Development and comparison of automated classifiers for glaucoma diagnosis using Stratus optical coherence tomography

Diagnosing glaucoma in primary eye care and the role of Artificial Intelligence applications for reducing the prevalence of undetected glaucoma in Australia

Glaucoma is the commonest cause of irreversible blindness worldwide, with over 70% of people affected remaining undiagnosed. Early detection is crucial for halting progressive visual impairment in glaucoma patients, as there is no cure available. This narrative review aims to: identify reasons for the significant under-diagnosis of glaucoma globally, particularly in Australia, elucidate the role of primary healthcare in glaucoma diagnosis using Australian healthcare as an example, and discuss how recent advances in artificial intelligence (AI) can be implemented to improve diagnostic outcomes. Glaucoma is a prevalent disease in ageing populations and can have improved visual outcomes through appropriate treatment, making it essential for general medical practice. In countries such as Australia, New Zealand, Canada, USA, and the UK, optometrists serve as the gatekeepers for primary eye care, and glaucoma detection often falls on their shoulders. However, there is significant variation in the capacity for glaucoma diagnosis among eye professionals. Automation with Artificial Intelligence (AI) analysis of optic nerve photos can help optometrists identify high-risk changes and mitigate the challenges of image interpretation rapidly and consistently. Despite its potential, there are significant barriers and challenges to address before AI can be deployed in primary healthcare settings, including external validation, high quality real-world implementation, protection of privacy and cybersecurity, and medico-legal implications. Overall, the incorporation of AI technology in primary healthcare has the potential to reduce the global prevalence of undiagnosed glaucoma cases by improving diagnostic accuracy and efficiency.

The utilization of artificial intelligence in glaucoma: diagnosis versus screening

With advancements in the implementation of artificial intelligence (AI) in different ophthalmology disciplines, it continues to have a significant impact on glaucoma diagnosis and screening. This article explores the distinct roles of AI in specialized ophthalmology clinics and general practice, highlighting the critical balance between sensitivity and specificity in diagnostic and screening models. Screening models prioritize sensitivity to detect potential glaucoma cases efficiently, while diagnostic models emphasize specificity to confirm disease with high accuracy. AI applications, primarily using machine learning (ML) and deep learning (DL), have been successful in detecting glaucomatous optic neuropathy from colored fundus photographs and other retinal imaging modalities. Diagnostic models integrate data extracted from various forms of modalities (including tests that assess structural optic nerve damage as well as those evaluating functional damage) to provide a more nuanced, accurate and thorough approach to diagnosing glaucoma. As AI continues to evolve, the collaboration between technology and clinical expertise should focus more on improving specificity of glaucoma diagnostic models to assess ophthalmologists to revolutionize glaucoma diagnosis and improve patients care.

Advancement in Understanding Glaucoma: A Comprehensive Review

Glaucoma, a silent thief of sight, remains a significant cause of irreversible blindness due to a substantial number of undiagnosed and untreated cases. To combat this insidious disease effectively, a multifaceted approach is imperative. Early detection is paramount in the battle against glaucoma. Patient history, including family history, plays a pivotal role in identifying those at risk. A comprehensive understanding of a patient's genetic predisposition can significantly enhance the accuracy of diagnosis and detection of suspicious cases. Treatment options include prescription eye drops, oral medicines, laser treatment, surgery, or a combination of approaches. Trabeculectomy involves the surgical creation of an aqueous humor drainage channel, while laser trabeculoplasty enhances aqueous outflow by modifying the trabecular meshwork. However, these procedures pose certain risks and complications. Exploration of alternative treatments with lower risks is underway. These innovative approaches hold promise in reducing the burdens associated with conventional treatments such as trabeculectomy. However, the effectiveness of these alternatives in the long term remains a subject of ongoing research. Neuroprotective drugs have also been in development to halt the progression of glaucoma. However, their success remains uncertain due to challenges, such as a lack of understanding of the underlying mechanisms, scarcity of suitable drugs, and regulatory hurdles in gaining approval. In essence, the overarching goal of glaucoma therapy is to reduce intraocular pressure through various means - medications, laser procedures, or innovative methods. The aim is to slow down the disease's progression, thereby preserving vision and improving the patient's quality of life. In conclusion, addressing the challenge of glaucoma requires a comprehensive approach encompassing early detection, innovative treatments, and ongoing research into potential cures. Only through concerted efforts can we hope to reduce the impact of this sight-stealing disease on individuals and society as a whole.

Characteristics of a large, labeled dataset for the training of artificial intelligence for glaucoma screening with fundus photographs

Purpose

Significant visual impairment due to glaucoma is largely caused by the disease being detected too late.

Objective

To build a labeled data set for training artificial intelligence (AI) algorithms for glaucoma screening by fundus photography, to assess the accuracy of the graders, and to characterize the features of all eyes with referable glaucoma (RG).

Design

Cross-sectional study.

Subjects

Color fundus photographs (CFPs) of 113 893 eyes of 60 357 individuals were obtained from EyePACS, California, United States, from a population screening program for diabetic retinopathy.

Methods

Carefully selected graders (ophthalmologists and optometrists) graded the images. To qualify, they had to pass the European Optic Disc Assessment Trial optic disc assessment with ≥ 85% accuracy and 92% specificity. Of 90 candidates, 30 passed. Each image of the EyePACS set was then scored by varying random pairs of graders as "RG," "no referable glaucoma (NRG)," or "ungradable (UG)." In case of disagreement, a glaucoma specialist made the final grading. Referable glaucoma was scored if visual field damage was expected. In case of RG, graders were instructed to mark up to 10 relevant glaucomatous features.

Main Outcome Measures

Qualitative features in eyes with RG.

Results

The performance of each grader was monitored; if the sensitivity and specificity dropped below 80% and 95%, respectively (the final grade served as reference), they exited the study and their gradings were redone by other graders. In all, 20 graders qualified; their mean sensitivity and specificity (standard deviation [SD]) were 85.6% (5.7) and 96.1% (2.8), respectively. The 2 graders agreed in 92.45% of the images (Gwet's AC2, expressing the inter-rater reliability, was 0.917). Of all gradings, the sensitivity and specificity (95% confidence interval) were 86.0 (85.2-86.7)% and 96.4 (96.3-96.5)%, respectively. Of all gradable eyes (n = 111 183; 97.62%) the prevalence of RG was 4.38%. The most common features of RG were the appearance of the neuroretinal rim (NRR) inferiorly and superiorly.

Conclusions

A large data set of CFPs was put together of sufficient quality to develop AI screening solutions for glaucoma. The most common features of RG were the appearance of the NRR inferiorly and superiorly. Disc hemorrhages were a rare feature of RG.

Financial Disclosures

Proprietary or commercial disclosure may be found after the references.

