Search

PurposeTo compare convolutional neural network (CNN) analysis of en face vessel density images to gradient boosting classifier (GBC) analysis of instrument-provided, feature-based optical coherence tomography angiography (OCTA) vessel density measurements and OCT retinal nerve fiber layer (RNFL) thickness measurements for classifying healthy and glaucomatous eyes.DesignComparison of diagnostic approaches.MethodsA total of 130 eyes of 80 healthy individuals and 275 eyes of 185 glaucoma patients with optic nerve head (ONH) OCTA and OCT imaging were included. Classification performance of a VGG16 CNN trained and tested on entire en face 4.5×4.5-mm radial peripapillary capillary OCTA ONH images was compared to the performance of separate GBC models trained and tested on standard OCTA and OCT measurements. Five-fold cross-validation was used to test predictions for CNNs and GBCs. Areas under the precision recall curves (AUPRC) were calculated to control for training/test set size imbalance and were compared.ResultsAdjusted AUPRCs for GBC models were 0.89 (95% CI=0.82, 0.92) for whole image vessel density GBC, 0.89 (0.83, 0.92) for whole image capillary density GBC, 0.91 (0.88, 0.93) for combined whole image vessel and whole image capillary density GBC, and 0.93 (0.91, 095) for RNFL thickness GBC. The adjusted AUPRC using CNN analysis of en face vessel density images was 0.97 (0.95, 0.99) resulting in significantly improved classification compared to GBC OCTA-based results and GBC OCT-based results (P ≤ 0.01 for all comparisons).ConclusionDeep learning en face image analysis improves on feature-based GBC models for classifying healthy and glaucoma eyes.

Uveal melanoma, the most common intraocular primary cancer in adults, is characterized by striking variability in metastatic tendencies. BAP1 deletion in the primary tumor is associated with uveal melanoma metastasis, but it cannot always be resolved by bulk DNA sequencing of heterogeneous tumors. Here, we show that assessment of BAP1 methylation is an accurate and readily clinically actionable assay to accurately identify high-risk uveal melanoma patients.

Purpose To compare gradient-boosting classifier (GBC) analysis of optical coherence tomography angiography (OCTA)-measured vessel density (VD) and OCT-measured tissue thickness to standard OCTA VD and OCT thickness parameters for classifying healthy eyes and eyes with early to moderate glaucoma. Design Comparison of diagnostic tools. Methods A total of 180 healthy eyes and 193 glaucomatous eyes with OCTA and OCT imaging of the macula and optic nerve head (ONH) were studied. Four GBCs were evaluated that combined 1) all macula VD and thickness measurements (Macula GBC), 2) all ONH VD and thickness measurements (ONH GBC), 3) all VD measurements from the macula and ONH (vessel density GBC), and 4) all thickness measurements from the macula and ONH (thickness GBC). ROC curve (AUROC) analyses compared the diagnostic accuracy of GBCs to that of standard instrument-provided parameters. A fifth GBC that combined all parameters (full GBC) also was investigated. Results GBCs had better diagnostic accuracy than standard OCTA and OCT parameters with AUROCs ranging from 0.90 to 0.93 and 0.64 to 0.91, respectively. The full GBC (AUROC = 0.93) performed significantly better than the ONH GBC (AUROC = 0.91; P = .036) and the vessel density GBC (AUROC = 0.90; P = .010). All other GBCs performed similarly. The mean relative influence of each parameter included in the full GBC identified a combination of macular thickness and ONH VD measurements as the greatest contributors. Conclusions GBCs that combine OCTA and OCT macula and ONH measurements can improve diagnostic accuracy for glaucoma detection compared to most but not all instrument provided parameters.

Purpose To assess the accuracy of deep learning models to predict glaucoma development from fundus photographs several years before disease onset. Design Algorithm development for predicting glaucoma using data from a prospective longitudinal study. Participants A total of 66 721 fundus photographs from 3272 eyes of 1636 subjects who participated in the Ocular Hypertension Treatment Study (OHTS) were included. Main Outcome Measures Accuracy and area under the curve (AUC). Methods Fundus photographs and visual fields were carefully examined by 2 independent readers from the optic disc and visual field reading centers of the OHTS. When an abnormality was detected by the readers, the subject was recalled for retesting to confirm the abnormality and for further confirmation by an end point committee. By using 66 721 fundus photographs, deep learning models were trained and validated using 85% of the fundus photographs and further retested (validated) on the remaining (held-out) 15% of the fundus photographs. Results The AUC of the deep learning model in predicting glaucoma development 4 to 7 years before disease onset was 0.77 (95% confidence interval [CI], 0.75–0.79). The accuracy of the model in predicting glaucoma development approximately 1 to 3 years before disease onset was 0.88 (95% CI, 0.86–0.91). The accuracy of the model in detecting glaucoma after onset was 0.95 (95% CI, 0.94–0.96). Conclusions Deep learning models can predict glaucoma development before disease onset with reasonable accuracy. Eyes with visual field abnormality but not glaucomatous optic neuropathy had a higher tendency to be missed by deep learning algorithms.

