Longitudinal evaluation of RTVue optical coherence tomography in patients with glaucoma and suspected glaucoma and stable visual fields / Avaliação longitudinal da tomografia de coerência óptica RTVue em pacientes glaucomatosos e suspeitos de glaucoma com campos visuais estáveis

ABSTRACT Purpose: To longitudinally compare isolated structural parameters obtained using RTVue optical coherence tomography in patients with glaucoma and suspected glaucoma with stable visual fields. Methods: All patients were required to have a reliable SITA Standard 24-2 Humphrey Visual Field test. Visual field stability was defined as having <5 points with p<5% and/or having no points with p<1% and/or p<0.05% in the glaucoma progression analysis comparison graph. Furthermore, the glaucoma assessment strategy was used in optical coherence tomography. Results: The study included 75 eyes from 75 patients, 43 of which had glaucoma and 32 had suspected glaucoma. The mean visual field intervals were 29.57 ± 9.65 months between the first and third tests. No visual field parameter variations (mean deviation, pattern standard deviation, and visual field index) and no retinal nerve fiber layer or optic disk parameter variations between the first and third tests were observed (p>0.05 for all), and no retinal nerve fiber layer parameter variations throughout the study were observed, except for optic disk parameters presenting with cup volume changes (p=0.004). However, ganglion complex cells presented a progressively decreased average ganglion cell complex parameter, with a variability of -0.98% ± 3.71% (p=0.04) between the first and third tests. By contrast, the global loss volume progressively increased throughout the study, with a variability of 14.71% ± 44.52% (p=0.04) between the first and third tests. The inferior ganglion cell complex parameter was significantly decreased between the first and third tests (p=0.02). Conclusion: The present findings suggest that patients with glaucoma or suspected glaucoma with stable visual fields may present structural ganglion complex cell progression as assessed using RTVue optical coherence tomography.

RESUMO Objetivo: Comparar longitudinalmente os parâmetros estruturais isolados obtidos através da tomografia de coerência óptica RTVue em pacientes glaucomatosos e suspeitos de glaucoma com campos visuais estáveis. Métodos: Todos os incluídos deveriam ter Campimetria Computadorizada Humphrey Sita Standard 24-2 confiáveis. A estabilidade campimétrica foi definida se apresentassem menos de cinco pontos com p<5% e/ou nenhum ponto com p<1% e/ou p<0,05% no gráfico de comparação do Glaucoma Progression Analysis. Para a tomografia de coerência óptica, foi utilizado a estratégia de avaliação para glaucoma. Resultados: Foram incluídos 75 olhos de 75 pacientes: 43 com glaucoma e 32 suspeitos. A média dos intervalos do campo visual entre o 1o e 3o exame, foi de 29,57 ± 9,65 meses. Não houve variação para os parâmetros do campo visual (desvio médio, desvio padrão e índice da função visual) entre o primeiro e o último exame (p>0,05 para todos). Não houve variação dos parâmetros da camada de fibras nervosas da retina ao longo do estudo, enquanto que para os parâmetros do disco óptico, apenas cup volume apresentou mudança (p=0,004). Em relação à camada de células ganglionares da retina, notou-se uma redução progressiva na espessura média da Ganglionar Complex Cells com uma variabilidade entre o primeiro e último exame de -0,98 ± 3,71% (p=0,04). Quanto ao Global loss volume, houve um aumento progressivo ao longo do estudo com uma variabilidade entre o primeiro e último exame de 14,71 ± 44,52% (p=0,04). O parâmetro inferior do Ganglionar Complex Cells também reduziu significativamente entre o 1o e 3o exames (p=0,02). Os demais parâmetros da tomografia de coerência óptica RTVue se mantiveram estáveis entre o 1o e 3o exames. Conclusão: Os presentes achados sugerem que pacientes glaucomatosos ou com suspeita de glaucoma e com campos visuais estáveis, podem apresentar progressão estrutural na camada de células ganglionares da retina avaliada por meio da tomografia de coerência óptica RTVue.

Extent of glaucomatous damage on the first presentation.