Diseased thyroid tissue classification in OCT images using deep learning: Towards surgical decision support

Intraoperative guidance tools for thyroid surgery based on optical coherence tomography (OCT) could aid distinguish between normal and diseased tissue. However, OCT images are difficult to interpret, thus, real-time automatic analysis could support the clinical decision-making. In this study, several deep learning models were investigated for thyroid disease classification on 2D and 3D OCT data obtained from ex vivo specimens of 22 patients undergoing surgery and diagnosed with several thyroid pathologies. Additionally, two open-access datasets were used to evaluate the custom models. On the thyroid dataset, the best performance was achieved by the 3D vision transformer model with a Matthew's correlation coefficient (MCC) of 0.79 (accuracy = 0.90) for the normal-versus-abnormal classification. On the open-access datasets, the custom models achieved the best performance (MCC > 0.88, accuracy > 0.96). Results obtained for the normal-versus-abnormal classification suggest OCT, complemented with deep learning-based analysis, as a tool for real-time automatic diseased tissue identification in thyroid surgery.

Performances of Machine Learning in Detecting Glaucoma Using Fundus and Retinal Optical Coherence Tomography Images: A Meta-Analysis

PURPOSE

To evaluate the performance of machine learning (ML) in detecting glaucoma using fundus and retinal optical coherence tomography (OCT) images.

DESIGN

Meta-analysis.

METHODS

PubMed and EMBASE were searched on August 11, 2021. A bivariate random-effects model was used to pool ML's diagnostic sensitivity, specificity, and area under the curve (AUC). Subgroup analyses were performed based on ML classifier categories and dataset types.

RESULTS

One hundred and five studies (3.3%) were retrieved. Seventy-three (69.5%), 30 (28.6%), and 2 (1.9%) studies tested ML using fundus, OCT, and both image types, respectively. Total testing data numbers were 197,174 for fundus and 16,039 for OCT. Overall, ML showed excellent performances for both fundus (pooled sensitivity = 0.92 [95% CI, 0.91-0.93]; specificity = 0.93 [95% CI, 0.91-0.94]; and AUC = 0.97 [95% CI, 0.95-0.98]) and OCT (pooled sensitivity = 0.90 [95% CI, 0.86-0.92]; specificity = 0.91 [95% CI, 0.89-0.92]; and AUC = 0.96 [95% CI, 0.93-0.97]). ML performed similarly using all data and external data for fundus and the external test result of OCT was less robust (AUC = 0.87). When comparing different classifier categories, although support vector machine showed the highest performance (pooled sensitivity, specificity, and AUC ranges, 0.92-0.96, 0.95-0.97, and 0.96-0.99, respectively), results by neural network and others were still good (pooled sensitivity, specificity, and AUC ranges, 0.88-0.93, 0.90-0.93, 0.95-0.97, respectively). When analyzed based on dataset types, ML demonstrated consistent performances on clinical datasets (fundus AUC = 0.98 [95% CI, 0.97-0.99] and OCT AUC = 0.95 [95% 0.93-0.97]).

CONCLUSIONS

Performance of ML in detecting glaucoma compares favorably to that of experts and is promising for clinical application. Future prospective studies are needed to better evaluate its real-world utility.

Glaucoma Detection Using Support Vector Machine Method Based on Spectralis OCT

Spectralis optical coherence tomography (OCT) provided more detailed parameters in the peripapillary and macular areas among the OCT machines, but it is not easy to understand the enormous information (114 features) generated from Spectralis OCT in glaucoma assessment. Machine learning methodology has been well-applied in glaucoma detection in recent years and has the ability to process a large amount of information at once. Here we aimed to analyze the diagnostic capability of Spectralis OCT parameters on glaucoma detection using Support Vector Machine (SVM) classification method in our population. Our results showed that applying all OCT features with the SVM method had good capability in the detection of glaucomatous eyes (area under curve (AUC) = 0.82), as well as discriminating normal eyes from early, moderate, or severe glaucomatous eyes (AUC = 0.78, 0.89, and 0.93, respectively). Apart from using all OCT features, the minimum rim width (MRW) may be good feature groups to discriminate early glaucomatous from normal eyes (AUC = 0.78). The combination of peripapillary and macular parameters, including MRW_temporal inferior (TI), MRW_global (G), ganglion cell layer (GCL)_outer temporal (T2), GCL_inner inferior (I1), peripapillary nerve fiber layer thickness (ppNFLT)_temporal superior (TS), and GCL_inner temporal (T1), provided better results (AUC = 0.84). This study showed promise in glaucoma management in the Taiwanese population. However, further validation study is needed to test the performance of our proposed model in the real world.

[Artificial intelligence and glaucoma: A literature review]

In recent years, research in artificial intelligence (AI) has experienced an unprecedented surge in the field of ophthalmology, in particular glaucoma. The diagnosis and follow-up of glaucoma is complex and relies on a body of clinical evidence and ancillary tests. This large amount of information from structural and functional testing of the optic nerve and macula makes glaucoma a particularly appropriate field for the application of AI. In this paper, we will review work using AI in the field of glaucoma, whether for screening, diagnosis or detection of progression. Many AI strategies have shown promising results for glaucoma detection using fundus photography, optical coherence tomography, or automated perimetry. The combination of these imaging modalities increases the performance of AI algorithms, with results comparable to those of humans. We will discuss potential applications as well as obstacles and limitations to the deployment and validation of such models. While there is no doubt that AI has the potential to revolutionize glaucoma management and screening, research in the coming years will need to address unavoidable questions regarding the clinical significance of such results and the explicability of the predictions.

Comparison of Different Machine Learning Classifiers for Glaucoma Diagnosis Based on Spectralis OCT

Early detection is important in glaucoma management. By using optical coherence tomography (OCT), the subtle structural changes caused by glaucoma can be detected. Though OCT provided abundant parameters for comprehensive information, clinicians may be confused once the results conflict. Machine learning classifiers (MLCs) are good tools for considering numerous parameters and generating reliable diagnoses in glaucoma practice. Here we aim to compare different MLCs based on Spectralis OCT parameters, including circumpapillary retinal nerve fiber layer (cRNFL) thickness, Bruch’s membrane opening-minimum rim width (BMO-MRW), Early Treatment Diabetes Retinopathy Study (ETDRS) macular thickness, and posterior pole asymmetry analysis (PPAA), in discriminating normal from glaucomatous eyes. Five MLCs were proposed, namely conditional inference trees (CIT), logistic model tree (LMT), C5.0 decision tree, random forest (RF), and extreme gradient boosting (XGBoost). Logistic regression (LGR) was used as a benchmark for comparison. RF was shown to be the best model. Ganglion cell layer measurements were the most important predictors in early glaucoma detection and cRNFL measurements were more important as the glaucoma severity increased. The global, temporal, inferior, superotemporal, and inferotemporal sites were relatively influential locations among all parameters. Clinicians should cautiously integrate the Spectralis OCT results into the entire clinical picture when diagnosing glaucoma.

