Glaucomatous Eyes Research Articles

The relationship between corneal calcareous degeneration and various ophthalmic diseases in dogs: A retrospective study

AbstractObjectiveTo investigate ocular factors that influence the development of corneal calcareous degeneration (CCD) in dogs.Animals and ProceduresThe medical records of 96 eyes of dogs with CCD and 288 eyes without CCD were retrospectively reviewed. Dogs with evidence of causative systemic illness associated with CCD were excluded from the study. Logistic regression analysis was used to identify the ocular factors associated with the development of CCD. To identify the effect of phosphate‐containing eyedrops on CCD, the application periods of phosphate‐containing antiglaucoma eyedrops were compared between the glaucomatous eyes in the CCD and non‐CCD groups.ResultsIncreased age, brachycephalic breed, keratoconjunctivitis sicca, advanced cataract, history of phacoemulsification, and topical corticosteroid application were significantly associated with CCD development. Glaucoma was significantly overrepresented in the non‐CCD group, and the application period of phosphate‐containing antiglaucoma eyedrops was significantly longer in eyes with CCD than in those without CCD.ConclusionsOphthalmic diseases requiring long‐term management of ocular inflammation and long‐term application of phosphate‐containing eyedrops may contribute to the development of CCD. Glaucoma is overrepresented in dogs without CCD, which is thought to be due to the differences in predisposed age and breeds between dogs with glaucoma and CCD.

Papillomacular bundle defect (PMBD) in glaucoma patients with high myopia: frequency and risk factors

Little is known about the papillomacular bundle defect (PMBD) in glaucoma. As such, we investigated the frequency of PMBD in glaucoma patients with high myopia, and its risk factors. In this retrospective, cross-sectional study, retinal nerve fiber layer (RNFL) defect was analyzed in 92 glaucomatous eyes with high myopia (axial length of 26.0 mm or more or an average spherical value of − 6.0 diopters or less). After dividing them into two groups with and without PMBD, the clinical characteristics of the groups were compared and analyzed. The mean age of the patients was 52.1 ± 10.5 years, and there were 53 males and 39 females. PMBD were observed in 55 eyes (59.8%). There was no significant intergroup difference in baseline or follow-up intraocular pressure (IOP). Parapapillary atrophy (PPA)-to-disc-area ratio (OR 3.83, CI: 1.58–10.27, p = 0.010), lamina cribrosa defect (LCD; OR 2.92, CI: 1.14–8.13, p = 0.031) and central visual field defect (CVFD; OR 3.56, CI: 1.38–9.58, p = 0.010) were significantly associated with the PMBD..

Long-term variability of retinal nerve fibre layer thickness measurement in patients with glaucoma of African and European descents

BackgroundTo examine long-term retinal nerve fibre layer thickness (RNFLT) variability and associated clinical factors in African (AD) and European descent (ED) individuals with glaucoma.MethodsThis retrospective cohort study included glaucoma eyes...

Time to Glaucoma Progression Detection by Optical Coherence Tomography in Individuals of African and European Descents

To examine the time to detectable retinal nerve fiber layer thickness (RNFLT) progression by optical coherence tomography (OCT) among glaucoma patients of African descent (AD) and European descent (ED). Retrospective cohort study. AD and ED glaucoma eyes from the Diagnostic Innovations in Glaucoma Study (DIGS)/African Descent and Glaucoma Evaluation Study (ADAGES) with ≥2 years/4 visits of optic nerve head RNFLT measurements were included after homogenization on age, diagnosis, and baseline visual field (VF) measurement. RNFLT variability estimates based on linear mixed-effects models were used to simulate longitudinal RNFLT data for both races. Times to trend-based RNFLT progression detection were calculated under standardized scenarios (same RNFLT baseline/thinning rates for both races) and real-world scenarios (AD and ED cohort-specific RNFLT baseline/thinning rates). We included 332 and 542 eyes (216 and 317 participants) of AD and ED, respectively. In standardized scenarios, the time to detect RNFLT progression appeared to be similar (difference, <0.2 years) for AD and ED across different assumed RNFLT thinning rates/baseline. In real-world scenarios, compared to ED, AD had a faster RNFLT thinning rate (-0.8 vs -0.6 µm/y) and thicker baseline RNFLT (84.6 vs 81.8 µm). With a faster thinning rate, the mean (SD) time to progression detection was shorter in AD (4.8 [2.0] vs ED: 5.4 [2.4] years), and the 5-year progression rate appeared to be higher (AD: 59% vs ED: 47%). Time to progression detection was similar for both races when assuming identical RNFLT baseline/thinning rates, and shorter in AD eyes under real-world simulation when AD had faster RNFLT thinning. In contrast to prior results on VF, which detected progression later in AD eyes than in ED eyes, OCT may detect progression more consistently across these races.

