Linear discriminant analysis and artificial neural network for glaucoma diagnosis using scanning laser polarimetry–variable cornea compensation measurements in Taiwan Chinese population

Artificial intelligence in glaucoma: opportunities, challenges, and future directions

Artificial intelligence (AI) has shown excellent diagnostic performance in detecting various complex problems related to many areas of healthcare including ophthalmology. AI diagnostic systems developed from fundus images have become state-of-the-art tools in diagnosing retinal conditions and glaucoma as well as other ocular diseases. However, designing and implementing AI models using large imaging data is challenging. In this study, we review different machine learning (ML) and deep learning (DL) techniques applied to multiple modalities of retinal data, such as fundus images and visual fields for glaucoma detection, progression assessment, staging and so on. We summarize findings and provide several taxonomies to help the reader understand the evolution of conventional and emerging AI models in glaucoma. We discuss opportunities and challenges facing AI application in glaucoma and highlight some key themes from the existing literature that may help to explore future studies. Our goal in this systematic review is to help readers and researchers to understand critical aspects of AI related to glaucoma as well as determine the necessary steps and requirements for the successful development of AI models in glaucoma.

Multi-stage glaucoma classification using pre-trained convolutional neural networks and voting-based classifier fusion

Aim: To design an automated glaucoma detection system for early detection of glaucoma using fundus images. Background: Glaucoma is a serious eye problem that can cause vision loss and even permanent blindness. Early detection and prevention are crucial for effective treatment. Traditional diagnostic approaches are time consuming, manual, and often inaccurate, thus making automated glaucoma diagnosis necessary. Objective: To propose an automated glaucoma stage classification model using pre-trained deep convolutional neural network (CNN) models and classifier fusion. Methods: The proposed model utilized five pre-trained CNN models: ResNet50, AlexNet, VGG19, DenseNet-201, and Inception-ResNet-v2. The model was tested using four public datasets: ACRIMA, RIM-ONE, Harvard Dataverse (HVD), and Drishti. Classifier fusion was created to merge the decisions of all CNN models using the maximum voting-based approach. Results: The proposed model achieved an area under the curve of 1 and an accuracy of 99.57% for the ACRIMA dataset. The HVD dataset had an area under the curve of 0.97 and an accuracy of 85.43%. The accuracy rates for Drishti and RIM-ONE were 90.55 and 94.95%, respectively. The experimental results showed that the proposed model performed better than the state-of-the-art methods in classifying glaucoma in its early stages. Understanding the model output includes both attribution-based methods such as activations and gradient class activation map and perturbation-based methods such as locally interpretable model-agnostic explanations and occlusion sensitivity, which generate heatmaps of various sections of an image for model prediction. Conclusion: The proposed automated glaucoma stage classification model using pre-trained CNN models and classifier fusion is an effective method for the early detection of glaucoma. The results indicate high accuracy rates and superior performance compared to the existing methods.

Serafini R, Lakhtakia S, Valliani AA, Warburton AJ, Patel A, Zhou D, Sklar B, Chelnis J, Elahi E. Portable hardware & software technologies for addressing ophthalmic health disparities: A systematic review

Vision impairment continues to be a major global problem, as the WHO estimates 2.2 billion people struggling with vision loss or blindness. One billion of these cases, however, can be prevented by expanding diagnostic capabilities. Direct global healthcare costs associated with these conditions totaled $255 billion in 2010, with a rapid upward projection to $294 billion in 2020. Accordingly, WHO proposed 2030 targets to enhance integration and patient-centered vision care by expanding refractive error and cataract worldwide coverage. Due to the limitations in cost and portability of adapted vision screening models, there is a clear need for new, more accessible vision testing tools in vision care. This comparative, systematic review highlights the need for new ophthalmic equipment and approaches while looking at existing and emerging technologies that could expand the capacity for disease identification and access to diagnostic tools. Specifically, the review focuses on portable hardware- and software-centered strategies that can be deployed in remote locations for detection of ophthalmic conditions and refractive error. Advancements in portable hardware, automated software screening tools, and big data-centric analytics, including machine learning, may provide an avenue for improving ophthalmic healthcare.

Histogram of Oriented Gradients (HOG)-Based Artificial Neural Network (ANN) Classifier for Glaucoma Detection

Glaucoma is a severe condition of the optic nerve resulting in the loss of eyesight. The proposed methodology has introduced the extraction of HOG (histogram of oriented gradients) features from the retinal fundus image. After the removal of HOG features, the authors compare the performance of five different machine learning techniques like k-nearest neighbour (KNN), support vector machine (SVM), linear discriminant analysis (LDA), naïve bayes, and artificial neural network. The process of image classification is based on analyzing the numerical properties of the obtained image features and classifying the data into different categories. In the paper, the authors intend to classify whether the image belongs to the glaucomatous category or the healthy category. After the application of the different classification algorithms to the test data and further analysis of the results, they could conclude that the SVM classifier provided an accuracy of 90%, KNN 86%, Naïve Bayes 96%, LDA 86%, and ANN 96.90% on the dataset in hand.