Optical Coherence Tomography and Glaucoma

Early detection and monitoring are critical to the diagnosis and management of glaucoma, a progressive optic neuropathy that causes irreversible blindness. Optical coherence tomography (OCT) has become a commonly utilized imaging modality that aids in the detection and monitoring of structural glaucomatous damage. Since its inception in 1991, OCT has progressed through multiple iterations, from time-domain OCT, to spectral-domain OCT, to swept-source OCT, all of which have progressively improved the resolution and speed of scans. Even newer technological advancements and OCT applications, such as adaptive optics, visible-light OCT, and OCT-angiography, have enriched the use of OCT in the evaluation of glaucoma. This article reviews current commercial and state-of-the-art OCT technologies and analytic techniques in the context of their utility for glaucoma diagnosis and management, as well as promising future directions.

A Deep Learning Model and a Dataset for Diagnosing Ophthalmology Diseases

The rapid development of artificial neural network techniques, especially convolutional neural networks, encouraged the researchers to adapt such techniques in the medical domain. Specifically, to provide assist tools to help the professionals in patients’ diagnosis. The main problem faced by the researchers in the medical domain is the lack of available annotated datasets which can be used to train and evaluate large and complex deep neural networks. In this paper, to assist researchers who are interested in applying deep learning techniques to aid the ophthalmologists in diagnosing eye-related diseases, we provide an optical coherence tomography dataset with collaboration with ophthalmologists from the King Abdullah University Hospital, Irbid, Jordan. This dataset consists of 21,991 OCT images distributed over seven eye diseases in addition to normal images (no disease), namely, Choroidal Neovascularisation, Full Macular Hole (Full Thickness), Partial Macular Hole, Central Serous Retinopathy, Geographic atrophy, Macular Retinal Oedema, and Vitreomacular Traction. To the best of our knowledge, this dataset is the largest of its kind, where images belong to actual patients from Jordan and the annotation was carried out by ophthalmologists. Two classification tasks were applied to this dataset; a binary classification to distinguish between images which belong to healthy eyes (normal) and images which belong to diseased eyes (abnormal). The second classification task is a multi-class classification, where the deep neural network is trained to distinguish between the seven diseases listed above in addition to the normal case. In both classification tasks, the U-Net neural network was modified and subsequently utilised. This modification adds an additional block of layers to the original U-Net model to become capable of handling classification as the original network is used for image segmentation. The results of the binary classification were equal to 84.90% and 69.50% as accuracy and quadratic weighted kappa, respectively. The results of the multi-class classification, by contrast, were equal to 63.68% and 66.06% as accuracy and quadratic weighted kappa, respectively.

Visual Field Inference From Optical Coherence Tomography Using Deep Learning Algorithms: A Comparison Between Devices

Purpose

To develop a deep learning model to estimate the visual field (VF) from spectral-domain optical coherence tomography (SD-OCT) and swept-source OCT (SS-OCT) and to compare the performance between them.

Methods

Two deep learning models based on Inception-ResNet-v2 were trained to estimate 24-2 VF from SS-OCT and SD-OCT images. The estimation performance of the two models was evaluated by using the root mean square error between the actual and estimated VF. The performance was also compared among different glaucoma severities, Garway-Heath sectorizations, and central/peripheral regions.

Results

The training dataset comprised images of 4391 eyes from 2350 subjects, and the test dataset was obtained from another 243 subjects (243 eyes). In all subjects, the global estimation errors were 5.29 ± 2.68 dB (SD-OCT) and 4.51 ± 2.54 dB (SS-OCT), and the estimation error of SS-OCT was significantly lower than that of SD-OCT (P < 0.001). In the analysis of sectors, SS-OCT showed better performance in all sectors except for the inferonasal sector in normal vision and early glaucoma. In advanced glaucoma, the estimation error of the central region was worsened in both OCTs, but SS-OCT was still significantly better in the peripheral region.

Conclusions

Our deep learning model estimated the VF 24-2 better with a wide field image of SS-OCT than did with retinal nerve fiber layer and ganglion cell–inner plexiform layer images of SD-OCT.

Translational Relevance

This deep learning method can help clinicians to determine the VF from OCT images. OCT manufacturers can equip this system to provide additional VF data.

Artificial Intelligence Algorithms to Diagnose Glaucoma and Detect Glaucoma Progression: Translation to Clinical Practice

Purpose

This concise review aims to explore the potential for the clinical implementation of artificial intelligence (AI) strategies for detecting glaucoma and monitoring glaucoma progression.

Methods

Nonsystematic literature review using the search combinations “Artificial Intelligence,” “Deep Learning,” “Machine Learning,” “Neural Networks,” “Bayesian Networks,” “Glaucoma Diagnosis,” and “Glaucoma Progression.” Information on sensitivity and specificity regarding glaucoma diagnosis and progression analysis as well as methodological details were extracted.

Results

Numerous AI strategies provide promising levels of specificity and sensitivity for structural (e.g. optical coherence tomography [OCT] imaging, fundus photography) and functional (visual field [VF] testing) test modalities used for the detection of glaucoma. Area under receiver operating curve (AROC) values of > 0.90 were achieved with every modality. Combining structural and functional inputs has been shown to even more improve the diagnostic ability. Regarding glaucoma progression, AI strategies can detect progression earlier than conventional methods or potentially from one single VF test.

Conclusions

AI algorithms applied to fundus photographs for screening purposes may provide good results using a simple and widely accessible test. However, for patients who are likely to have glaucoma more sophisticated methods should be used including data from OCT and perimetry. Outputs may serve as an adjunct to assist clinical decision making, whereas also enhancing the efficiency, productivity, and quality of the delivery of glaucoma care. Patients with diagnosed glaucoma may benefit from future algorithms to evaluate their risk of progression. Challenges are yet to be overcome, including the external validity of AI strategies, a move from a “black box” toward “explainable AI,” and likely regulatory hurdles. However, it is clear that AI can enhance the role of specialist clinicians and will inevitably shape the future of the delivery of glaucoma care to the next generation.

Translational Relevance

The promising levels of diagnostic accuracy reported by AI strategies across the modalities used in clinical practice for glaucoma detection can pave the way for the development of reliable models appropriate for their translation into clinical practice. Future incorporation of AI into healthcare models may help address the current limitations of access and timely management of patients with glaucoma across the world.

Dual-input convolutional neural network for glaucoma diagnosis using spectral-domain optical coherence tomography

BACKGROUND/AIMS

To evaluate, with spectral-domain optical coherence tomography (SD-OCT), the glaucoma-diagnostic ability of a deep-learning classifier.