Comparative Study of Additive IOP Lowering Effects and Safety Profile of Ripasudil 0.4% Combined with Timolol 0.5% and Ripasudil 0.4% Combined with Bimatoprost 0.03% In Primary Open Angle Glaucoma in Indian Eyes

Comparative Study of Additive IOP Lowering Effects and Safety Profile of Ripasudil 0.4% Combined with Timolol 0.5% and Ripasudil 0.4% Combined with Bimatoprost 0.03% In Primary Open Angle Glaucoma in Indian Eyes

Performance Metrics of AdaBoost and Random Forest in Multi-Class Eye Disease Identification: An Imbalanced Dataset Approach

This study presents a comprehensive evaluation of AdaBoost and Random Forest Classifier algorithms in the classification of eye diseases, focusing on a challenging scenario involving an imbalanced dataset. Eye diseases, particularly Cataract, Diabetic Retinopathy, Glaucoma, and Normal eye conditions, pose significant diagnostic challenges, and the advent of machine learning offers promising avenues for enhancing diagnostic accuracy. Our research utilizes a dataset preprocessed with Canny edge detection for image segmentation and Hu Moments for feature extraction, providing a robust foundation for the comparative analysis. The performance of the algorithms is assessed using a 5-fold cross-validation approach, with accuracy, precision, recall, and F1-score as the key metrics. The results indicate that the Random Forest Classifier outperforms AdaBoost across these metrics, albeit with moderate overall performance. This finding underscores the potential and limitations of using advanced machine learning techniques for medical image analysis, particularly in the context of imbalanced datasets. The study contributes to the field by providing insights into the effectiveness of different machine learning algorithms in handling the complexities of medical image classification. For future research, it recommends exploring a diverse range of image processing techniques, delving into other sophisticated machine learning models, and extending the study to encompass a wider array of eye diseases. These findings have practical implications in guiding the selection of machine learning tools for medical diagnostics and highlight the need for continuous improvement in automated systems for enhanced patient care.

An efficient framework for optic disk segmentation and classification of Glaucoma on fundus images

Accurately segmenting the Optic Disk is a crucial step in classifying Glaucoma using Fundus images. Machine learning and artificial intelligence techniques are widely used in Glaucoma detection, and the main indicators observed in Fundus images are the presence of Papillary Atrophy, Cup to Disc Ratio values, diminishing Neural Retinal Rim (NRR), the Inferior Superior Nasal Temporal (ISNT) rule, and Cup Diameter. In this research, we investigated various segmentation and classification techniques that can be applied to Optic disk segmentation and classification of normal and glaucomatous eyes. The proposed method will be beneficial to clinicians and healthcare workers in facilities with limited resources. In this paper, histogram processing is used to determine the type of image, and based on this information; we decide whether the image requires segmentation. Some datasets in the standard dataset contain complete retinal images while others include segmented optic disks. The segmented images are directly given as input for classification using the proposed Convolutional Neural Network (CNN). For complete retinal images, segmentation is performed using the Simple Linear Iterative clustering (SLIC) and normalized graph cut algorithm. The proposed framework's performance is compared with that of pretrained neural networks, including VGG19, InceptionV3, and ResNet50V2, using major metrics. We trained and tested these architectures with 3115 images from six standard datasets. Our proposed framework outperforms all with an accuracy of 96.33 %.