AIM: The aim is to determine the magnitude of glaucomatous damage in the asymptomatic subjects identified with primary glaucoma for the first time and thus to evaluate the significance and efficacy of screening measures for glaucoma. MATERIALS AND METHODS: An observational retrospective cohort study of 100 asymptomatic patients of age more than 40 years, diagnosed with and under treatment for primary glaucoma was performed. Patients were categorized into having early, moderate, and severe glaucoma, according to standard automated perimetry (SAP) mean deviation (MD) in the worse eye (<-6, -6 to -12 and >-12 dB, respectively). Risk factors were correlated with the severity of glaucoma at presentation and statistically analyzed. RESULTS: About 32%, 33%, and 35% of patients were found to have early, moderate, and severe stages of glaucoma with average MD of -3.51 ±1.53, -8.65 ±1.64, -17.15 ± 5.13 on SAP, respectively. The association of risk factors such as age (P = 0.006) and glaucoma awareness (P = 0.044) with the severity of glaucoma was statistically significant. There was no direct statistical correlation found between gender, history of diabetes mellitus, family history of glaucoma, intraocular pressure, central corneal thickness, the angle width, and the severity of glaucoma in our study. CONCLUSION: Majority of cases with primary glaucoma show no symptoms until advanced irreversible stages. Early screening and proper treatment are the only ways to halt its progression. In spite of available facilities, 68% of patients in our study were found to have moderate-to-severe stages of glaucoma. This indicates that our screening measures should reach the masses at the primary level, with a focus on awareness programs.

Glaucoma suspects referred by general ophthalmologists to a tertiary center in Brazil: outcomes of the glaucoma specialist assessment / Suspeitos de glaucoma encaminhados do oftalmologista geral para um centro terciário no Brasil: resultados da avaliação do especialista em glaucoma

ABSTRACT Purpose: To characterize patients with suspected glaucoma who were referred to the clinic for suspected glaucoma in a tertiary public hospital in southern Brazil and to evaluate differences in functional and structural damages between patients diagnosed with different types of glaucoma, those with normal eye examination results, and those who remained as glaucoma suspects. Methods: This is a cohort study of patients referred by general ophthalmologists to the clinic for suspected glaucoma at Hospital Nossa Senhora da Conceição, Porto Alegre, Brazil, between March 2016 and December 2018. The patients were followed up until they had undergone reliable examinations (eye examination, visual field screening, and optic coherence tomography for classification as normal and having a suspected glaucoma, glaucoma with an elevated intraocular pressure, normotensive glaucoma, or ocular hypertension. Results: A total of 135 patients were included in this study. Of the patients, 117 subjects completed all examinations and met the inclusion criteria. Most patients were normal (36.8%), followed by those with suspected glaucoma (25.64%), normal tension glaucoma (18.8%), glaucoma with elevated intraocular pressure (12%), and ocular hypertensive (6%). The main reason for referral was increased optic nerve head cupping. The patients with normal tension glaucoma were older than the other subjects on average (p=0.03). In addition, the normal tension glaucoma group had a significantly worse baseline visual field index and mean deviation of the visual field than the normal, glaucoma suspect, and ocular hypertensive groups. The circumpapillary retinal nerve fiber layer on OCT was thinner on average in the normal tension glaucoma group than in the normal and glaucoma suspect groups (p<0.002) but did not significantly differ between the glaucoma group with elevated intraocular pressure and the other groups. Conclusions: Patients with normal tension glaucoma tend to be diagnosed later because of their normal intraocular pressures; thus, the optic nerve cupping must be greater to raise the suspicion of glaucoma. In this study, we found that the patients with normal tension glaucoma had worse disease at the time of diagnosis.