A 68-year-old woman with a recent history of blurring in the left eye had undergone mastectomy for breast cancer 20 years ago. A series of bone metastases started 5 years after her diagnosis. Examination of the optic nerve head of the left eye revealed an isolated peripapillary mass. Indocyanine green angiography displayed vessels within the mass, and fluorescein angiography demonstrated hyperfluorescence of the mass from vascular leakage plus lobular spots of blocked fluorescence. B-scan ultrasound revealed a hyperechoic-elevated nodular mass on the optic disc. Spectral-domain optical coherence tomography displayed a mass of spherules. Magnetic resonance imaging of the brain demonstrated metastatic tumors. She was diagnosed with an optic disk metastasis from her breast carcinoma.

Chromosomal instability (CIN) and epigenetic alterations have been implicated in tumor progression and metastasis; yet how these two hallmarks of cancer are related remains poorly understood. By integrating genetic, epigenetic, and functional analyses at the single cell level, we show that progression of uveal melanoma (UM), the most common intraocular primary cancer in adults, is driven by loss of Polycomb Repressive Complex 1 (PRC1) in a subpopulation of tumor cells. This leads to transcriptional de-repression of PRC1-target genes and mitotic chromosome segregation errors. Ensuing CIN leads to the formation of rupture-prone micronuclei, exposing genomic double-stranded DNA (dsDNA) to the cytosol. This provokes tumor cell-intrinsic inflammatory signaling, mediated by aberrant activation of the cGAS-STING pathway. PRC1 inhibition promotes nuclear enlargement, induces a transcriptional response that is associated with significantly worse patient survival and clinical outcomes, and enhances migration that is rescued upon pharmacologic inhibition of CIN or STING. Thus, deregulation of PRC1 can promote tumor progression by inducing CIN and represents an opportunity for early therapeutic intervention., The molecular underpinnings driving uveal melanoma (UM) progression are unknown. Here the authors show that loss of Polycomb Repressive Complex 1 triggers chromosomal instability, which promotes inflammatory signaling and migration in UM.

To investigate the diagnostic accuracy of deep learning (DL) algorithms to detect primary open-angle glaucoma (POAG) trained on fundus photographs from the Ocular Hypertension Treatment Study (OHTS). 66,715 photographs from 3,272 eyes were used to train and test a ResNet-50 model to detect the OHTS Endpoint Committee POAG determination based on optic disc (n=287 eyes, 3,502 photographs) and/or visual field (n=198 eyes, 2,300 visual fields) changes. OHTS training, validation and testing sets were randomly determined using an 85-5-10 percentage split by subject. Three independent test sets were used to estimate the generalizability of the model: UCSD Diagnostic Innovations in Glaucoma Study (DIGS, USA), ACRIMA (Spain) and Large-scale Attention-based Glaucoma (LAG, China). The DL model achieved an AUROC (95% CI) of 0.88 (0.82, 0.92) for the overall OHTS POAG endpoint. For the OHTS endpoints based on optic disc changes or visual field changes, AUROCs were 0.91 (0.88, 0.94) and 0.86 (0.76, 0.93), respectively. False-positive rates (at 90% specificity) were higher in photographs of eyes that later developed POAG by disc or visual field (19.1%) compared to eyes that did not develop POAG (7.3%) during their OHTS follow-up. The diagnostic accuracy of the DL model developed on the OHTS optic disc endpoint applied to 3 independent datasets was lower with AUROC ranging from 0.74 to 0.79. High diagnostic accuracy of the current model suggests that DL can be used to automate the determination of POAG for clinical trials and management. In addition, the higher false-positive rate in early photographs of eyes that later developed POAG suggests that DL models detected POAG in some eyes earlier than the OHTS Endpoint Committee.