[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.

Machine learning models based on the dimensionality reduction of standard automated perimetry data for glaucoma diagnosis

INTRODUCTION

Visual field testing via standard automated perimetry (SAP) is a commonly used glaucoma diagnosis method. Applying machine learning techniques to the visual field test results, a valid clinical diagnosis of glaucoma solely based on the SAP data is provided. In order to reflect structural-functional patterns of glaucoma on the automated diagnostic models, we propose composite variables derived from anatomically grouped visual field clusters to improve the prediction performance. A set of machine learning-based diagnostic models are designed that implement different input data manipulation, dimensionality reduction, and classification methods.

METHODS

Visual field testing data of 375 healthy and 257 glaucomatous eyes were used to build the diagnostic models. Three kinds of composite variables derived from the Garway-Heath map and the glaucoma hemifield test (GHT) sector map were included in the input variables in addition to the 52 SAP visual filed locations. Dimensionality reduction was conducted to select important variables so as to alleviate high-dimensionality problems. To validate the proposed methods, we applied four classifiers-linear discriminant analysis, naïve Bayes classifier, support vector machines, and artificial neural networks-and four dimensionality reduction methods-Pearson correlation coefficient-based variable selection, Markov blanket variable selection, the minimum redundancy maximum relevance algorithm, and principal component analysis- and compared their classification performances.

RESULTS

For all tested combinations, the classification performance improved when the proposed composite variables and dimensionality reduction techniques were implemented. The combination of total deviation values, the GHT sector map, support vector machines, and Markov blanket variable selection obtains the best performance: an area under the receiver operating characteristic curve (AUC) of 0.912.

CONCLUSION

A glaucoma diagnosis model giving an AUC of 0.912 was constructed by applying machine learning techniques to SAP data. The results show that dimensionality reduction not only reduces dimensions of the input space but also enhances the classification performance. The variable selection results show that the proposed composite variables from visual field clustering play a key role in the diagnosis model.

Machine Learning Techniques in Clinical Vision Sciences

This review presents and discusses the contribution of machine learning techniques for diagnosis and disease monitoring in the context of clinical vision science. Many ocular diseases leading to blindness can be halted or delayed when detected and treated at its earliest stages. With the recent developments in diagnostic devices, imaging and genomics, new sources of data for early disease detection and patients' management are now available. Machine learning techniques emerged in the biomedical sciences as clinical decision-support techniques to improve sensitivity and specificity of disease detection and monitoring, increasing objectively the clinical decision-making process. This manuscript presents a review in multimodal ocular disease diagnosis and monitoring based on machine learning approaches. In the first section, the technical issues related to the different machine learning approaches will be present. Machine learning techniques are used to automatically recognize complex patterns in a given dataset. These techniques allows creating homogeneous groups (unsupervised learning), or creating a classifier predicting group membership of new cases (supervised learning), when a group label is available for each case. To ensure a good performance of the machine learning techniques in a given dataset, all possible sources of bias should be removed or minimized. For that, the representativeness of the input dataset for the true population should be confirmed, the noise should be removed, the missing data should be treated and the data dimensionally (i.e., the number of parameters/features and the number of cases in the dataset) should be adjusted. The application of machine learning techniques in ocular disease diagnosis and monitoring will be presented and discussed in the second section of this manuscript. To show the clinical benefits of machine learning in clinical vision sciences, several examples will be presented in glaucoma, age-related macular degeneration, and diabetic retinopathy, these ocular pathologies being the major causes of irreversible visual impairment.

Optic nerve head and fibre layer imaging for diagnosing glaucoma

BACKGROUND

The diagnosis of glaucoma is traditionally based on the finding of optic nerve head (ONH) damage assessed subjectively by ophthalmoscopy or photography or by corresponding damage to the visual field assessed by automated perimetry, or both. Diagnostic assessments are usually required when ophthalmologists or primary eye care professionals find elevated intraocular pressure (IOP) or a suspect appearance of the ONH. Imaging tests such as confocal scanning laser ophthalmoscopy (HRT), optical coherence tomography (OCT) and scanning laser polarimetry (SLP, as used by the GDx instrument), provide an objective measure of the structural changes of retinal nerve fibre layer (RNFL) thickness and ONH parameters occurring in glaucoma.

OBJECTIVES

To determine the diagnostic accuracy of HRT, OCT and GDx for diagnosing manifest glaucoma by detecting ONH and RNFL damage.

SEARCH METHODS

We searched several databases for this review. The most recent searches were on 19 February 2015.