METHODS

A total of 777 Cirrus high-definition SD-OCT image sets of the retinal nerve fibre layer (RNFL) and ganglion cell-inner plexiform layer (GCIPL) of 315 normal subjects, 219 patients with early-stage primary open-angle glaucoma (POAG) and 243 patients with moderate-to-severe-stage POAG were aggregated. The image sets were divided into a training data set (252 normal, 174 early POAG and 195 moderate-to-severe POAG) and a test data set (63 normal, 45 early POAG and 48 moderate-to-severe POAG). The visual geometry group (VGG16)-based dual-input convolutional neural network (DICNN) was adopted for the glaucoma diagnoses. Unlike other networks, the DICNN structure takes two images (both RNFL and GCIPL) as inputs. The glaucoma-diagnostic ability was computed according to both accuracy and area under the receiver operating characteristic curve (AUC).

RESULTS

For the test data set, DICNN could distinguish between patients with glaucoma and normal subjects accurately (accuracy=92.793%, AUC=0.957 (95% CI 0.943 to 0.966), sensitivity=0.896 (95% CI 0.896 to 0.917), specificity=0.952 (95% CI 0.921 to 0.952)). For distinguishing between patients with early-stage glaucoma and normal subjects, DICNN's diagnostic ability (accuracy=85.185%, AUC=0.869 (95% CI 0.825 to 0.879), sensitivity=0.921 (95% CI 0.813 to 0.905), specificity=0.756 (95% CI 0.610 to 0.790)]) was higher than convolutional neural network algorithms that trained with RNFL or GCIPL separately.

CONCLUSION

The deep-learning algorithm using SD-OCT can distinguish normal subjects not only from established patients with glaucoma but also from patients with early-stage glaucoma. The deep-learning model with DICNN, as trained by both RNFL and GCIPL thickness map data, showed a high diagnostic ability for discriminatingpatients with early-stage glaucoma from normal subjects.

Artificial intelligence and deep learning in glaucoma: Current state and future prospects

Over the past few years, there has been an unprecedented and tremendous excitement for artificial intelligence (AI) research in the field of Ophthalmology; this has naturally been translated to glaucoma-a progressive optic neuropathy characterized by retinal ganglion cell axon loss and associated visual field defects. In this review, we aim to discuss how AI may have a unique opportunity to tackle the many challenges faced in the glaucoma clinic. This is because glaucoma remains poorly understood with difficulties in providing early diagnosis and prognosis accurately and in a timely fashion. In the short term, AI could also become a game changer by paving the way for the first cost-effective glaucoma screening campaigns. While there are undeniable technical and clinical challenges ahead, and more so than for other ophthalmic disorders whereby AI is already booming, we strongly believe that glaucoma specialists should embrace AI as a companion to their practice. Finally, this review will also remind ourselves that glaucoma is a complex group of disorders with a multitude of physiological manifestations that cannot yet be observed clinically. AI in glaucoma is here to stay, but it will not be the only tool to solve glaucoma.

The potential application of artificial intelligence for diagnosis and management of glaucoma in adults

BACKGROUND

Glaucoma is the most frequent cause of irreversible blindness worldwide. There is no cure, but early detection and treatment can slow the progression and prevent loss of vision. It has been suggested that artificial intelligence (AI) has potential application for detection and management of glaucoma.

SOURCES OF DATA

This literature review is based on articles published in peer-reviewed journals.

AREAS OF AGREEMENT

There have been significant advances in both AI and imaging techniques that are able to identify the early signs of glaucomatous damage. Machine and deep learning algorithms show capabilities equivalent to human experts, if not superior.

AREAS OF CONTROVERSY

Concerns that the increased reliance on AI may lead to deskilling of clinicians.

GROWING POINTS

AI has potential to be used in virtual review clinics, telemedicine and as a training tool for junior doctors. Unsupervised AI techniques offer the potential of uncovering currently unrecognized patterns of disease. If this promise is fulfilled, AI may then be of use in challenging cases or where a second opinion is desirable.

AREAS TIMELY FOR DEVELOPING RESEARCH

There is a need to determine the external validity of deep learning algorithms and to better understand how the 'black box' paradigm reaches results.

Glaucoma management in the era of artificial intelligence

Glaucoma is a result of irreversible damage to the retinal ganglion cells. While an early intervention could minimise the risk of vision loss in glaucoma, its asymptomatic nature makes it difficult to diagnose until a late stage. The diagnosis of glaucoma is a complicated and expensive effort that is heavily dependent on the experience and expertise of a clinician. The application of artificial intelligence (AI) algorithms in ophthalmology has improved our understanding of many retinal, macular, choroidal and corneal pathologies. With the advent of deep learning, a number of tools for the classification, segmentation and enhancement of ocular images have been developed. Over the years, several AI techniques have been proposed to help detect glaucoma by analysis of functional and/or structural evaluations of the eye. Moreover, the use of AI has also been explored to improve the reliability of ascribing disease prognosis. This review summarises the role of AI in the diagnosis and prognosis of glaucoma, discusses the advantages and challenges of using AI systems in clinics and predicts likely areas of future progress.

[Deep learning and neuronal networks in ophthalmology : Applications in the field of optical coherence tomography]

Deep learning is increasingly becoming the focus of various imaging methods in medicine. Due to the large number of different imaging modalities, ophthalmology is particularly suitable for this field of application. This article gives a general overview on the topic of deep learning and its current applications in the field of optical coherence tomography. For the benefit of the reader it focuses on the clinical rather than the technical aspects.

Data Driven Approach for Eye Disease Classification with Machine Learning

Medical health systems have been concentrating on artificial intelligence techniques for speedy diagnosis. However, the recording of health data in a standard form still requires attention so that machine learning can be more accurate and reliable by considering multiple features. The aim of this study is to develop a general framework for recording diagnostic data in an international standard format to facilitate prediction of disease diagnosis based on symptoms using machine learning algorithms. Efforts were made to ensure error-free data entry by developing a user-friendly interface. Furthermore, multiple machine learning algorithms including Decision Tree, Random Forest, Naive Bayes and Neural Network algorithms were used to analyze patient data based on multiple features, including age, illness history and clinical observations. This data was formatted according to structured hierarchies designed by medical experts, whereas diagnosis was made as per the ICD-10 coding developed by the American Academy of Ophthalmology. Furthermore, the system is designed to evolve through self-learning by adding new classifications for both diagnosis and symptoms. The classification results from tree-based methods demonstrated that the proposed framework performs satisfactorily, given a sufficient amount of data. Owing to a structured data arrangement, the random forest and decision tree algorithms’ prediction rate is more than 90% as compared to more complex methods such as neural networks and the naïve Bayes algorithm.