Myo/Nog Cells Increase in Response to Elevated Intraocular Pressure and Mitigate Ganglion Cell Death in a Mouse Model of Glaucoma

Glaucoma is one of the leading causes of blindness worldwide. Decreased aqueous humor drainage causes an increase in intraocular pressure (IOP), which in turn damages the ganglion cells of the retina and optic nerve. A mouse model of glaucoma was used to examine the behavior of Myo/Nog (M/N) cells, which were previously shown to respond to cataract surgery and retinopathy induced by hypoxia, light damage, and intravitreal injection of human retinal pigment epithelial cells. M/N cells express the skeletal-muscle-specific transcription factor MyoD, the bone morphogenetic protein inhibitor Noggin, and brain-specific angiogenesis inhibitor 1 (BAI1). Glaucoma was induced by injecting microbeads into the anterior chamber (AC) of the right eye to obstruct the flow of aqueous humor into the trabecular meshwork. IOP was elevated within three days of addition of microbeads. Loss of retinal ganglion cells (RGCs) and thinning of the ganglion cell layer–nerve fiber layer (GCL-NFL) was observed in tissue sections by day 32. The injection of microbeads resulted in an increase in BAI1-positive (+) M/N cells in the trabecular meshwork, ciliary body, canal of Schlemm, cornea, and ganglion cell layer (GCL). M/N cells ingested microbeads. The effect of further increasing the population of M/N cells on IOP and RGC loss was determined by injecting BAI1+ cells isolated from the brain into the AC of both eyes. Exogenous M/N cells prelabeled with CellTracker™ Red were found in the same tissues as the endogenous population of M/N cells in eyes with and without elevated IOP. The addition of M/N cells did not significantly reduce IOP in bead-injected eyes. However, there were significantly more RGCs and the NFL was thicker in glaucomatous eyes with M/N cell supplementation than eyes injected with phosphate-buffered saline. The numbers of RGCs and NFL thickness were similar in glaucomatous and non-glaucomatous eyes after adding M/N cells. These results demonstrate that endogenous M/N cells respond to elevated IOP in the anterior and posterior segments in response to induction of glaucoma. M/N cells’ mitigation of RGC loss may reflect a neuroprotective effect within the retina, as opposed to a significant drop in IOP.

Glaucoma after penetrating keratoplasty and keratoprosthesis

To compare visual and glaucoma outcomes in patients with known glaucoma after a penetrating keratoplasty (PKP) or a Boston Keratoprosthesis Type 1 (KPro) as a second corneal replacement procedure. Retrospective interventional case series. Charts of 141 eyes that underwent either a PKP or KPro at the Centre hospitalier de l'Université de Montréal after one failed PKP from 2008 to 2020 were reviewed. Forty-six eyes with preoperative glaucoma were included. Data collected included demographics, indication for the initial surgery, best corrected visual acuity (BCVA), concurrent ocular disorders, number of glaucoma medications, need for glaucoma surgery, cup-to-disc ratios (CDRs), mean RNFL thickness, and visual field (VF) characteristics. Primary outcomes were glaucoma progression trends. Secondary outcomes were visual outcomes and need for additional procedures. Mean follow-up was 4.7years for the PKP and 7.3 for the KPro group (P<0.007). 30.6% of PKP compared to 70.5% of KPro patients were diagnosed with glaucoma preoperatively. Glaucoma worsened similarly in both groups; this is based on an analysis of the number of glaucoma medications, CDR, need for glaucoma surgery, and characteristic VF changes. Patients in the PKP group required significantly more regrafts than patients in the KPro group (31.8 vs. 8.3%; P=0.045). A preoperative diagnosis of glaucoma does not preclude KPro implantation. In glaucomatous eyes, the disease progressed similarly in both groups. Since both procedures increase the risk of worsening glaucoma, close follow-up is recommended. KPro may decrease the need for further corneal transplantation surgery.

Detection and agreement of event-based OCT and OCTA analysis for glaucoma progression.