RESUMO Objetivo: Caracterizar a população com suspeita de glaucoma encaminhada a um centro público terciário no sul do Brasil e avaliar diferenças no dano dos parâmetros funcionais e estruturais entre os pacientes diagnosticados com diferentes tipos de glaucoma e aqueles classificados como normais e aqueles mantidos como suspeitos de glaucoma. Métodos: Esta é uma coorte dos pacientes encaminhados para o setor de glaucoma suspeito do Hospital Nossa Senhora da Conceição, Porto Alegre - BR, no período de março de 2016 a dezembro de 2018. Os pacientes foram acompanhados até obterem exames confiáveis (exame oftalmológico completo, campimetria visual, tomografia de coerência óptica) para serem classificados como: normal, glaucoma suspeito, glaucoma com pressão intraocular elevada, glaucoma de pressão normal ou hipertenso ocular. Resultados: Um total de 135 pacientes foram incluídos neste estudo, sendo que destes, 117 pacientes completaram todos os exames e foram incluídos neste estudo. A maioria dos pacientes foi considerada normal (36,8%), seguido por glaucoma suspeito (25,64%), glaucoma de pressão normal (18,8%), glaucoma com pressão intraocular elevada (12%) e hipertensão ocular (6%). A principal razão para encaminhamento foi escavação do nervo óptico aumentada. Pacientes com glaucoma de pressão normal eram em média mais velhos que os demais (p=0,03). Esses também apresentavam índice de campo visual e desvio médio da campimetria visual piores que sujeitos normal, com suspeita de glaucoma e hipertensos oculares, e tinham a camada de fibra nervosa medida pela tomografia de coerência óptica mais fina que normais e suspeitos de glaucoma (p<0,002). Os pacientes com glaucoma de pressão elevada não diferiram significativamente dos outros grupos. Conclusão: Pacientes com glaucoma de pressão normal tendem a ser diagnosticados mais tardiamente devido ao fato da pressão intraocular não estar elevada, logo a escavação do disco óptico deve ser maior para gerar a suspeita de glaucoma. Neste estudo, paciente com glaucoma de pressão normal apresentaram doença mais avançada no momento do diagnóstico em comparação com os outros grupos.

Dependence of corneal hysteresis on non-central corneal thickness in healthy subjects.

OBJECTIVE: To evaluate the dependence of corneal hysteresis (CH) on non-central corneal thickness. METHODS: Cross-sectional study of 1561 eyes of 1561 healthy volunteers with IOP less than 21mmHg, open angles on gonioscopy and no prior eye surgeries or local or systemic diseases. Pentacam-Scheimpflug technology was employed to segment the cornea into 6 circular zones centered on the apex (zones 1-6) and to determine the mean corneal thickness of these areas. CH was measured with ORA. Univariate and multivariate linear regression models adjusted for age and sex were created to model the dependence of CH on corneal thickness in zones 1 to 6. RESULTS: In the univariate linear regression models, we found that CH was dependent on mean corneal thickness of zone 1 (B=0,004; R2=0.95%; P<0.001), zone 2 (B=0,004; R2=0.57%; P=0.002), zone 4 (B=0,005; R2=1.50%; P<0.001) and zone 6 (B=0,003; R2=0.92%; P<0.001). Similar results were obtained in the multivariate model (R2=3.46%; P<0.001). CONCLUSION: This study suggests a significant dependence of CH on non-central corneal thickness. The model of corneal thickness segmentation into circular zones centered on the corneal apex is able to explain 3.47% of the variation in CH measurements.

24-2 SITA Standard versus 24-2 SITA Faster in Perimetry-Naive Normal Subjects.

PURPOSE: To compare the Swedish Interactive Thresholding Algorithm (SITA) Standard (SS) and SITA Faster (SFR) strategies in normal individuals undergoing standard automated perimetry (SAP) for the first time. DESIGN: Randomized, comparative, observational case series. PARTICIPANTS: Seventy-four perimetry-naive healthy individuals. METHODS: All individuals underwent SAP 24-2 testing with the Humphrey Field Analyzer III (model 850 Zeiss) using the SS and SFR strategies. One eye of each individual was tested. Test order between strategies was randomized, and an interval of 15 minutes was allowed between the tests. MAIN OUTCOME MEASURES: The following variables were compared: test time, foveal threshold, false-positive errors, number of unreliable tests, mean deviation (MD), visual field index (VFI), pattern standard deviation (PSD), glaucoma hemifield test (GHT), and number of depressed points deviating at P < 5%, P < 2%, P < 1%, and P < 0.5% on the total and pattern deviation probability maps. Specificity of the SS and SFR strategies were compared using Anderson's criteria for abnormal visual fields. RESULTS: The SFR tests were 60.4% shorter in time compared with SS (P < 0.001) and were associated with a significantly lower PSD (1.75 ± 0.80 decibel [dB] vs. 2.15 ± 1.25 dB; P = 0.016). There were no significant differences regarding the MD, VFI, foveal threshold, GHT, and number of points depressed at P < 5%, P < 2%, P < 1%, and P < 0.5% on the total deviation and pattern deviation probability maps between SS and SFR. When all exams were analyzed and any of Anderson's criteria was applied, the specificity was 68% with SFR and 61% with SS (P = 0.250). The specificities observed with SFR and SS when only the first or second exams were analyzed were also similar (63% vs. 64% and 72% vs. 58%, respectively, P > 0.05). CONCLUSIONS: The SS and SFR were associated with similar specificities in perimetry-naive individuals. The SFR did not increase the number of depressed points in the total and pattern deviation probability maps. Ophthalmologists should be aware that both strategies are associated with disturbingly high false-positive rates in perimetry-naive individuals. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references.

