory performance of the proposal, achieving an Average Precision of 64.90%, 31.36%, and 33.55% in the E-Ophtha, ROC, and DDR public datasets. Overall, the proposed approach outperforms existing deep learning alternatives while providing a more straightforward detection method that can be effectively applied to raw unprocessed retinal images. ; This work is supported by Instituto de Salud Carlos III, Government of Spain, and the European Regional Development Fund (ERDF) of the European Union (EU) through the DTS18/00136 research project; Ministerio de Ciencia e Innovación, Government of Spain, through the RTI2018-095894-B-I00 and PID2019-108435RB-I00 research projects; Xunta de Galicia, Spain and the European Social Fund (ESF) of the EU through the predoctoral grant contract ref. ED481A-2017/328; Consellería de Cultura, Educación e Universidade, Xunta de Galicia, through Grupos de Referencia Competitiva, grant ref. ED431C 2020/24. CITIC, Centro de Investigación de Galicia ref. ED431G 2019/01, receives financial support from Consellería de Cultura, Educación e Universidade, Xunta de Galicia , through the ERDF (80%) and Secretaría Xeral de Universidades (20%). Funding for open access charge: Universidade da Coruña/CISUG ; Xunta de Galicia; ED481A-2017/328 ; Xunta de Galicia; ED431C 2020/24 ; Xunta de Galicia; ED431G 2019/01
[Abstract] Data scarcity represents an important constraint for the training of deep neural networks in medical imaging. Medical image labeling, especially if pixel-level annotations are required, is an expensive task that needs expert intervention and usually results in a reduced number of annotated samples. In contrast, extensive amounts of unlabeled data are produced in the daily clinical practice, including paired multimodal images from patients that were subjected to multiple imaging tests. This work proposes a novel self-supervised multimodal reconstruction task that takes advantage of this unlabeled multimodal data for learning about the domain without human supervision. Paired multimodal data is a rich source of clinical information that can be naturally exploited by trying to estimate one image modality from others. This multimodal reconstruction requires the recognition of domain-specific patterns that can be used to complement the training of image analysis tasks in the same domain for which annotated data is scarce. In this work, a set of experiments is performed using a multimodal setting of retinography and fluorescein angiography pairs that offer complementary information about the eye fundus. The evaluations performed on different public datasets, which include pathological and healthy data samples, demonstrate that a network trained for self-supervised multimodal reconstruction of angiography from retinography achieves unsupervised recognition of important retinal structures. These results indicate that the proposed self-supervised task provides relevant cues for image analysis tasks in the same domain. ; This work is supported by Instituto de Salud Carlos III, Government of Spain, and the European Regional Development Fund (ERDF) of the European Union (EU) through the DTS18/00136 research project, and by Ministerio de Economía, Industria y Competitividad, Government of Spain, through the DPI2015-69948-R research project. The authors of this work also receive financial support from the ERDF and Xunta de Galicia through Grupo de Referencia Competitiva, Ref. ED431C 2016-047, and from the European Social Fund (ESF) of the EU and Xunta de Galicia through the predoctoral grant contract Ref. ED481A-2017/328. CITIC, Centro de Investigación de Galicia Ref. ED431G 2019/01, receives financial support from Consellería de Educación, Universidade e Formación Profesional, Xunta de Galicia, through the ERDF (80%) and Secretaría Xeral de Universidades (20%) ; Xunta de Galicia; ED431C 2016-047 ; Xunta de Galicia; ED481A-2017/328 ; Xunta de Galicia; ED431G 2019/01