To examine event-based glaucoma progression using optical coherence tomography (OCT) and OCT angiography (OCTA). In this retrospective study, glaucoma eyes with ≥2-year and 4-visits of OCT/OCTA imaging were included. Peripapillary capillary density (CD) and retinal nerve fibre layer thickness (RNFL) were obtained from 4.5 mm × 4.5 mm optic nerve head (ONH) scans. Event-based OCT/OCTA progression was defined as decreases in ONH measurements exceeding test-retest variability on ≥2 consecutive visits. Visual field (VF) progression was defined as significant VF mean deviation worsening rates on ≥2 consecutive visits. Inter-instrument agreement on progression detection was compared using kappa(κ) statistics. Among 147 eyes (89 participants), OCTA and OCT identified 33(22%) and 25(17%) progressors, respectively. They showed slight agreement (κ = 0.06), with 7(5%) eyes categorized as progressors by both. When incorporating both instruments, the rate of progressors identified increased to 34%. Similar agreement was observed in diagnosis- and severity-stratified analyses (κ < 0.10). Compared to progressors identified only by OCT, progressors identified only by OCTA tended to have thinner baseline RNFL and worse baseline VF. VF progression was identified in 11(7%) eyes. OCT and VF showed fair agreement (κ = 0.26), with 6(4%) eyes categorized as progressors by both. OCTA and VF showed slight agreement (κ = 0.08), with 4(3%) eyes categorized as progressors by both. OCT and OCTA showed limited agreement on event-based progression detection, with OCT showing better agreement with VF. Both OCT and OCTA detected more progressors than VF. OCT and OCTA may provide valuable, yet different and complementary, information about glaucoma progression.

Fellow-Eye Comparison between Phaco-Tanito Microhook Trabeculotomy and Phaco-iStent Inject W.

This study aims to compare the surgical efficacy and safety of the Tanito microhook trabeculotomy (TMH-CE) and iStent inject W (Inject-CE) when performed in combination with cataract surgery on the eyes of glaucoma patients. A total of 78 glaucomatous eyes from 39 participants were retrospectively analyzed. Intraocular pressure (IOP), the number of antiglaucoma medications, best-corrected visual acuity (BCVA), anterior chamber flare (ACF), and corneal endothelial cell density (CECD) were all evaluated preoperatively and at multiple postoperative time points. The preoperative IOP was significantly higher in the TMH-CE (19.6 ± 6.7 mmHg) than in the Inject-CE (15.7 ± 3.8 mmHg) (p < 0.0001). At the 12-month follow-up, reductions in IOP and the number of medications were more pronounced in the TMH-CE (6.6 mmHg, 27.6% and -1.1, respectively) group than Inject-CE (2.7 mmHg, 12.4% and -0.7, respectively) (p < 0.0001 and p = 0.0034), while the IOP and medication-number levels were identical between TMH-CE (13.0 ± 3.3 mmHg and 1.3 ± 0.9, respectively) and Inject-CE (12.9 ± 2.6 mmHg and 1.9 ± 0.9, respectively) (p = 0.88 and p > 0.99, respectively). The TMH-CE group exhibited a higher ACF, a higher frequency of layered hyphema, and a greater anterior chamber floating red blood cells score in the early postoperative periods. Despite these differences, the changes in BCVA, ACF, and CECD were equivalent between the two groups in later follow-up periods. TMH-CE provides a more significant IOP reduction and medication-number reduction compared to Inject-CE, while Inject-CE shows quicker BCVA recovery. This study provides valuable insights for ophthalmologists choosing the most suitable surgical approach for glaucoma and cataract patients.

Controlled release, chitosan-tethered luteolin phytocubosomes; Formulation optimization to in-vivo antiglaucoma and anti-inflammatory ocular evaluation

A chitosan-coated luteolin-loaded phytocubosomal system was prepared to improve the pharmacodynamic performance of luteolin in the treatment of glaucoma and ocular inflammation after topical ocular administration. Luteolin, a potent anti-oxidant herbal drug with poor aqueous solubility, was complexed with phospholipid. The prepared phytocubosomes were coated with chitosan, producing homogenously distributed nanosized particles (258 ± 9.05 nm) with a positive charge (+49 ± 6.09 mV), improved EE% (96 %), and increased concentration of encapsulated drug to 288 μg/ml. Polarized light microscopy revealed a cubic phase. Chitosan-coated phytocubosomes showed a sustained drug release profile (38 % over 24 h) and improved anti-oxidant activity (IC50 of 32 μg/ml). Ex vivo transcorneal permeation was higher by 3.60 folds compared to luteolin suspension. Irritancy tests confirmed their safety in ocular tissues after single and multiple administrations. The pharmacodynamic studies on glaucomatous rabbit eyes demonstrated 6.46-, 3.88-, and 1.89-fold reductions in IOP of chitosan-coated phytocubosomes compared to luteolin suspension, cubosomes, and phytocubosomes, respectively. Pharmacodynamic anti-inflammatory studies revealed faster recovery capabilities of chitosan-coated phytocubosomes over other formulations. Thus, chitosan-coated phytocubosomes could be a promising ocular hybrid system for delivering herbal lipophilic drugs such as luteolin.