In-Depth Evaluation of Saliency Maps for Interpreting Convolutional Neural Network Decisions in the Diagnosis of Glaucoma Based on Fundus Imaging.

Glaucoma, a leading cause of blindness, damages the optic nerve, making early diagnosis challenging due to no initial symptoms. Fundus eye images taken with a non-mydriatic retinograph help diagnose glaucoma by revealing structural changes, including the optic disc and cup. This research aims to thoroughly analyze saliency maps in interpreting convolutional neural network decisions for diagnosing glaucoma from fundus images. These maps highlight the most influential image regions guiding the network's decisions. Various network architectures were trained and tested on 739 optic nerve head images, with nine saliency methods used. Some other popular datasets were also used for further validation. The results reveal disparities among saliency maps, with some consensus between the folds corresponding to the same architecture. Concerning the significance of optic disc sectors, there is generally a lack of agreement with standard medical criteria. The background, nasal, and temporal sectors emerge as particularly influential for neural network decisions, showing a likelihood of being the most relevant ranging from 14.55% to 28.16% on average across all evaluated datasets. We can conclude that saliency maps are usually difficult to interpret and even the areas indicated as the most relevant can be very unintuitive. Therefore, its usefulness as an explanatory tool may be compromised, at least in problems such as the one addressed in this study, where the features defining the model prediction are generally not consistently reflected in relevant regions of the saliency maps, and they even cannot always be related to those used as medical standards.

Asynchronous feature regularization and cross-modal distillation for OCT based glaucoma diagnosis.

Glaucoma has become a major cause of vision loss. Early-stage diagnosis of glaucoma is critical for treatment planning to avoid irreversible vision damage. Meanwhile, interpreting the rapidly accumulated medical data from ophthalmic exams is cumbersome and resource-intensive. Therefore, automated methods are highly desired to assist ophthalmologists in achieving fast and accurate glaucoma diagnosis. Deep learning has achieved great successes in diagnosing glaucoma by analyzing data from different kinds of tests, such as peripapillary optical coherence tomography (OCT) and visual field (VF) testing. Nevertheless, applying these developed models to clinical practice is still challenging because of various limiting factors. OCT models present worse glaucoma diagnosis performances compared to those achieved by OCT&VF based models, whereas VF is time-consuming and highly variable, which can restrict the wide employment of OCT&VF models. To this end, we develop a novel deep learning framework that leverages the OCT&VF model to enhance the performance of the OCT model. To transfer the complementary knowledge from the structural and functional assessments to the OCT model, a cross-modal knowledge transfer method is designed by integrating a designed distillation loss and a proposed asynchronous feature regularization (AFR) module. We demonstrate the effectiveness of the proposed method for glaucoma diagnosis by utilizing a public OCT&VF dataset and evaluating it on an external OCT dataset. Our final model with only OCT inputs achieves the accuracy of 87.4% (3.1% absolute improvement) and AUC of 92.3%, which are on par with the OCT&VF joint model. Moreover, results on the external dataset sufficiently indicate the effectiveness and generalization capability of our model.

Comparison between the Recommendations of Glaucoma Specialists and OCT Report Specialists for Further Ophthalmic Evaluation in a Community-Based Screening Study.