Author(s): Bowd, Christopher; Belghith, Akram; Christopher, Mark; Goldbaum, Michael H; Fazio, Massimo A; Girkin, Christopher A; Liebmann, Jeffrey M; de Moraes, Carlos Gustavo; Weinreb, Robert N; Zangwill, Linda M | Abstract: PurposeTo compare change over time in eye-specific optical coherence tomography (OCT) retinal nerve fiber layer (RNFL)-based region-of-interest (ROI) maps developed using unsupervised deep-learning auto-encoders (DL-AE) to circumpapillary RNFL (cpRNFL) thickness for the detection of glaucomatous progression.MethodsForty-four progressing glaucoma eyes (by stereophotograph assessment), 189 nonprogressing glaucoma eyes (by stereophotograph assessment), and 109 healthy eyes were followed for ≥3 years with ≥4 visits using OCT. The San Diego Automated Layer Segmentation Algorithm was used to automatically segment the RNFL layer from raw three-dimensional OCT images. For each longitudinal series, DL-AEs were used to generate individualized eye-based ROI maps by identifying RNFL regions of likely progression and no change. Sensitivities and specificities for detecting change over time and rates of change over time were compared for the DL-AE ROI and global cpRNFL thickness measurements derived from a 2.22-mm to 3.45-mm annulus centered on the optic disc.ResultsThe sensitivity for detecting change in progressing eyes was greater for DL-AE ROIs than for global cpRNFL annulus thicknesses (0.90 and 0.63, respectively). The specificity for detecting not likely progression in nonprogressing eyes was similar (0.92 and 0.93, respectively). The mean rates of change in DL-AE ROI were significantly faster than for cpRNFL annulus thickness in progressing eyes (-1.28 µm/y vs. -0.83 µm/y) and nonprogressing eyes (-1.03 µm/y vs. -0.78 µm/y).ConclusionsEye-specific ROIs identified using DL-AE analysis of OCT images show promise for improving assessment of glaucomatous progression.Translational relevanceThe detection and monitoring of structural glaucomatous progression can be improved by considering eye-specific regions of likely progression identified using deep learning.

We develop an Artificial Intelligence (AI) framework for glaucoma precognition from baseline confocal scanning laser ophthalmoscopy imaging data, using a convolutional neural network (CNN) model. The proposed framework extracts ‘deep features’ from convolutional layers of the CNN model, which are used as input to the ensemble learning classifier in order to identify patients that will likely convert to glaucoma after few years. The prediction model achieved area under the receiver operating characteristic curve (AUC) of 0.83 using the data from baseline visit. The model predicted the onset of glaucoma more accurately than known glaucoma risk factors, Glaucoma Probability Score (GPS) and Moorfields Regression Analysis (MRA) parameters of the Heidelberg Retinal Tomograph (HRT) software. The proposed AI construct provides a highly specific and sensitive model that can predict the onset of glaucoma from baseline HRT parameters and has the potential to provide clinicians valuable information regarding the onset of glaucoma.

BackgroundCardiovascular disease is the leading cause of mortality and disability worldwide. A noninvasive test that can detect underlying cardiovascular disease has the potential to identify patients at risk prior to the occurrence of adverse cardiovascular events. We sought to determine whether an easily observed imaging finding indicative of retinal ischemia, which we term 'retinal ischemic perivascular lesions' (RIPLs), could serve as a biomarker for cardiovascular disease.MethodsWe reviewed optical coherence tomography (OCT) scans of individuals, with no underlying retinal pathology, obtained at UC San Diego Health from July 2014 to July 2019. We identified 84 patients with documented cardiovascular disease and 76 healthy controls. OCT scans were assessed for evidence of RIPLs. In addition, the 10-year atherosclerotic cardiovascular disease (ASCVD) risk calculator was used to risk-stratify the subjects into four different categories.FindingsPatients with documented cardiovascular disease had higher number of RIPLs compared to healthy controls (2.8vs 0.8, p37). The number of RIPLs in individuals with intermediate and high 10-year ASCVD risk scores was higher than in those with low ASCVD risk scores (1.7vs 0.64, p=0.02 and 2.9vs 0.64, p 0.002, respectively).InterpretationThe presence of RIPLs, which are anatomical markers of prior retinal ischemic infarcts, is suggestive of coexisting cardiovascular disease. RIPLs detection, obtained from routine retinal scans, may thus provide an additional biomarker to identify patients at risk of developing adverse cardiovascular events.FundingNone.