A feature agnostic approach for glaucoma detection in OCT volumes

Optical coherence tomography (OCT) based measurements of retinal layer thickness, such as the retinal nerve fibre layer (RNFL) and the ganglion cell with inner plexiform layer (GCIPL) are commonly employed for the diagnosis and monitoring of glaucoma. Previously, machine learning techniques have relied on segmentation-based imaging features such as the peripapillary RNFL thickness and the cup-to-disc ratio. Here, we propose a deep learning technique that classifies eyes as healthy or glaucomatous directly from raw, unsegmented OCT volumes of the optic nerve head (ONH) using a 3D Convolutional Neural Network (CNN). We compared the accuracy of this technique with various feature-based machine learning algorithms and demonstrated the superiority of the proposed deep learning based method. Logistic regression was found to be the best performing classical machine learning technique with an AUC of 0.89. In direct comparison, the deep learning approach achieved a substantially higher AUC of 0.94 with the additional advantage of providing insight into which regions of an OCT volume are important for glaucoma detection. Computing Class Activation Maps (CAM), we found that the CNN identified neuroretinal rim and optic disc cupping as well as the lamina cribrosa (LC) and its surrounding areas as the regions significantly associated with the glaucoma classification. These regions anatomically correspond to the well established and commonly used clinical markers for glaucoma diagnosis such as increased cup volume, cup diameter, and neuroretinal rim thinning at the superior and inferior segments.

[Can Big Data change our practices?]

The European Medicines Agency has defined Big Data by the "3 V's": Volume, Velocity and Variety. These large databases allow access to real life data on patient care. They are particularly suited for studies of adverse events and pharmacoepidemiology. Deep learning is a collection of algorithms used in machine learning, used to model high-level abstractions in data using model architectures, which are composed of multiple nonlinear transformations. This article shows how Big Data and Deep Learning can help in ophthalmology, pointing out their advantages and disadvantages. A literature review is presented in this article illustrating the uses of Deep Learning in ophthalmology.

Artificial intelligence in glaucoma

PURPOSE OF REVIEW

The use of computers has become increasingly relevant to medical decision-making, and artificial intelligence methods have recently demonstrated significant advances in medicine. We therefore provide an overview of current artificial intelligence methods and their applications, to help the practicing ophthalmologist understand their potential impact on glaucoma care.

RECENT FINDINGS

Techniques used in artificial intelligence can successfully analyze and categorize data from visual fields, optic nerve structure [e.g., optical coherence tomography (OCT) and fundus photography], ocular biomechanical properties, and a combination thereof to identify disease severity, determine disease progression, and/or recommend referral for specialized care. Algorithms have become increasingly complex in recent years, utilizing both supervised and unsupervised methods of artificial intelligence. Impressive performance of these algorithms on previously unseen data has been reported, often outperforming standard global indices and expert observers. However, there remains no clearly defined gold standard for determining the presence and severity of glaucoma, which undermines the training of these algorithms. To improve upon existing methodologies, future work must employ more robust definitions of disease, optimize data inputs for artificial intelligence analysis, and improve methods of extracting knowledge from learned results.

SUMMARY

Artificial intelligence has the potential to revolutionize the screening, diagnosis, and classification of glaucoma, both through the automated processing of large data sets, and by earlier detection of new disease patterns. In addition, artificial intelligence holds promise for fundamentally changing research aimed at understanding the development, progression, and treatment of glaucoma, by identifying novel risk factors and by evaluating the importance of existing ones.

Nerve Fiber Flux Analysis Using Wide-Field Swept-Source Optical Coherence Tomography

PURPOSE

To devise a method to quantify nerve fibers over their arcuate courses over an extended peripapillary area using optical coherence tomography (OCT).

METHODS

Participants were imaged with 8 × 8-mm volumetric OCT scans centered at the optic disc. A new quantity, nerve fiber flux (NFF), represents the cross-sectional area transected perpendicular to the nerve fibers. The peripapillary area was divided into 64 tracks with equal flux. An iterative algorithm traced the trajectory of the tracks assuming that the relative distribution of the NFF was conserved with compensation for fiber connections to ganglion cells on the macular side. Average trajectory was averaged from normal eyes and use to calculate the NFF maps for glaucomatous eyes. The NFF maps were divided into eight sectors that correspond to visual field regions.

RESULTS

There were 24 healthy and 10 glaucomatous eyes enrolled. The algorithm converged on similar patterns of NFL tracks for all healthy eyes. In glaucomatous eyes, NFF correlated with visual field sensitivity in the arcuate sectors (Spearman ρ = 0.53-0.62). Focal nerve fiber loss in glaucomatous eyes appeared as uniform tracks of NFF defects that followed the expected arcuate fiber trajectory.

CONCLUSIONS

Using an algorithm based on the conservation of flux, we derived nerve fiber trajectories in the peripapillary area. The NFF map is useful for the visualization of focal defects and quantification of sector nerve fiber loss from wide-area volumetric OCT scans.

TRANSLATIONAL RELEVANCE

NFF provides a cumulative measure of volumetric loss along nerve fiber tracks and could improve the detection of focal glaucoma damage.

Age and axial length on peripapillary retinal nerve fiber layer thickness measured by optical coherence tomography in nonglaucomatous Taiwanese participants

Purpose

This study investigates the influence of age and axial length (AL) on the peripapillary retinal nerve fiber layer (RNFL) thickness, as measured by optical coherence tomography (OCT).

Methods

Healthy patients visiting an eye clinic at a county hospital were recruited. All participants underwent comprehensive ophthalmologic examinations, and their retinas were scanned using 3D OCT-1000. In total, 223 patients with 446 eyes were included. The mean age and AL were 42.07 ± 13.16 (21–76) years and 25.38 ± 1.73 (21.19–30.37) mm, respectively.

Results

The average RNFL thickness decreased by 2.71 μm for every 10-year increase in age (P < 0.001). Age-related RNFL thinning was more significant in participants older than 41 years (-0.24μm/year; P = 0.015). The earliest sector showing a significant decline in RNFL thickness was after 35 years of age (-0.70μm/year; P = 0.011) at the superior quadrant and at the 1–2 o’clock hour (-1.42μm/year; P = 0.009). Meanwhile, the maximal rate of age-associated RNFL decay was observed in these two regions as well. The reduction of RNFL with age progression did not differ in eyes with long AL (> 27 mm; -0.16μm/year) or those with short AL (< 25 mm; -0.22μm/year). For every 1-mm-greater AL, RNFL was thinner by 1.78 μm (P < 0.001). The inferior quadrant showed the greatest tendency of RNFL decline with longer AL (4.46 μm/mm; P < 0.001).

Conclusions

The factors of age and AL should be considered when interpreting the results. Significantly age-associated RNFL thinning was found in participants older than 41 years. Reduction of RNFL thickness with increasing age was not affected by AL. Topographic variations in RNFL thinning were observed in that the maximal decline of RNFL thickness with advancing age at the superior quadrant whereas with elongation of AL at the inferior quadrant.