PURPOSE: To compare glaucoma referral patterns between glaucoma and OCT report specialists and to determine what influence, if any, a designated OCT reading could have on a glaucoma specialist's judgments. DESIGN: Retrospective, exploratory study. SUBJECTS: We included 483 eyes (243 individuals) from high-risk New York City neighborhoods screened as part of a mobile van glaucoma screening study from July 2017 to October 2017. METHODS: All participants underwent comprehensive testing, including visual acuity, commercial OCT imaging, gonioscopy, intraocular pressure, frequency-doubling testing, and funduscopic assessment. Three glaucoma specialists independently evaluated all the collected data to determine whether a further glaucoma workup referral was recommended. Two OCT report specialists evaluated only the OCT image for each eye using the commercial report as well as a specialized, customized report. In phase II, the glaucoma specialists then re-evaluated a subset of these eyes, this time with an OCT report specialist's judgments made available. MAIN OUTCOME MEASURES: Comparison of glaucoma specialist referrals made by glaucoma specialists versus OCT report specialists. RESULTS: Intergrader agreement between glaucoma specialists was 60% (κ = 0.43) and between report specialists was 95% (κ = 0.77). There was an agreement between a single OCT report specialist and the consensus (2 of 3) of glaucoma specialists in 74% of eyes (κ= 0.32). Of the eyes studied, 25% were referred for further glaucoma evaluation by the glaucoma specialists alone and 1% were referred for further glaucoma workup by only the report specialist. With the addition of the report specialist's judgments, referral pattern changes varied by glaucoma specialist but overall agreement increased to 85% (κ = 0.53). CONCLUSIONS: There was a fair level of agreement regarding glaucoma referral recommendations between glaucoma specialists with access to comprehensive screening data and OCT report specialists with access to only OCT data. Overall agreement increased when the designated OCT evaluation was made available to the glaucoma specialists. These results may aid in the design of future large-scale glaucoma screening studies.

Risk of Missed Diagnosis of Primary Open-Angle Glaucoma by Eye Care Providers.

Purpose: To evaluate the efficacy of opportunistic case finding in glaucoma detection and to determine factors associated with failure of glaucoma detection by eye health providers. Methods: This study was conducted on 154 new definite primary open-angle glaucoma (POAG) patients presenting to our glaucoma clinic. A questionnaire was prepared to determine if these subjects had sought eye care up to 12 months before presentation. The type of eye care provider and the principal reason for the visit were probed. The primary outcome measure was the frequency of a correct glaucoma diagnosis in their index visit. The secondary outcomes were factors associated with missed POAG diagnosis. Results: The great majority of study subjects (132 cases, 85.7%) had sought at least one ocular examination within 1 year before presentation. Among these patients, 73 cases (55.3%) had remained undiagnosed after the examination. Among the probed variables, age, gender, visual acuity, visual field defects, intraocular pressure, cup/disc ratio, nerve fiber layer thickness of the worse eye at presentation, and family history of glaucoma were comparable between correctly diagnosed and missed POAGs. The only factors significantly associated with missed POAG diagnosis were lack of significant refractive errors and visiting an optometrist rather than an ophthalmologist. Conclusions: The efficacy of opportunistic case finding for POAG seems to be less than ideal in our settings. Lack of a significant refractive error and visiting an optometrist rather than an ophthalmologist were associated with a missed diagnosis of POAG. These observations reflect the need to adopt policies to improve glaucoma screening by eye care providers.

Artificial Intelligence to Aid Glaucoma Diagnosis and Monitoring: State of the Art and New Directions.

Recent developments in the use of artificial intelligence in the diagnosis and monitoring of glaucoma are discussed. To set the context and fix terminology, a brief historic overview of artificial intelligence is provided, along with some fundamentals of statistical modeling. Next, recent applications of artificial intelligence techniques in glaucoma diagnosis and the monitoring of glaucoma progression are reviewed, including the classification of visual field images and the detection of glaucomatous change in retinal nerve fiber layer thickness. Current challenges in the direct application of artificial intelligence to further our understating of this disease are also outlined. The article also discusses how the combined use of mathematical modeling and artificial intelligence may help to address these challenges, along with stronger communication between data scientists and clinicians.

Diagnosing glaucoma on imbalanced data with self-ensemble dual-curriculum learning.