Purpose To develop deep learning (DL) systems estimating visual function from macula-centered spectral-domain (SD) OCT images. Design Evaluation of a diagnostic technology. Participants A total of 2408 10-2 visual field (VF) SD OCT pairs and 2999 24-2 VF SD OCT pairs collected from 645 healthy and glaucoma subjects (1222 eyes). Methods Deep learning models were trained on thickness maps from Spectralis macula SD OCT to estimate 10-2 and 24-2 VF mean deviation (MD) and pattern standard deviation (PSD). Individual and combined DL models were trained using thickness data from 6 layers (retinal nerve fiber layer [RNFL], ganglion cell layer [GCL], inner plexiform layer [IPL], ganglion cell-IPL [GCIPL], ganglion cell complex [GCC] and retina). Linear regression of mean layer thicknesses were used for comparison. Main Outcome Measures Deep learning models were evaluated using R2 and mean absolute error (MAE) compared with 10-2 and 24-2 VF measurements. Results Combined DL models estimating 10-2 achieved R2 of 0.82 (95% confidence interval [CI], 0.68–0.89) for MD and 0.69 (95% CI, 0.55–0.81) for PSD and MAEs of 1.9 dB (95% CI, 1.6–2.4 dB) for MD and 1.5 dB (95% CI, 1.2–1.9 dB) for PSD. This was significantly better than mean thickness estimates for 10-2 MD (0.61 [95% CI, 0.47–0.71] and 3.0 dB [95% CI, 2.5–3.5 dB]) and 10-2 PSD (0.46 [95% CI, 0.31–0.60] and 2.3 dB [95% CI, 1.8–2.7 dB]). Combined DL models estimating 24-2 achieved R2 of 0.79 (95% CI, 0.72–0.84) for MD and 0.68 (95% CI, 0.53–0.79) for PSD and MAEs of 2.1 dB (95% CI, 1.8–2.5 dB) for MD and 1.5 dB (95% CI, 1.3–1.9 dB) for PSD. This was significantly better than mean thickness estimates for 24-2 MD (0.41 [95% CI, 0.26–0.57] and 3.4 dB [95% CI, 2.7–4.5 dB]) and 24-2 PSD (0.38 [95% CI, 0.20–0.57] and 2.4 dB [95% CI, 2.0–2.8 dB]). The GCIPL (R2 = 0.79) and GCC (R2 = 0.75) had the highest performance estimating 10-2 and 24-2 MD, respectively. Conclusions Deep learning models improved estimates of functional loss from SD OCT imaging. Accurate estimates can help clinicians to individualize VF testing to patients.

Deep archetypal analysis (DAA) has recently been proposed as an unsupervised approach for discovering latent structures in data. However, while a few approaches have used classical archetypal analysis (AA), DAA has not been incorporated in medical image analysis as yet. The purpose of this study is to develop a precognition framework to identify preclinical signs of glaucomatous vision loss using convex representations derived from DAA. We first develop an AA structure and a novel DAA framework to recognize hidden patterns of visual functional loss, and then project visual field data over the identified patterns to obtain a representation for glaucoma precognition several years prior to disease onset. We then develop a glaucoma classification framework using class-balanced bagging with neural networks to address the class imbalance problem. In contrast to other classification approaches, DAA, applied to a unique prospective longitudinal dataset with approximately eight years of visual field tests from normal eyes that developed glaucoma, has allowed visualization of the early signs of glaucoma and development of a construct for glaucoma precognition. Our findings suggest that our proposed glaucoma precognition approach could significantly advance state-of-the-art glaucoma prediction.

G protein mutations are common in uveal melanomas, and the vast majority target amino acid residue Q209 in either GNAQ or GNA11. The GNAQ R183Q mutation is found in a small fraction of uveal melanomas. We report a patient with an unusual presentation of uveal melanoma arising at an early age in the setting of congenital skin and ocular surface melanosis. A 34-year-old Hispanic female with congenital bilateral nevus of Ota and ocular surface melanosis presented with progressive loss of visual acuity and was found to have a juxtapapillary uveal melanoma. She was treated with brachytherapy, but the tumor relapsed. She underwent enucleation that revealed mixed spindle and epithelioid uveal melanoma cells with no extraocular or lymphovascular spread. Next-generation sequencing performed on DNA isolated from the enucleation specimen identified a GNAQ R183Q mutation and a PMS1 truncation mutation. Cytogenetic profiling revealed no monosomy 3. These findings raise the possibility that uveal melanomas bearing G protein R183 mutations may have distinct clinicopathologic profiles compared to those with Q209 mutations. Furthermore, this is the first reported case of a mutation in the mismatch repair gene PMS1 associated with uveal melanoma.