Retinal angiomatous proliferations (RAP) are one rare exudative maculopathy, occurring only in 10% of all cases of age-related macular degeneration (AMD). Objective: The aim of our survey was to describe a rare clinical case of RAP with bilateral affection and discuss the diagnostic methods and possible clinical therapies. Material and methods: We present a case of a 73-year-old patient with bilateral occurrence of RAP examined and followed at Eye Clinic of Military Medical Academy, Sofia. He underwent a complete ophthalmologic examination, including visual acuity, Amsler test and fluorescein angiography. Optical coherence tomography (OCT) was also done with the spectral OCT device RTVue OPTOVUE. The standard retinal programs – HD Line, 3D Macular Test and MM5 quantitative retinal assessment tests have been used. Results: The visual acuity of the patient on his first visit was 20/10 for the right eye, and 20/50 for the left eye, with drusen and significant edema in the posterior pole. On the fluorescein angiographic pictures, intraretinal proliferations were found in the left eye and retino-choroidal anastomoses in the right one. The ОСТ examination confirmed the diagnosis of RAP and enabled us to establish the stage of the disease. In the left eye, the intraretinal proliferations did not involve the subretinal space, so this was RAP I. While in the right eye connections between the retinal and choroidal vessels were seen, as well as serous detachment of the RPE – RAP stage III. We used anti-VEGF therapy in the right eye. Conclusion: Retinal angiomatous proliferations represent an exudative maculopathy, with specific FA and OCT characteristics, differing it from the typical age related macular degeneration. Precise diagnosis and stages of the disease are possible to establish only with the help of optical coherence tomography. RAP is an illness, which, due to its specific pathogenesis, clinical picture and bilateral affection, has to be carefully diagnosed, and monitoring of both eyes has to be taken in consideration.
Chenghui Zhang, Suyuan Wang, Mingxia Li, Yunhong Wu Department of Endocrinology and Metabolism, Hospital of Chengdu Office of People's Government of Tibetan Autonomous Region, Chengdu, People's Republic of ChinaCorrespondence: Yunhong WuDepartment of Endocrinology and Metabolism, Hospital of Chengdu Office of People's Government of Tibetan Autonomous Region, No. 20 Ximianqiao Street, Chengdu, Sichuan 610041, People's Republic of ChinaEmail Wu_yunhong@163.comAim: To explore the association between the atherosclerosis and diabetic retinopathy (DR) in Chinese patients with type 2 diabetes mellitus (T2DM).Methods: This hospital-based cross-sectional study included 949 patients (700 males and 249 females) with T2DM. The atherosclerotic parameters were assessed using the cardio-ankle vascular index (CAVI), ankle-brachial index (ABI), and carotid plaque. DR was assessed and graded using digital retinal photography and fundus fluorescein angiography as either nonproliferative DR (NPDR) or proliferative DR (PDR). Multiple logistic regression analysis was performed to identify the associations between the atherosclerotic parameters and DR status.Results: The prevalence of DR was 23.6% in total patients, including 167 (17.6%) patients with NPDR and 57 (6.0%) patients with PDR. Patients with NPDR and PDR were more likely to have higher prevalence of increased CAVI, increased ABI, and carotid plaque than those without DR. In multivariable adjusted logistic regression analysis, patients with NPDR showed an odds ratio (OR) of 2.59 [95% confidence interval (CI), 1.61– 4.19] for increased CAVI, 1.99 (0.62– 6.34) for increased ABI, and 1.75 (1.13– 2.71) for carotid plaque. Patients with PDR showed an OR of 7.83 (3.52– 17.41) for increased CAVI, 10.65 (3.33– 34.04) for increased ABI, and 11.40 (2.67– 48.63) for carotid plaque.Conclusion: Both NPDR and PDR were independently associated with increased CAVI and presence of carotid plaque in Chinese patients with T2DM.Keywords: atherosclerosis, cardio-ankle vascular index, diabetic retinopathy, ankle-brachial index, carotid plaque