Glaucoma diagnosis often suffers from two types of data imbalances: 1) class imbalance, i.e., the non-glaucoma majority cases occupy most of the data; 2) rare cases, i.e., few cases present the uncommon retinopathy e.g., bayoneting or physiologic cupping. This dual-imbalances make glaucoma diagnosis model easy to be dominated by the majority cases but cannot correctly classify the minority and/or rare ones. In this paper, we propose an adaptive re-balancing strategy in the feature space, Self-Ensemble Dual-Curriculum learning (SEDC), to improve the glaucoma diagnosis on imbalanced data by augmenting feature distribution with feature distilling and feature re-weighting. Firstly, the self-ensembling (SEL) is developed to reinforce the discriminative ability of feature representations for the minority class and rare cases by distilling the features learned from the abundant majority cases. Secondly, the dual-curriculum (DCL) is designed to adaptively re-weight the imbalanced data in the feature space to learn a balanced decision function for accurate glaucoma diagnosis. Benefiting from feature distilling and re-weighting, the proposed SEDC fairly represents fundus images, regardless of the majority or rare cases, by augmenting the feature distribution to obtains the optimal decision boundary for accurate glaucoma diagnosis on the imbalanced dataset. Experimental results on three challenging glaucoma datasets show that our SEDC successfully delivers accurate glaucoma diagnosis by the adaptive re-balancing strategy, with the average mean value of Accuracy 0.9712, Sensitivity 0.9520, Specificity 0.9816, AUC 0.9928, F2-score 0.9547. Ablation and comparison studies demonstrate that our method outperforms state-of-the-art methods and traditional re-balancing strategies. The experiment also shows that the adaptive re-balancing strategy proposed in our method provides a more effective training approach with optimal convergence performance. It endows our SEDC a great advantage to handle the disease diagnosis on imbalanced data distribution.

Personal Computer-Based Visual Field Testing as an Alternative to Standard Automated Perimetry.

Introduction Standard automated perimetry (SAP) is the gold standard of visual field assessment in patients with neuro-ophthalmic conditions. Glaucoma is a progressive optic neuropathy characterized by damage to the ganglion cell complex with corresponding visual field defects and intraocular pressure (IOP) being the only modifiable ocular risk factor. Recent advances in technology have paved the way for remote screening and monitoring of visual field defects with the aid of a computer or tablet-based software. One such personal computer (PC)-based software is 'Specvis', which has shown promising reliability as compared to SAP. The primary objective of this study was to compare Specvis and Humphrey Field Analyzer (HFA) visual field reports in the graphical domain while secondary objectives were to estimate the ease of use of Specvis in comparison to HFA and comparison of test duration between Specvis and HFA.  Materials and methods This was a cross-sectional validation study performed at a tertiary care ophthalmology institute in Rawalpindi, Pakistan. Subjects presenting to the outpatient department were recruited based on consecutive sampling technique and were divided into healthy and diseased groups. Basic data collection instrument after informed consent was filled with demographic data, ophthalmic data, disease condition, and attached with analysis reports of both HFA and Specvis for assessment by three senior ophthalmology consultants independently. A total of 218 eyes of 109 subjects were included in this pilot study. SAP was done on the VF 30-2 program using HFA 3. The same patient then performed the visual field assessment on a PC with Specvis installed and settings adjusted to match the VF 30-2 program of HFA as closely as possible. Visual fields of a subject obtained from HFA and Specvis were then coupled and sent to three different senior ophthalmologists. The assessment was done by comparing the greyscale visual field printouts in the graphical domain and scored based on a 5-point Likert scale which were then analyzed for inter-observer reliability. After each test, all subjects were asked to rate the difficulty level of performing the test on HFA and Specvis based on a 5-point Likert scale. The duration of the test performed on HFA and Specvis was also noted for comparison. Results We observed male preponderance in our study participants (n=128, 58.72%). The majority of the participants were non-diseased (n=170, 77.98%) while advanced glaucoma was the commonest disease in the diseased group (n = 22, 10.09%). The mean age of the participants was 40.71 (SD=15.24). The observations for the HFA test duration had an average of 213.33 seconds (SD=33.49, Min=174.00, Max=314.00) while the Specvis test duration had an average of 267.36 seconds (SD=35.98, Min=228.00, Max=370.00). A significant positive correlation was observed between score 1, score 2, and score 3 given by the three ophthalmologists. A significant negative correlation was observed between ease of using HFA and age, with a correlation of -.28. A significant negative correlation was also observed between ease of using Specvis and age. Conclusion Specvis, a computer-based free open-source software used in our study, can give promising results in diagnosing as well as monitoring the progression of visual field defects. It can act as a significantly cost-effective and readily available bridge between visual field examination by confrontation method and SAP.