Glaucoma history and risk factors

Apart from the risk of developing glaucoma there is also the risk that it is not detected and irreversible loss of vision ensues. Some studies of methods of glaucoma diagnosis have examined the results of instrument-based examinations with great if not complete reliance on objective findings in arriving at a diagnosis. The very valuable advances in glaucoma detection instrument technologies, and apparent increasing dependence on them, may have led to reduced consideration of information available from a patient history in those studies. Dependence on objective evidence of glaucomatous pathology may reduce the possibility of detecting glaucoma suspects or patients at risk for becoming glaucoma suspects. A valid positive family history of glaucoma is very valuable information. However, negative family histories can often be unreliable due to large numbers of glaucoma cases being undiagnosed. No evidence of family history is appropriate rather than no family history. In addition the unreliability of a negative family history is increased when patients with glaucoma fail to inform their family members. A finding of no family history can only be stated as no known family history. In examining the potential diagnostic contribution from a patient history, this review considers, age, frailty, race, type and degree of refractive error, systemic hyper- and hypotension, vasospasm, migraine, pigmentary dispersion syndrome, pseudoexfoliation syndrome, obstructive sleep apnea syndrome, diabetes, medication interactions and side effects, the degree of exposure to intraocular and intracranial pressure elevations and fluctuations, smoking, and symptoms in addition to genetics and family history of the disease.

Diagnostic Ability of Retinal Nerve Fiber Layer Thickness Deviation Map for Localized and Diffuse Retinal Nerve Fiber Layer Defects

Purpose. To evaluate the diagnostic ability of the retinal nerve fiber layer (RNFL) deviation map for glaucoma with localized or diffuse RNFL defects. Methods. Eyes of 139 glaucoma patients and 165 healthy subjects were enrolled. All participants were imaged with Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, CA, USA). A RNFL defect was defined as at least 10 contiguous red (<1% level) superpixels in RNFL deviation map. The area, location, and angular width of RNFL defects were automatically measured. We compared sensitivities, specificities, and area under the receiver operating characteristic curves (AUCs) of RNFL deviation map and circumpapillary RNFL thickness for localized and diffuse RNFL defects. Subgroup analysis was performed according to the severity of glaucoma. Results. For localized defects, the area of RNFL defects (AUC, 0.991; sensitivity, 97%; specificity, 90%) in deviation map showed a higher diagnostic performance (p = 0.002) than the best circumpapillary RNFL parameter (inferior RNFL thickness; AUC, 0.914; sensitivity, 79%; specificity, 92%). For diffuse defects, there was no significant difference between the RNFL deviation map and circumpapillary RNFL parameters. In mild glaucoma with localized defect, RNFL deviation map showed a better diagnostic performance than circumpapillary RNFL measurement. Conclusions. RNFL deviation map is a useful tool for evaluating glaucoma regardless of localized or diffuse defect type and has advantages over circumpapillary RNFL measurement for detecting localized RNFL defects.

Clinical Use of an Optical Coherence Tomography Linear Discriminant Function for Differentiating Glaucoma From Normal Eyes

PURPOSE

To determine and validate the diagnostic ability of a linear discriminant function (LDF) based on retinal nerve fiber layer (RNFL) and ganglion cell-inner plexiform layer (GCIPL) thickness obtained using high-definition optical coherence tomography (Cirrus HD-OCT) for discriminating between healthy controls and early glaucoma subjects.

METHODS

We prospectively selected 214 healthy controls and 152 glaucoma subjects (teaching set) and another independent sample of 86 healthy controls and 71 glaucoma subjects (validating set). Two scans, including 1 macular and 1 peripapillary RNFL scan, were obtained. After calculating the LDF in the teaching set using the binary logistic regression analysis, receiver operating characteristic curves were plotted and compared between the OCT-provided parameters and LDF in the validating set.

RESULTS

The proposed LDF was 16.529-(0.132×superior RNFL)-(0.064×inferior RNFL)+(0.039×12 o'clock RNFL)+(0.038×1 o'clock RNFL)+(0.084×superior GCIPL)-(0.144×minimum GCIPL). The highest area under the receiver operating characteristic (AUROC) curve was obtained for LDF in both sets (AUROC=0.95 and 0.96). In the validating set, the LDF showed significantly higher AUROC than the best RNFL (inferior RNFL=0.91) and GCIPL parameter (minimum GCIPL=0.88). The LDF yielded a sensitivity of 93.0% at a fixed specificity of 85.0%.

CONCLUSIONS

The LDF showed better diagnostic ability for differentiating between healthy and early glaucoma subjects than individual OCT parameters. A classification algorithm based on the LDF can be used in the OCT analysis for glaucoma diagnosis.

[Evaluation of the new Canon HS-100 SD-OCT: Reproducibility of macular ganglion cell complex (GCC) thickness measurement in normal, hypertensive and glaucomatous eyes]

PURPOSE

To evaluate intra- and interobserver reproducibility of macular GCC thickness measurement by automated segmentation on the Canon HS-100 SD-OCT (Tokyo, Japan) in normal (N), hypertensive (OHT) and glaucomatous eyes.

METHODS

A total of 179 eyes of 93 patients were included: 90 N, 28 OHT and 36 early glaucoma and 25 advanced glaucoma. All patients underwent a complete ophthalmologic exam, central corneal thickness and 24-2 standard automated perimetry (HFA SITA standard). Each of two observers performed three macular acquisitions with the Canon OCT HS-100. Acquisitions were analyzed with the Glaucoma 3D mode, which estimated the macular GCC thickness in global, superior and inferior hemisectors, and in eight separate macular areas. Reproducibility was assessed by intraclass correlation coefficient (ICC), coefficient of variation (CV) and test-retest variability (TRTV) calculated as 1.96 times the standard deviation.

RESULTS

Mean GCC thickness was respectively 92.4 μm, 89.0 μm, 80.7 μm and 71.2 μm in N, OHT, early and advanced glaucomatous eyes. In all groups, intra- and interobserver reproducibility ranged respectively for ICC from 89.8 to 99.8% and from 90.2 to 99.4%, for CV from 0.43 to 1.95% and from 0.58 to 2.16% and for TRTV from 0.8 to 3.22 μm and from 1.04 to 3.53 μm. GCC thickness measurements using the new HS-100 SD-OCT were highly reproducible. However, in the advanced glaucoma group, while the reproducibility of GCC thickness measurement is good in the average, superior and inferior hemisectors of the macula, it was slightly less for the paracentral sectors, especially inferior. These sectors correspond generally to the areas most affected by glaucoma.

CONCLUSION

The reproducibility of GCC thickness measurements using the new Canon HS-100 SD-OCT is high for normal, OHT, and glaucomatous eyes. It is thus a reliable and reproducible ancillary test available to the clinician for the examination of glaucomatous optic neuropathies.

Quantitative analysis of localized retinal nerve fiber layer defects using spectral domain optical coherence tomography

PURPOSE

To compare the topographic features of localized retinal nerve fiber layer (RNFL) defects presented in red-free RNFL photography and spectral domain optical coherence tomography (SD-OCT), and to evaluate the correlation with structural and functional parameters.

METHODS

Sixty eyes with localized RNFL defects in red-free RNFL photographs were included. RNFL thickness map and significance map were obtained by SD-OCT. The angular location, angular width, and area of localized RNFL defects were measured and compared among RNFL thickness map, significance map (red, <1% level; yellow, <5% level), and RNFL photograph. The RNFL defect areas were analyzed by their correlations with structural and functional parameters.