Author(s): Christopher, Mark; Nakahara, Kenichi; Bowd, Christopher; Proudfoot, James A; Belghith, Akram; Goldbaum, Michael H; Rezapour, Jasmin; Weinreb, Robert N; Fazio, Massimo A; Girkin, Christopher A; Liebmann, Jeffrey M; De Moraes, Gustavo; Murata, Hiroshi; Tokumo, Kana; Shibata, Naoto; Fujino, Yuri; Matsuura, Masato; Kiuchi, Yoshiaki; Tanito, Masaki; Asaoka, Ryo; Zangwill, Linda M | Abstract: PurposeTo compare performance of independently developed deep learning algorithms for detecting glaucoma from fundus photographs and to evaluate strategies for incorporating new data into models.MethodsTwo fundus photograph datasets from the Diagnostic Innovations in Glaucoma Study/African Descent and Glaucoma Evaluation Study and Matsue Red Cross Hospital were used to independently develop deep learning algorithms for detection of glaucoma at the University of California, San Diego, and the University of Tokyo. We compared three versions of the University of California, San Diego, and University of Tokyo models: original (no retraining), sequential (retraining only on new data), and combined (training on combined data). Independent datasets were used to test the algorithms.ResultsThe original University of California, San Diego and University of Tokyo models performed similarly (area under the receiver operating characteristic curve = 0.96 and 0.97, respectively) for detection of glaucoma in the Matsue Red Cross Hospital dataset, but not the Diagnostic Innovations in Glaucoma Study/African Descent and Glaucoma Evaluation Study data (0.79 and 0.92; P l .001), respectively. Model performance was higher when classifying moderate-to-severe compared with mild disease (area under the receiver operating characteristic curve = 0.98 and 0.91; P l .001), respectively. Models trained with the combined strategy generally had better performance across all datasets than the original strategy.ConclusionsDeep learning glaucoma detection can achieve high accuracy across diverse datasets with appropriate training strategies. Because model performance was influenced by the severity of disease, labeling, training strategies, and population characteristics, reporting accuracy stratified by relevant covariates is important for cross study comparisons.Translational relevanceHigh sensitivity and specificity of deep learning algorithms for moderate-to-severe glaucoma across diverse populations suggest a role for artificial intelligence in the detection of glaucoma in primary care.

PURPOSE: To compare the characteristics of the retinal and choroidal lesions including choroidal nevus, choroidal melanoma and congenital hypertrophy of the retina pigment epithelium using conventional color fundus photography (CFP) and multicolor imaging (MCI). METHODS: The paired images of patients with retinal or choroidal lesions were assessed for the visibility of lesion’s border, halo and drusen using a grading scale (0–2). The area of the lesion was measured on both imaging modalities. The same grading was also done on the individual color channels of MCI for a further evaluation. RESULTS: Thirty-three eyes of 33 patients were included. There were no significant differences in the mean border, drusen and halo visibility scores between the two imaging modalities (p=0.12, p=0.70, p=0.35). However, the mean area of the lesion was significantly smaller on MCI than that on CFP (14.9±3.3 versus 18.7±3.4 mm(2), p=0.01). CONCLUSION: The appearance of choroidal and/ or retinal lesions on MCI may be different than that on CFP. Though MCI can provide similar information with CFP for the features of retinal and/ or choroidal lesions including border, halo and drusen; the infrared light reflection on MCI underestimates the extent of the choroidal lesion by 33%.

Purpose Uveal melanomas are associated with characteristic genetic changes. Germline mutations in mismatch repair (MMR) genes and microsatellite instability have been implicated in the development of numerous malignant neoplasms such as colon and ovarian cancers. The frequency of MMR defects in uveal melanomas has yet to be determined. Methods Here, we analyzed the frequency of MMR gene mutations in uveal melanoma specimens from the University of California, San Diego (UCSD), The Cancer Genome Atlas (TGCA), and the Catalogue of Somatic Mutations in Cancer (COSMIC). Results We identified only two mutations in a MMR gene: one premature stop codon in the PMS gene within the UCSD cohort (0.5% frequency) and one in-frame deletion in MSH3 within the COSMIC database (0.8% frequency). We report copy number variation of MLH1 in monosomy 3 and show decreased mRNA expression of MLH1 in uveal melanoma specimens with monosomy 3. Expression levels of MLH1 were not found to correlate with the observed number of total mutations. Conclusion Overall, we show that mutations in MMR genes in uveal melanoma specimens are exceedingly rare, and although one copy of MLH1 is lost in monosomy 3, it does not seem to have pathologic consequences in uveal melanoma pathogenesis.