Assessment of retinal microcirculation alterations in glaucoma by optical coherence tomography angiography / 中华实验眼科杂志

Microcirculation alterations of the human retina are of significant relevance with glaucoma.Optical coherence tomography angiography (OCTA) is a non-invasive examination that provides signals of the retina and retinal microcirculation.It is currently widely used in research and screening for glaucoma due to its effective detection of each layer in the retinal microcirculation.Vessel density of radial peripapillary capillary (RPC) is positively correlated with the retinal nerve fiber layer thickness, which is crucial in the assessment of glaucoma.Peripapillary region and macula are regions of interest in OCTA analysis for microcirculation.Vessel density of the whole retina and RPC in the peripapillary region decreases significantly, which matches the thinned retinal fiber layer thickness and visual field defects, and is relevant to the disease severity.As for the macular region, vessel density declines while the area of the foveal avascular zone increases.Greater changes in late-stage glaucoma and normal-tension glaucoma are detected by OCTA compared with other stages and types of glaucoma.OCTA imaging may be influenced by high myopia and intraocular pressure, and the peripapillary region is of greater diagnostic value than macula in microcirculation changes.In conclusion, OCTA can serve as a new technique for the assessment of retinal microcirculation in glaucoma.This review summarized the characteristics of retinal microcirculation in OCTA images and its change in peripapillary and macular region in glaucoma eyes.Influencing factors associated with peripapillary and macular microcirculation changes in OCTA images and evaluation of peripapillary and macular microcirculation in glaucoma by OCTA and their diagnostic values were reviewed.

Central Corneal Thickness Among Filipino Patients in an Ambulatory Eye Surgery Center Using Anterior Segment Optical Coherence Tomography.

OBJECTIVE: The purpose of the study was to determine the central corneal thickness (CCT) among Filipino patients that may contribute to different glaucoma diagnosis using the anterior segment optical coherence tomography in an ambulatory eye surgery center. METHODS: A single-center retrospective, cross-sectional study design including 1232 eyes of 641 patients of the Asian Eye Institute, Makati, Philippines from January 2019 to December 2019 who had their CCT measured with Visante anterior segment optical coherence tomography (AS-OCT). CCT was correlated with age, sex, presence of diabetes and/or hypertension, and glaucoma diagnosis. RESULTS: Among 641 patients who had their CCT measured by Visante AS-OCT, 723 eyes of 369 patients were included. Nearly half of the study population were normal or glaucoma suspects. The mean CCT among Filipino patients was 535.59 ± 34.06 µm. Ocular hypertensive patients had the thickest CCT, while normal tension glaucoma patients had the thinnest CCT. After adjusting for multiple variables, CCT had a direct relationship with the presence of diabetes, IOP level and the diagnosis of ocular hypertension, while inverse relationship with age. Most of the patients presenting with angle closure glaucoma were females aged 60 and above. CONCLUSION: Visante AS-OCT is a non-contact and non-aerosol generating instrument allaying the fear of disease transmission from contact or aerosolization of tears. Our study confirms similar relationships of CCT with age, presence of diabetes, IOP level, and diagnosis of ocular hypertension or normal tension glaucoma among Filipino patients with the available literature from other ethnicities.

Automatic glaucoma detection based on transfer induced attention network.

BACKGROUND: Glaucoma is one of the causes that leads to irreversible vision loss. Automatic glaucoma detection based on fundus images has been widely studied in recent years. However, existing methods mainly depend on a considerable amount of labeled data to train the model, which is a serious constraint for real-world glaucoma detection. METHODS: In this paper, we introduce a transfer learning technique that leverages the fundus feature learned from similar ophthalmic data to facilitate diagnosing glaucoma. Specifically, a Transfer Induced Attention Network (TIA-Net) for automatic glaucoma detection is proposed, which extracts the discriminative features that fully characterize the glaucoma-related deep patterns under limited supervision. By integrating the channel-wise attention and maximum mean discrepancy, our proposed method can achieve a smooth transition between general and specific features, thus enhancing the feature transferability. RESULTS: To delimit the boundary between general and specific features precisely, we first investigate how many layers should be transferred during training with the source dataset network. Next, we compare our proposed model to previously mentioned methods and analyze their performance. Finally, with the advantages of the model design, we provide a transparent and interpretable transferring visualization by highlighting the key specific features in each fundus image. We evaluate the effectiveness of TIA-Net on two real clinical datasets and achieve an accuracy of 85.7%/76.6%, sensitivity of 84.9%/75.3%, specificity of 86.9%/77.2%, and AUC of 0.929 and 0.835, far better than other state-of-the-art methods. CONCLUSION: Different from previous studies applied classic CNN models to transfer features from the non-medical dataset, we leverage knowledge from the similar ophthalmic dataset and propose an attention-based deep transfer learning model for the glaucoma diagnosis task. Extensive experiments on two real clinical datasets show that our TIA-Net outperforms other state-of-the-art methods, and meanwhile, it has certain medical value and significance for the early diagnosis of other medical tasks.