Segmental biomechanics of the normal and glaucomatous human aqueous outflow pathway

The conventional aqueous outflow pathway, encompassing the trabecular meshwork (TM), juxtacanalicular connective tissue (JCT), and inner wall endothelium of Schlemm's canal (SC), governs intraocular pressure (IOP) regulation. This study targets the biomechanics of low-flow (LF) and high-flow (HF) regions within the aqueous humor outflow pathway in normal and glaucomatous human donor eyes, using a combined experimental and computational approach. LF and HF TM/JCT/SC complex tissues from normal and glaucomatous eyes underwent uniaxial tensile testing. Dynamic motion of the TM/JCT/SC complex was recorded using customized green-light optical coherence tomography during SC pressurization in cannulated anterior segment wedges. A hyperviscoelastic model quantified TM/JCT/SC complex properties. A fluid-structure interaction model simulated tissue-aqueous humor interaction. FluoSpheres were introduced into the pathway via negative pressure in the SC, with their motion tracked using two-photon excitation microscopy. Tensile test results revealed that the elastic moduli of the LF and HF regions in glaucomatous eyes are 3.5- and 1.5-fold stiffer than the normal eyes, respectively. The FE results also showed significantly larger shear moduli in the TM, JCT, and SC of the glaucomatous eyes compared to the normal subjects. The LF regions in normal eyes demonstrated larger elastic moduli compared to the HF regions in glaucomatous eyes. The resultant strain in the outflow tissues and velocity of the aqueous humor in the FSI models were in good agreement with the digital volume correlation and 3D particle image velocimetry data, respectively. This study uncovers stiffer biomechanical responses in glaucomatous eyes, with LF regions stiffer than HF regions in both normal and glaucomatous eyes. Statement of significanceThis study delves into the biomechanics of the conventional aqueous outflow pathway, a crucial regulator of intraocular pressure and ocular health. By analyzing mechanical differences in low-flow and high-flow regions of normal and glaucomatous eyes, this research unveils the stiffer response in glaucomatous eyes. The distinction between regions' properties offers insights into aqueous humor outflow regulation, while the integration of experimental and computational methods enhances credibility. These findings have potential implications for disease management and present a vital step toward innovative ophthalmic interventions. This study advances our understanding of glaucoma's biomechanical basis and its broader impact on ocular health.

Interpretable surrogate models to approximate the predictions of convolutional neural networks in glaucoma diagnosis

Deep learning systems, especially in critical fields like medicine, suffer from a significant drawback, their black box nature, which lacks mechanisms for explaining or interpreting their decisions. In this regard, our research aims to evaluate the use of surrogate models for interpreting convolutional neural network (CNN) decisions in glaucoma diagnosis. Our approach is novel in that we approximate the original model with an interpretable one and also change the input features, replacing pixels with tabular geometric features of the optic disc, cup, and neuroretinal rim. We trained CNNs with two types of images: original images of the optic nerve head and simplified images showing only the disc and cup contours on a uniform background. Decision trees were used as surrogate models due to their simplicity and visualization properties, while saliency maps were calculated for some images for comparison. The experiments carried out with 1271 images of healthy subjects and 721 images of glaucomatous eyes demonstrate that decision trees can closely approximate the predictions of neural networks trained on simplified contour images, with R-squared values near 0.9 for VGG19, Resnet50, InceptionV3 and Xception architectures. Saliency maps proved difficult to interpret and showed inconsistent results across architectures, in contrast to the decision trees. Additionally, some decision trees trained as surrogate models outperformed a decision tree trained on the actual outcomes without surrogation. Decision trees may be a more interpretable alternative to saliency methods. Moreover, the fact that we matched the performance of a decision tree without surrogation to that obtained by decision trees using knowledge distillation from neural networks is a great advantage since decision trees are inherently interpretable. Therefore, based on our findings, we think this approach would be the most recommendable choice for specialists as a diagnostic tool.