Background Many CpGs become hyper or hypo-methylated with age. Multiple methods have been developed by Horvath et al. to estimate DNA methylation (DNAm) age including Pan-tissue, Skin & Blood, PhenoAge, and GrimAge. Pan-tissue and Skin & Blood try to estimate chronological age in the normal population whereas PhenoAge and GrimAge use surrogate markers associated with mortality to estimate biological age and its departure from chronological age. Here, we applied Horvath’s four methods to calculate and compare DNAm age in 499 subjects with type 1 diabetes (T1D) from the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) study using DNAm data measured by Illumina EPIC array in the whole blood. Association of the four DNAm ages with development of diabetic complications including cardiovascular diseases (CVD), nephropathy, retinopathy, and neuropathy, and their risk factors were investigated. Results Pan-tissue and GrimAge were higher whereas Skin & Blood and PhenoAge were lower than chronological age (p < 0.0001). DNAm age was not associated with the risk of CVD or retinopathy over 18–20 years after DNAm measurement. However, higher PhenoAge (β = 0.023, p = 0.007) and GrimAge (β = 0.029, p = 0.002) were associated with higher albumin excretion rate (AER), an indicator of diabetic renal disease, measured over time. GrimAge was also associated with development of both diabetic peripheral neuropathy (OR = 1.07, p = 9.24E−3) and cardiovascular autonomic neuropathy (OR = 1.06, p = 0.011). Both HbA1c (β = 0.38, p = 0.026) and T1D duration (β = 0.01, p = 0.043) were associated with higher PhenoAge. Employment (β = − 1.99, p = 0.045) and leisure time (β = − 0.81, p = 0.022) physical activity were associated with lower Pan-tissue and Skin & Blood, respectively. BMI (β = 0.09, p = 0.048) and current smoking (β = 7.13, p = 9.03E−50) were positively associated with Skin & Blood and GrimAge, respectively. Blood pressure, lipid levels, pulse rate, and alcohol consumption were not associated with DNAm age regardless of the method used. Conclusions Various methods of measuring DNAm age are sub-optimal in detecting people at higher risk of developing diabetic complications although some work better than the others.

Goal: The purpose of this study was to identify clinically relevant patterns of glaucomatous vision loss through convex representation to predict glaucoma several years prior to disease onset. Methods: We developed a deep archetypal analysis to identify patterns of glaucomatous vision loss, and then projected visual fields over the identified patterns. Projections provided a representation that was more accurate in detecting glaucomatous vision loss, thus, more appropriate for recognizing preclinical signs of glaucoma prior to disease development. To overcome the class imbalance in prediction, we implemented a class-balanced bagging with neural networks. Results: Using original visual field as features of the class-balanced bagging classification provided an area under the receiver-operating characteristic curve (AUC) of 0.55 for predicting glaucoma approximately four years prior to disease development. Using convex representation of the visual fields as input features provided an AUC of 0.61 while using deep convex representation as input features improved the AUC to 0.71. Relevance vector machine (RVM) achieved an AUC of 0.64. Conclusion: Deep archetypal analysis representation of visual functional features with balanced bagging classification could serve as an automated tool for predicting glaucoma. Significance: Glaucoma is the second leading cause of worldwide blindness. Most people with glaucoma have no early symptoms or pain, delaying diagnosis in many patients until they reach late irreversible vision loss stages. In fact, about 50% of people with glaucoma are unaware they have the disease. Deep archetypal analysis models may impact clinical practice in effectively identifying at-risk glaucoma patients well prior to disease development., Glaucoma is the second leading cause of worldwide blindness. Most people with glaucoma have no early symptoms or pain, delaying diagnosis in many patients until they have irreversible vision loss. In fact, about 50% of people with glaucoma are unaware they have the disease. Deep archetypal analysis models may impact clinical practice in effectively identifying at-risk glaucoma patients well prior to disease development.

PurposeTo assess the accuracy of deep learning models to predict glaucoma development from fundus photographs several years prior to disease onset.DesignA deep learning model for prediction of glaucomatous optic neuropathy or visual field abnormality from color fundus photographs.ParticipantsWe retrospectively included 66,721 fundus photographs from 3,272 eyes of 1,636 subjects to develop deep leaning models.MethodFundus photographs and visual fields were carefully examined by two independent readers from the optic disc and visual field reading centers of the ocular hypertension treatment study (OHTS). When an abnormality was detected by the readers, subject was recalled for re-testing to confirm the abnormality and further confirmation by an endpoint committee. Using OHTS data, deep learning models were trained and tested using 85% of the fundus photographs and further validated (re-tested) on the remaining (held-out) 15% of the fundus photographs.Main Outcome MeasuresAccuracy and area under the receiver-operating characteristic curve (AUC).ResultsThe AUC of the deep learning model in predicting glaucoma development 4-7 years prior to disease onset was 0.77 (95% confidence interval 0.75, 0.79). The accuracy of the model in predicting glaucoma development about 1-3 years prior to disease onset was 0.88 (0.86, 0.91). The accuracy of the model in detecting glaucoma after onset was 0.95 (0.94, 0.96).ConclusionsDeep learning models can predict glaucoma development prior to disease onset with reasonable accuracy. Eyes with visual field abnormality but not glaucomatous optic neuropathy had a higher tendency to be missed by deep learning algorithms.

Tuberous sclerosis or tuberous sclerosis complex (TSC), one of the phakomatoses, is characterized by hamartomas of the heart, kidney, brain, skin and eyes. Ophthalmologic examinations are required in all cases of TSC. Retinal hamartomas are the most common ocular finding in tuberous sclerosis. The majority of hamartomas are non-progressive; however, lesions with subretinal fluid and progression have been reported. This paper details the genetics, clinical features and ocular findings of TSC and reviews potential therapeutic options for ophthalmic manifestations.