Diagnostic capability of different morphological parameters for primary open-angle glaucoma in the Chinese population.

BACKGROUND: To assess the diagnostic capability of novel Bruch's membrane opening (BMO)-based disc parameters, the BMO-minimum rim width (BMO-MRW) and the BMO-minimum rim area (BMO-MRA) in the Chinese population and compare them to the retinal nerve fiber layer (RNFL) from optical coherence tomography (OCT) and the rim area (RA) from the Heidelberg retinal tomograph-III (HRT-III). METHODS: In total, 200 eyes of 77 healthy and 123 primary open-angle glaucoma (POAG) subjects were included in this cross-sectional study. All participants underwent the visual field test and structural measurements by OCT and HRT-III. The areas under the receiver operating characteristic curves (AUCs) of different structural parameters were calculated to assess their diagnostic power and compared using the DeLong test. RESULTS: In populations with different characteristics, the BMO-MRW and BMO-MRA had better diagnostic power than the RA. In discriminating between all POAG subjects and healthy controls and between early-stage patients and controls, the global BMO-MRW had comparable AUCs with the RNFL, but the BMO-MRA had lower AUCs than the RNFL. In healthy subjects with macrodiscs, both the global and sectoral BMO-MRW were thinner than those in healthy subjects with normal disc size. The AUCs of BMO-MRA, BMO-MRW and RNFL in subjects with macrodiscs were comparable. Additionally, in the myopic population, the BMO-MRA and BMO-MRW had comparable AUCs with the RNFL. CONCLUSIONS: The BMO-MRW had comparable diagnostic power with the RNFL, and compared with BMO-MRW, the BMO-MRA might have advantages in certain populations, such as macrodiscs. All OCT-derived parameters exceeded the RA in diagnostic capability.

An enhanced deep image model for glaucoma diagnosis using feature-based detection in retinal fundus.

This paper proposes a deep image analysis-based model for glaucoma diagnosis that uses several features to detect the formation of glaucoma in retinal fundus. These features are combined with most extracted parameters like inferior, superior, nasal, and temporal region area, and cup-to-disc ratio that overall forms a deep image analysis. This proposed model is exercised to investigate the various aspects related to the prediction of glaucoma in retinal fundus images that help the ophthalmologist in making better decisions for the human eye. The proposed model is presented with the combination of four machine learning algorithms that provide the classification accuracy of 98.60% while other existing models like support vector machine (SVM), K-nearest neighbors (KNN), and Naïve Bayes provide individually with accuracies of 97.61%, 90.47%, and 95.23% respectively. These results clearly demonstrate that this proposed model offers the best methodology to an early diagnosis of glaucoma in retinal fundus.

Artificial intelligence (AI) impacting diagnosis of glaucoma and understanding the regulatory aspects of AI-based software as medical device.

Glaucoma, the group of eye diseases is characterized by increased intraocular pressure, optic neuropathy and visual field defect patterns. Early and correct diagnosis of glaucoma can prevent irreversible vision loss and glaucomatous structural damages to the eye. However, greater chances of misdiagnosis by the currently used conventional methods for diagnosis open up ways for more advanced techniques like the use of artificial intelligence (AI). Artificial intelligence coupled with optical coherence tomography imaging creates an algorithm that can be effectively used to make a model of complex data for detection as well as diagnosis of glaucoma. The present review is an attempt to provide state-of-the-art information on various AI techniques used in the diagnosis and assessment of glaucoma. The second part of the review is focused on understanding how the AI along with machine learning (ML) can be potentially used to be subjected for software as a medical device (SaMD) in precise diagnosis or early detection of disease conditions.