Usability of an Eye Drop Delivery Aid for Single-Dose Instillation: Results from a Market Research Study.

One of the challenges of treating chronic ocular diseases like vernal keratoconjunctivitis (VKC), glaucoma, and ocular surface disease is patient adherence to topical medication. To support correct eye drop instillation, a variety of delivery aids have been developed for both single-dose and conventional multi-dose containers. To evaluate Dropaid™ Single-dose, an eye drop delivery aid designed for single-dose containers, a usability study was conducted on 30 parents and caregivers of patients with VKC. After assessing the ability to squeeze a single eye drop from the single-dose container onto the eye of a pediatric medical dummy, the delivery aid was evaluated using an 11-point Likert scale on a variety of characteristics: from "very difficult" (-5) to "very easy" (+5). The majority of participants rated the task of opening the single-dose units (SDUs), preparing, and positioning the Dropaid™ Single-dose device as "very easy". When providing a single eye drop from the container, 87% of participants rated the Dropaid™ Single-dose device as either "very easy" or "easy", with a median rating score of +5.0 (interquartile range [IQR], 4.0-5.0). For general ease of use, 84% considered the delivery aid as either "very easy" or "easy", with a median score of +4.0 (IQR, 3.0-5.0). Most participants (93%) rated Dropaid™ Single-dose as either "very comfortable" or "comfortable" to hold, with a median score of +5.0 (IQR, 4.0-5.0). The Dropaid™ Single-dose delivery aid demonstrated rapid learning and ease of use across all stages of application, including opening the container, eye drop administration, and handling comfort. Although designed for use with single-dose containers to help eye drop instillation in patients with VKC, Dropaid™ Single-dose may provide a wider utility across a range of other ocular diseases such as glaucoma and dry eye.

The associations and mediators between visual disabilities and anxiety disorders in middle-aged and older adults: A population-based study.

Visual disabilities significantly impact an individual's mental health. Little is known about the prospective relationship between visual disabilities and anxiety disorders and the underlying effects of modifiable risk factors. Our analysis was based on 117,252 participants from the U.K. Biobank, with baseline data collected between 2006 and 2010. Habitual visual acuity was measured by a standardized logarithmic chart, and ocular disorders reported using questionnaires were collected at baseline. Incident hospitalized anxiety recorded using longitudinal linkage with hospital inpatient data, lifetime anxiety disorder, and current anxiety symptoms assessed by a comprehensive online mental health questionnaire were identified over a 10-year follow-up. After adjustments for confounding factors, one-line worse visual acuity (0.1 logarithm of the minimum angle of resolution [logMAR]) was associated with an increased risk of incident hospitalized anxiety (HR = 1.05, 95% CI = 1.01-1.08), lifetime anxiety disorder (OR = 1.07, 95% CI [1.01-1.12]), and current anxiety scores (β = 0.028, 95% CI [0.002-0.054]). Besides poorer visual acuity, the longitudinal analysis also supported that each ocular disorder (including cataracts, glaucoma, macular degeneration, and diabetes-related eye disease) was significantly associated with at least two anxiety outcomes. Mediation analyses highlighted that subsequent onsets of eye diseases, especially cataracts, and lower socioeconomic status (SES) partly mediated the association between poorer visual acuity and anxiety disorders. This study demonstrates an overall association between visual disabilities and anxiety disorders in middle-aged and older adults. In particular, early interventions involving treatments for visual disabilities and effective psychological counseling services sensitive to socioeconomic status may help prevent anxiety in those living with poor vision. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Addressing Glaucoma in Myopic Eyes: Diagnostic and Surgical Challenges.

Epidemiological and genetic studies provide strong evidence supporting an association between myopia and glaucoma. The accurate detection of glaucoma in myopic eyes, especially those with high myopia, remains clinically challenging due to characteristic morphologic features of the myopic optic nerve in addition to limitations of current optic nerve imaging modalities. Distinguishing glaucoma from myopia is further complicated by overlapping perimetric findings. Therefore, longitudinal follow-up is essential to differentiate progressive structural and functional abnormalities indicative of glaucoma from defects that may result from myopia alone. Highly myopic eyes are at increased risk of complications from traditional incisional glaucoma surgery and may benefit from newer microinvasive glaucoma surgeries in select cases.