Sadullah Keles,1 Orhan Ates,1 Baki Kartal,2 Hamit Hakan Alp,3 Metin Ekinci,4 Erdinc Ceylan,2 Osman Ondas,5 Eren Arpali,2 Semih Dogan,6 Kenan Yildirim,7 Mevlut Sait Keles8 1Department of Ophthalmology, School of Medicine, Ataturk University, Erzurum, Turkey; 2Department of Ophthalmology, Regional Training and Research Hospital, Erzurum, Turkey; 3Department of Biochemistry, School of Medicine, Yuzuncu Yil University, Van, Turkey; 4Department of Ophthalmology, School of Medicine, Kafkas University, Kars, Turkey; 5Department of Ophthalmology, Erbaa Government Hospital, Tokat, Turkey; 6Department of Ophthalmology, Kolan Hospital, Istanbul, Turkey; 7Department of Ophthalmology, Igdir Government Hospital, Igdir, Turkey; 8Department of Biochemistry, School of Medicine, Ataturk University, Erzurum, Turkey Aim: To evaluate levels of homocysteine, asymmetric dimethylarginine (ADMA), and nitric oxide (NO), as well as activity of endothelial NO synthase (eNOS), in patients with age-related macular degeneration (AMD).Methods: The levels of homocysteine, ADMA, and NO and activity of eNOS in patients who were diagnosed with wet AMD by fundus fluorescein angiography (n=30) were compared to a control group with no retinal pathology (n=30).Results: Levels of homocysteine and ADMA were found to be significantly higher in the wet AMD group than in the control group (P<0.001), whereas NO levels and eNOS activity were higher in the control group (P<0.001). In the wet AMD group, we detected a 2.64- and 0.33-fold increase in the levels of ADMA and homocysteine, respectively, and a 0.49- and 2.41-fold decrease in the eNOS activity and NO level, respectively.Conclusion: Elevated levels of homocysteine and ADMA were observed in patients with wet AMD. Increased ADMA may be responsible for the diminished eNOS activity found in these patients, which in turn contributes to the decrease in NO levels, which likely plays a role in the pathogenesis of AMD. Keywords: age-related macular degeneration, homocysteine, asymmetric dimethylarginine, nitric oxide, endothelial nitric oxide synthase activity
[Abstract] Deep learning is becoming the reference paradigm for approaching many computer vision problems. Nevertheless, the training of deep neural networks typically requires a significantly large amount of annotated data, which is not always available. A proven approach to alleviate the scarcity of annotated data is transfer learning. However, in practice, the use of this technique typically relies on the availability of additional annotations, either from the same or natural domain. We propose a novel alternative that allows to apply transfer learning from unlabelled data of the same domain, which consists in the use of a multimodal reconstruction task. A neural network trained to generate one image modality from another must learn relevant patterns from the images to successfully solve the task. These learned patterns can then be used to solve additional tasks in the same domain, reducing the necessity of a large amount of annotated data. In this work, we apply the described idea to the localization and segmentation of the most important anatomical structures of the eye fundus in retinography. The objective is to reduce the amount of annotated data that is required to solve the different tasks using deep neural networks. For that purpose, a neural network is pre-trained using the self-supervised multimodal reconstruction of fluorescein angiography from retinography. Then, the network is fine-tuned on the different target tasks performed on the retinography. The obtained results demonstrate that the proposed self-supervised transfer learning strategy leads to state-of-the-art performance in all the studied tasks with a significant reduction of the required annotations. ; This work is supported by Instituto de Salud Carlos III, Government of Spain, and the European Regional Development Fund (ERDF) of the European Union (EU) through the DTS18/00136 research project, and by Ministerio de Economía, Industria y Competitividad, Government of Spain, through the DPI2015-69948-R research project. The authors of this work also receive financial support from the ERDF and Xunta de Galicia (Spain) through Grupo de Referencia Competitiva, ref. ED431C 2016-047, and from the European Social Fund (ESF) of the EU and Xunta de Galicia (Spain) through the predoctoral grant contract ref. ED481A-2017/328. CITIC, Centro de Investigación de Galicia ref. ED431G 2019/01, receives financial support from Consellería de Educación, Universidade e Formación Profesional, Xunta de Galicia (Spain) , through the ERDF (80%) and Secretaría Xeral de Universidades (20%) ; Xunta de Galicia; ED431C 2016-047 ; Xunta de Galicia ; ED481A-2017/328 ; Xunta de Galicia; ED431G 2019/01