RESULTS

The RNFL defect area of RNFL thickness map was significantly greater than those of red significance map, and smaller than those of RNFL photograph and yellow significance map (all P<0.001). RNFL thickness map showed a significantly narrower angular width than RNFL photograph and yellow significance map (all P<0.001). The area, angular width, and angular location of localized RNFL defects in RNFL photographs strongly correlated with those of RNFL thickness maps and significance maps (all r≥0.741, P<0.001). The relationship between RNFL defect area and structural-functional parameters was also significant.

CONCLUSIONS

The high topographic correlations in RNFL defects between RNFL photography and SD-OCT RNFL maps suggest the validity of SD-OCT RNFL imaging for detecting localized glaucomatous RNFL damage. In addition, structural and functional parameters are closely related to RNFL defect areas. Quantitative measurements of RNFL defects might be valuable for glaucoma diagnosis.

Systematic review of macular ganglion cell complex analysis using spectral domain optical coherence tomography for glaucoma assessment

AIM: To review the use of spectral domain optical coherence tomography (SD-OCT) for macular retinal ganglion cells (RGC) and ganglion cell complex (GCC) measurement in glaucoma assessment, specifically for early detection and detection of disease progression.

METHODS: A systematic review was performed by searching PubMed, Medline, and Web of Science for articles published in English through July 2014 describing the various macular SD-OCT scanning strategies developed for glaucoma assessment. The review focused on papers evaluating the use of macular RGC/GCC SD-OCT to detect early glaucoma and its progression. The search included keywords corresponding to the index test (macular ganglion cell/RGC/GCC/Spectral domain OCT), the target condition (glaucoma), and diagnostic performance. The RGC/GCC SD-OCT scanning strategies used to assess glaucoma of most commonly used SD-OCT instruments were described and compared. These included the Cirrus high definition-OCT (Carl Zeiss Meditec, Inc., Dublin, CA, United States), RTVue (Optovue, Inc., Fremont, CA, United States), Spectralis (Heidelberg Engineering, Heidelberg, Germany) and the 3D OCT 2000 (Topcon Corporation, Tokyo, Japan). Studies focusing on the ability of RGC/GCC SD-OCT to detect early glaucomatous damage and on the correlation between glaucomatous progression and RGC/GCC measurement by SD-OCT were reviewed.

RESULTS: According to the literature, macular RGC/GCC SD-OCT has high diagnostic power of preperimetric glaucoma, reliable discrimination ability to differentiate between healthy eyes and glaucomatous eyes, with good correlation with visual filed damage. The current data suggests that it may serve as a sensitive detection tool for glaucomatous structural progression even with mild functional progression as the rate of change of RGC/GCC thickness was found to be significantly higher in progressing than in stable eyes. Glaucoma assessment with RGC/GCC SD-OCT was comparable with and sometimes better than circumpapillary retinal nerve fiber layer thickness measurement.

CONCLUSION: An increasing body of evidence supports using macular RGC/GCC thickness as an indicator for early glaucoma. This might be a useful tool for monitoring disease progression.

Comparison of glaucoma diagnoses using Stratus and Cirrus optical coherence tomography in different glaucoma types in a Chinese population

PURPOSE

To compare the glaucoma diagnostic power of Stratus and Cirrus optical coherence tomographies (OCTs) in a Taiwan Chinese population with different glaucoma types.

PATIENTS AND METHODS

One eye each was chosen from 21 ocular hypertension (OH) patients, 27 glaucoma-suspect (GS) patients, 35 primary open-angle glaucoma (POAG) patients, 26 primary angle-closure glaucoma (PACG) patients, and 52 normal subjects. Early glaucoma (EG) was identified among glaucomatous eyes on the basis of the visual field severity (better than -9 dB). All participants were imaged using 2 OCT units at the same visit. The area under the receiver operator characteristic (AROC) curve was used to differentiate normal eyes from OH, GS, POAG, PACG, and EG eyes, and the sensitivity and specificity of each parameter from internal normative classifications were analyzed.

RESULTS

For normal versus OH eyes, the best AROC value was the average thickness (Stratus, 0.693; Cirrus, 0.697). For normal versus GS eyes, the best AROC value was the average thickness (Stratus, 0.807; Cirrus, 0.776). For normal versus POAG eyes, the best AROC value was the average thickness (Stratus, 0.943; Cirrus, 0.930). For normal versus PACG eyes, the best AROC value was the 5-o'clock hour (Stratus, 0.830; Cirrus, 0.817). For normal versus EG eyes, the best AROC value was the average thickness with Stratus (0.868) and the 5-o'clock hour with Cirrus (0.876). All sensitivities in the 5 groups were fair on the basis of the internal normal classification database of both OCTs.

CONCLUSIONS

Cirrus and Stratus OCTs showed equal diagnostic power in EG, OH, GS, POAG, and PACG eyes in a Taiwan Chinese population. The utility of the current internal databases of both OCT units for the Chinese population is an interesting issue that needs to be addressed in the future.

Reproducibility of peripapillary retinal nerve fiber layer thickness measured by spectral domain optical coherence tomography in pseudophakic eyes

Purpose

To assess the reproducibility of circumpapillary retinal nerve fiber layer (cpRNFL) thickness measurement (measurement agreement) and its color-coded classification (classification agreement) by Cirrus spectral domain optical coherence tomography (OCT) in pseudophakic eyes.

Methods

Two-hundred five participants having glaucoma or glaucoma suspected eyes underwent two repeated Cirrus OCT scans to measure cpRNFL thickness (optic disc cube 200 × 200). After classifying participants into three different groups according to their lens status (clear media, cataract, and pseudophakic), values of intra-class coefficient (ICC), coefficient of variance, and test-retest variability were compared between groups for average retinal nerve fiber layer (RNFL) thicknesses and that corresponding to four quadrant maps. Linear weighted kappa coefficients were calculated as indicators of agreement of color code classification in each group.

Results

ICC values were all excellent (generally defined as 0.75 to 1.00) for the average and quadrant RNFL thicknesses in all three groups. ICC values of the clear media group tended to be higher than those in the cataract and pseudophakic groups for all quadrants and average thickness. Especially in the superior and nasal quadrants, the ICC value of the cataract group was significantly lower than that of the clear media and pseudophakic groups. For average RNFL thickness, classification agreement (kappa) in three groups did not show a statistically significant difference. For quadrant maps, classification agreement (kappa) in the clear media group was higher than those in the other two groups.

Conclusions

Agreement of cpRNFL measurement and its color code classification between two repeated Cirrus OCT scans in pseudophakic eyes was as good as that in eyes with clear crystalline lens. More studies are required to ascertain the effect of lens status on the reproducibility of Cirrus OCT according to different stages of glaucoma patients.