Vessel Density Features of Optical Coherence Tomography Angiography for Classification of Glaucoma Using Machine Learning.

Machine learning (ML) based on the optical coherence tomography angiography vessel density features with different thresholds using a support vector machine (SVM) model provides excellent performance for glaucoma detection. To assess the classification performance of ML based on the 4 vessel density features of peripapillary optical coherence tomography angiography for glaucoma detection. Images from 119 eyes of 119 glaucoma patients and 76 eyes of 76 healthy individuals were included. Four vessel density features of optical coherence tomography angiography images were developed using a threshold-based segmentation method and were integrated into 3 models of machine learning classifiers. Images were divided into 70% training set and 30% test set. Classification performances of SVM, random forest, and Gaussian Naive Bayes models were evaluated with the area under the receiver operating characteristic curve (AUC) and accuracy. Glaucoma eyes had lower vessel densities at different thresholds. For differentiating glaucoma eyes, the best results were achieved with 70% and 100% thresholds, in which SVM classifier discriminated glaucoma from healthy eyes with an AUC of 1 and accuracy of 1. After SVM, the random forest classifier with 100% thresholds showed an AUC of 0.993 and an accuracy of 0.994. Furthermore, the AUC of our ML performance (SVM) was 0.96 in a subgroup analysis of mild and moderate glaucoma eyes. ML based on the combined peripapillary vessel density features of total vessels and capillaries in the whole image and ring image could provide excellent performance for glaucoma detection.

Classifying glaucoma exclusively with OCT: comparison of three clustering algorithms derived from machine learning.

To objectively classify eyes as either healthy or glaucoma based exclusively on data provided by peripapillary retinal nerve fiber layer (pRNFL) and ganglion cell-inner plexiform (GCIPL) measurements derived from spectral-domain optical coherence tomography (SD-OCT) using machine learning algorithms. Three clustering methods (k-means, hierarchical cluster analysis -HCA- and model-based clustering-MBC-) were used separately to classify a training sample of 109 eyes as either healthy or glaucomatous using solely 13 SD-OCT parameters: pRNFL average and sector thicknesses and GCIPL average and minimum values together with the six macular wedge-shaped regions. Then, the best-performing algorithm was applied to an independent test sample of 102 eyes to derive close estimates of its actual performance (external validation). In the training sample, accuracy was 91.7% for MBC, 81.7% for k-means and 78.9% for HCA (p value = 0.02). The best MBC model was that in which subgroups were allowed to have variable volume and shape and equal orientation. The MBC algorithm in the independent test sample correctly classified 98 out of 102 cases for an overall accuracy of 96.1% (95% CI, 92.3-99.8%), with a sensitivity of 94.3 and 100% specificity. The accuracy for pRNFL was 92.2% (95% CI, 86.9-97.4%) and for GCIPL 98.0% (95% CI, 95.3-100%). Clustering algorithms in general (and MBC in particular) seem promising methods to help discriminate between healthy and glaucomatous eyes using exclusively SD-OCT-derived parameters. Understanding the relative merits of one method over others may also provide insights into the nature of the disease.

Focused analysis of RNFL decay in glaucomatous eyes using circular statistics on high-resolution OCT data.

We generated Optical Coherence Tomography (OCT) data of much higher resolution than usual on retinal nerve fiber layer (RNFL) thickness of a given eye. These consist of measurements made at hundreds of angular-points defined on a circular coordinate system. Traditional analysis of OCT RNFL data does not utilize insightful characteristics such as its circularity and granularity for common downstream applications. To address this, we present a new circular statistical framework that defines an Angular Decay function and thereby provides a directionally precise representation of an eye with attention to patterns of focused RNFL loss. By applying to a clinical cohort of Asian Indian eyes, the generated circular data were modeled with a finite mixture of von Mises distributions, which led to an unsupervised identification in different age-groups of recurrent clusters of glaucomatous eyes with distinct directional signatures of RNFL decay. New indices of global and local RNFL loss were computed for comparing the structural differences between these glaucoma clusters across the age-groups and improving classification. Further, we built a catalog of directionally precise statistical distributions of RNFL thickness for the said population of normal eyes as stratified by their age and optic disc size.