A hierarchical approach to learn from visual field data was adopted to identify glaucomatous visual field defect patterns and to detect glaucomatous progression. The analysis pipeline included three stages, namely, clustering, glaucoma boundary limit detection, and glaucoma progression detection testing. First, cross-sectional visual field tests collected from each subject were clustered using a mixture of Gaussians and model parameters were estimated using expectation maximization. The visual field clusters were further estimated to recognize glaucomatous visual field defect patterns by decomposing each cluster into several axes. The glaucoma visual field defect patterns along each axis then were identified. To derive a definition of progression, the longitudinal visual fields of stable glaucoma eyes on the abnormal cluster axes were projected and the slope was approximated using linear regression (LR) to determine the confidence limit of each axis. For glaucoma progression detection, the longitudinal visual fields of each eye on the abnormal cluster axes were projected and the slope was approximated by LR. Progression was assigned if the progression rate was greater than the boundary limit of the stable eyes; otherwise, stability was assumed. The proposed method was compared to a recently developed progression detection method and to clinically available glaucoma progression detection software. The clinical accuracy of the proposed pipeline was as good as or better than the currently available methods. © 1964-2012 IEEE.

PurposeTo determine whether there is an association between hepatic lipase (LIPC) and age-related macular degeneration (AMD) in two independent Caucasian cohorts.MethodsA discovery cohort of 1626 patients with advanced AMD and 859 normal controls and a replication cohort of 2159 cases and 1150 controls were genotyped for two single-nucleotide polymorphisms (SNPs) in the promoter region of LIPC. The associations between the SNPs and AMD were examined by χ(2) tests.ResultsIn the discovery cohort, rs493258 and rs10468017 were both associated with advanced AMD (P=9.63E-3 and P=0.048, respectively). The association was corroborated in the replication cohort (P=4.48E-03 for rs493258 and P=0.015 for rs10468017). Combined analysis resulted in even more significant associations (P=1.21E-04 for rs493258 and P=1.67E-03 for rs10468017).ConclusionThe LIPC promoter variants rs493258 and rs10468017 were associated with advanced AMD in two independent Caucasian populations, confirming that LIPC polymorphisms may be a genetic risk factor for AMD in the Caucasian population.

Purpose To validate Gaussian mixture-model with expectation maximization (GEM) and variational Bayesian independent component analysis mixture-models (VIM) for detecting glaucomatous progression along visual field (VF) defect patterns (GEM–progression of patterns (POP) and VIM-POP). To compare GEM-POP and VIM-POP with other methods. Methods GEM and VIM models separated cross-sectional abnormal VFs from 859 eyes and normal VFs from 1117 eyes into abnormal and normal clusters. Clusters were decomposed into independent axes. The confidence limit (CL) of stability was established for each axis with a set of 84 stable eyes. Sensitivity for detecting progression was assessed in a sample of 83 eyes with known progressive glaucomatous optic neuropathy (PGON). Eyes were classified as progressed if any defect pattern progressed beyond the CL of stability. Performance of GEM-POP and VIM-POP was compared to point-wise linear regression (PLR), permutation analysis of PLR (PoPLR), and linear regression (LR) of mean deviation (MD), and visual field index (VFI). Results Sensitivity and specificity for detecting glaucomatous VFs were 89.9% and 93.8%, respectively, for GEM and 93.0% and 97.0%, respectively, for VIM. Receiver operating characteristic (ROC) curve areas for classifying progressed eyes were 0.82 for VIM-POP, 0.86 for GEM-POP, 0.81 for PoPLR, 0.69 for LR of MD, and 0.76 for LR of VFI. Conclusions GEM-POP was significantly more sensitive to PGON than PoPLR and linear regression of MD and VFI in our sample, while providing localized progression information. Translational Relevance Detection of glaucomatous progression can be improved by assessing longitudinal changes in localized patterns of glaucomatous defect identified by unsupervised machine learning.