Combining spectral domain optical coherence tomography structural parameters for the diagnosis of glaucoma with early visual field loss

PURPOSE

To create a multivariable predictive model for glaucoma with early visual field loss using a combination of spectral-domain optical coherence tomography (SD-OCT) parameters, and to compare the results with single variable models.

METHODS

Two hundred fifty-three subjects (149 healthy controls and 104 with early glaucoma) underwent optic disc and macular scanning using SD-OCT in one randomly selected eye per subject. Sixteen parameters (rim area, cup-to-disc area ratio, vertical cup-to-disc diameter ratio, average and quadrant RNFL thicknesses, average, minimum, and sectoral ganglion cell inner-plexiform layer [GCIPL] thicknesses) were collected and submitted to an exploratory factor analysis (EFA) followed by logistic regression with the backward elimination variable selection technique. Area under the curve (AUC) of the receiver operating characteristic (ROC), sensitivity, specificity, Akaike's information criterion (AIC), predicted probability, prediction interval length (PIL), and classification rates were used to determine the performances of the univariable and multivariable models.

RESULTS

The multivariable model had an AUC of 0.995 with 98.6% sensitivity, 96.0% specificity, and an AIC value of 43.29. Single variable models yielded AUCs of 0.943 to 0.987, sensitivities of 82.6% to 95.7%, specificities of 88.0% to 94.0%, and AICs of 113.16 to 59.64 (smaller is preferred). The EFA logistic regression model correctly classified 91.67% of cases with a median PIL of 0.050 in the validation set. Univariable models correctly classified 80.62% to 90.48% of cases with median PILs 1.9 to 3.0 times larger.

CONCLUSIONS

The multivariable model was successful in predicting glaucoma with early visual field loss and outperformed univariable models in terms of AUC, AIC, PILs, and classification rates.

Glaucoma Diagnostic Accuracy of Machine Learning Classifiers Using Retinal Nerve Fiber Layer and Optic Nerve Data from SD-OCT

Purpose. To investigate the diagnostic accuracy of machine learning classifiers (MLCs) using retinal nerve fiber layer (RNFL) and optic nerve (ON) parameters obtained with spectral domain optical coherence tomography (SD-OCT). Methods. Fifty-seven patients with early to moderate primary open angle glaucoma and 46 healthy patients were recruited. All 103 patients underwent a complete ophthalmological examination, achromatic standard automated perimetry, and imaging with SD-OCT. Receiver operating characteristic (ROC) curves were built for RNFL and ON parameters. Ten MLCs were tested. Areas under ROC curves (aROCs) obtained for each SD-OCT parameter and MLC were compared. Results. The mean age was 56.5 ± 8.9 years for healthy individuals and 59.9 ± 9.0 years for glaucoma patients (P = 0.054). Mean deviation values were −1.4 dB for healthy individuals and −4.0 dB for glaucoma patients (P < 0.001). SD-OCT parameters with the greatest aROCs were cup/disc area ratio (0.846) and average cup/disc (0.843). aROCs obtained with classifiers varied from 0.687 (CTREE) to 0.877 (RAN). The aROC obtained with RAN (0.877) was not significantly different from the aROC obtained with the best single SD-OCT parameter (0.846) (P = 0.542). Conclusion. MLCs showed good accuracy but did not improve the sensitivity and specificity of SD-OCT for the diagnosis of glaucoma.

Comparison of ability of time-domain and spectral-domain optical coherence tomography to detect diffuse retinal nerve fiber layer atrophy

PURPOSE

Our aim was to evaluate and compare diagnostic capabilities of time-domain (Stratus) and spectral-domain (Cirrus) optical coherence tomography (OCT) to detect diffuse retinal nerve fiber layer (RNFL) atrophy.

METHODS

This study assessed 101 eyes from 101 glaucoma patients with diffuse RNFL atrophy and 101 eyes from 101 age-matched healthy individuals. Two experienced glaucoma specialists graded red-free RNFL photographs of eyes with diffuse RNFL atrophy using a four-level grading system. The area under the receiver operating characteristic curves (AUC) of normal eyes was compared with that of eyes with diffuse atrophy. Sensitivity and specificity of each OCT device were calculated on the basis of its internal normative database.

RESULTS

The largest AUC for Stratus and Cirrus were obtained for average RNFL thicknesses (0.96 and 0.94, respectively). Comparison of the AUC with different RNFL atrophy grades revealed no significant difference between the two OCT devices. Using an internal normative database at a <5 % level, the overall sensitivity of Stratus ranged from 58.0 to 84.0 %, whereas that of Cirrus ranged from 75.0 to 87.0 %. According to the normative database, the highest Stratus sensitivity was obtained with the temporal-superior-nasal-inferior-temporal (TSNIT) thickness graph, and the highest Cirrus sensitivity with the TSNIT thickness graph and the deviation map.

CONCLUSIONS

The AUC obtained from Cirrus were comparable with those from Stratus. On the basis of their normative databases, these devices had similar diagnostic accuracy. Our results suggest that the diagnostic capabilities of the two instruments to detect diffuse RNFL atrophy are similar.

Detection of macular ganglion cell loss in preperimetric glaucoma patients with localized retinal nerve fibre defects by spectral-domain optical coherence tomography

BACKGROUND

To evaluate and compare the utility of ganglion cell complex with peripapillary retinal nerve fibre layer and optic nerve head measurements for detection of localized defects in patients with preperimetric glaucoma using spectral-domain optical coherence tomography.

DESIGN

Prospective study.

PARTICIPANTS

Preperimetric glaucoma patients.

METHODS

A total of 105 eyes with preperimetric glaucoma and 68 age- and refractive error-matched control eyes were enrolled. The ability to detect localized retinal nerve fibre layer defects by RTVue-100 spectral-domain optical coherence tomography (Optovue, Inc., Fremont, CA, USA) was assessed calculating the areas under receiver operating characteristic curves.

MAIN OUTCOME MEASURES

The ability to detect localized retinal nerve fibre layer defects by spectral-domain optical coherence tomography.

RESULTS

Global volume loss and superior ganglion cell complex thickness showed the largest area under receiver operating characteristic curve values (both areas under receiver operating characteristic curves 0.84, P < 0.001) among ganglion cell complex parameters. Average peripapillary retinal nerve fibre layer thickness afforded the best diagnostic capability (area under receiver operating characteristic curve 0.89, P < 0.001), whereas among optic nerve head parameters, the horizontal cup:disc ratio yielded the highest area under receiver operating characteristic curve (0.85, P < 0.001). No statistical difference was evident between the areas under receiver operating characteristic curves of the most informative parameters when the data were gathered from the three different sites (ganglion cell complex, peripapillary retinal nerve fibre layer, and optic nerve head) (P > 0.02).

CONCLUSIONS

Ganglion cell complex thickness was significantly reduced in eyes with preperimetric glaucoma. Ganglion cell complex imaging using spectral-domain optical coherence tomography may be a useful ancillary modality for detection of early macular changes in glaucomatous eyes with localized retinal nerve fibre layer defects.