SELECTION CRITERIA

We included prospective and retrospective cohort studies and case-control studies that evaluated the accuracy of OCT, HRT or the GDx for diagnosing glaucoma. We excluded population-based screening studies, since we planned to consider studies on self-referred people or participants in whom a risk factor for glaucoma had already been identified in primary care, such as elevated IOP or a family history of glaucoma. We only considered recent commercial versions of the tests: spectral domain OCT, HRT III and GDx VCC or ECC.

DATA COLLECTION AND ANALYSIS

We adopted standard Cochrane methods. We fitted a hierarchical summary ROC (HSROC) model using the METADAS macro in SAS software. After studies were selected, we decided to use 2 x 2 data at 0.95 specificity or closer in meta-analyses, since this was the most commonly-reported level.

MAIN RESULTS

We included 106 studies in this review, which analysed 16,260 eyes (8353 cases, 7907 controls) in total. Forty studies (5574 participants) assessed GDx, 18 studies (3550 participants) HRT, and 63 (9390 participants) OCT, with 12 of these studies comparing two or three tests. Regarding study quality, a case-control design in 103 studies raised concerns as it can overestimate accuracy and reduce the applicability of the results to daily practice. Twenty-four studies were sponsored by the manufacturer, and in 15 the potential conflict of interest was unclear.Comparisons made within each test were more reliable than those between tests, as they were mostly based on direct comparisons within each study.The Nerve Fibre Indicator yielded the highest accuracy (estimate, 95% confidence interval (CI)) among GDx parameters (sensitivity: 0.67, 0.55 to 0.77; specificity: 0.94, 0.92 to 0.95). For HRT measures, the Vertical Cup/Disc (C/D) ratio (sensitivity: 0.72, 0.60 to 0.68; specificity: 0.94, 0.92 to 0.95) was no different from other parameters. With OCT, the accuracy of average RNFL retinal thickness was similar to the inferior sector (0.72, 0.65 to 0.77; specificity: 0.93, 0.92 to 0.95) and, in different studies, to the vertical C/D ratio.Comparing the parameters with the highest diagnostic odds ratio (DOR) for each device in a single HSROC model, the performance of GDx, HRT and OCT was remarkably similar. At a sensitivity of 0.70 and a high specificity close to 0.95 as in most of these studies, in 1000 people referred by primary eye care, of whom 200 have manifest glaucoma, such as in those who have already undergone some functional or anatomic testing by optometrists, the best measures of GDx, HRT and OCT would miss about 60 cases out of the 200 patients with glaucoma, and would incorrectly refer 50 out of 800 patients without glaucoma. If prevalence were 5%, e.g. such as in people referred only because of family history of glaucoma, the corresponding figures would be 15 patients missed out of 50 with manifest glaucoma, avoiding referral of about 890 out of 950 non-glaucomatous people.Heterogeneity investigations found that sensitivity estimate was higher for studies with more severe glaucoma, expressed as worse average mean deviation (MD): 0.79 (0.74 to 0.83) for MD < -6 db versus 0.64 (0.60 to 0.69) for MD ≥ -6 db, at a similar summary specificity (0.93, 95% CI 0.92 to 0.94 and, respectively, 0.94; 95% CI 0.93 to 0.95; P < 0.0001 for the difference in relative DOR).

AUTHORS' CONCLUSIONS

The accuracy of imaging tests for detecting manifest glaucoma was variable across studies, but overall similar for different devices. Accuracy may have been overestimated due to the case-control design, which is a serious limitation of the current evidence base.We recommend that further diagnostic accuracy studies are carried out on patients selected consecutively at a defined step of the clinical pathway, providing a description of risk factors leading to referral and bearing in mind the consequences of false positives and false negatives in the setting in which the diagnostic question is made. Future research should report accuracy for each threshold of these continuous measures, or publish raw data.

AUTOMATED GLAUCOMA DETECTION USING HYBRID FEATURE EXTRACTION IN RETINAL FUNDUS IMAGES

Glaucoma is one of the most common causes of blindness. Robust mass screening may help to extend the symptom-free life for affected patients. To realize mass screening requires a cost-effective glaucoma detection method which integrates well with digital medical and administrative processes. To address these requirements, we propose a novel low cost automated glaucoma diagnosis system based on hybrid feature extraction from digital fundus images. The paper discusses a system for the automated identification of normal and glaucoma classes using higher order spectra (HOS), trace transform (TT), and discrete wavelet transform (DWT) features. The extracted features are fed to a support vector machine (SVM) classifier with linear, polynomial order 1, 2, 3 and radial basis function (RBF) in order to select the best kernel for automated decision making. In this work, the SVM classifier, with a polynomial order 2 kernel function, was able to identify glaucoma and normal images with an accuracy of 91.67%, and sensitivity and specificity of 90% and 93.33%, respectively. Furthermore, we propose a novel integrated index called Glaucoma Risk Index (GRI) which is composed from HOS, TT, and DWT features, to diagnose the unknown class using a single feature. We hope that this GRI will aid clinicians to make a faster glaucoma diagnosis during the mass screening of normal/glaucoma images.