Author(s): Christopher, Mark; Belghith, Akram; Weinreb, Robert N; Bowd, Christopher; Goldbaum, Michael H; Saunders, Luke J; Medeiros, Felipe A; Zangwill, Linda M | Abstract: Purpose:To apply computational techniques to wide-angle swept-source optical coherence tomography (SS-OCT) images to identify novel, glaucoma-related structural features and improve detection of glaucoma and prediction of future glaucomatous progression. Methods:Wide-angle SS-OCT, OCT circumpapillary retinal nerve fiber layer (cpRNFL) circle scans spectral-domain (SD)-OCT, standard automated perimetry (SAP), and frequency doubling technology (FDT) visual field tests were completed every 3 months for 2 years from a cohort of 28 healthy participants (56 eyes) and 93 glaucoma participants (179 eyes). RNFL thickness maps were extracted from segmented SS-OCT images and an unsupervised machine learning approach based on principal component analysis (PCA) was used to identify novel structural features. Area under the receiver operating characteristic curve (AUC) was used to assess diagnostic accuracy of RNFL PCA for detecting glaucoma and progression compared to SAP, FDT, and cpRNFL measures. Results:The RNFL PCA features were significantly associated with mean deviation (MD) in both SAP (R2 = 0.49, P l 0.0001) and FDT visual field testing (R2 = 0.48, P l 0.0001), and with mean circumpapillary RNFL thickness (cpRNFLt) from SD-OCT (R2 = 0.58, P l 0.0001). The identified features outperformed each of these measures in detecting glaucoma with an AUC of 0.95 for RNFL PCA compared to an 0.90 for mean cpRNFLt (P = 0.09), 0.86 for SAP MD (P = 0.034), and 0.83 for FDT MD (P = 0.021). Accuracy in predicting progression was also significantly higher for RNFL PCA compared to SAP MD, FDT MD, and mean cpRNFLt (P = 0.046, P = 0.007, and P = 0.044, respectively). Conclusions:A computational approach can identify structural features that improve glaucoma detection and progression prediction.

The purposes of this study were to evaluate with spectral domain-optical coherence tomography the relationship between the retina and overlying silicone oil tamponade after macular hole surgery and to evaluate how this relationship changes with patient positioning.We studied a retrospective consecutive case series of 10 eyes from 9 patients who underwent macular hole surgery with silicone oil tamponade and subsequent spectral domain-optical coherence tomography scans. Four of the included eyes were also imaged with patients in face-down posture to determine whether the silicone-retina apposition changes with prone positioning. Finally, a single patient was also scanned in the lateral and supine positions.The posterior surface of the silicone oil bubble was well visualized in all 10 eyes. In the majority of eyes (7 of 10), the oil tamponade bridged the macular hole, creating a prefoveal fluid space, but in 3 eyes the silicone oil filled the macular hole and was seen in touch with the underlying foveal depression or retinal pigment epithelium. In 75% of eyes (3 of 4), the silicone oil-retinal approximation did not vary with face-down position. Supine positioning clearly floated the silicone tamponade anteriorly and off the retinal surface.Silicone oil tamponade can either bridge macular holes or, in a novel finding, fill the underlying foveal depression or macular hole space. Generally, the oil position is stable between face-forward and prone spectral-domain optical coherence tomography images, suggesting that either of these patient positions allows waterproofing of the underlying macular hole. Finally, our images confirm that supine positioning should be avoided postoperatively because it leads to loss of oil-retinal tamponade.

Display Omitted We propose an optimization framework to detect glaucomatous change over time.We present the rate of glaucomatous progression over time graphically.Similar accuracy while less complex and faster algorithm than other methods. Detecting glaucomatous progression is an important aspect of glaucoma management. The assessment of longitudinal series of visual fields, measured using Standard Automated Perimetry (SAP), is considered the reference standard for this effort. We seek efficient techniques for determining progression from longitudinal visual fields by formulating the problem as an optimization framework, learned from a population of glaucoma data. The longitudinal data from each patients eye were used in a convex optimization framework to find a vector that is representative of the progression direction of the sample population, as a whole. Post-hoc analysis of longitudinal visual fields across the derived vector led to optimal progression (change) detection. The proposed method was compared to recently described progression detection methods and to linear regression of instrument-defined global indices, and showed slightly higher sensitivities at the highest specificities than other methods (a clinically desirable result). The proposed approach is simpler, faster, and more efficient for detecting glaucomatous changes, compared to our previously proposed machine learning-based methods, although it provides somewhat less information. This approach has potential application in glaucoma clinics for patient monitoring and in research centers for classification of study participants.

Glaucoma is a progressive optic neuropathy with characteristic structural changes in the optic nerve head reflected in the visual field. The visual-field sensitivity test is commonly used in a clinical setting to evaluate glaucoma. Standard automated perimetry (SAP) is a common computerized visual-field test whose output is amenable to machine learning. We compared the performance of a number of machine learning algorithms with STATPAC indexes mean deviation, pattern standard deviation, and corrected pattern standard deviation. The machine learning algorithms studied included multilayer perceptron (MLP), support vector machine (SVM), and linear (LDA) and quadratic discriminant analysis (QDA), Parzen window, mixture of Gaussian (MOG), and mixture of generalized Gaussian (MGG). MLP and SVM are classifiers that work directly on the decision boundary and fall under the discriminative paradigm. Generative classifiers, which first model the data probability density and then perform classification via Bayes' rule, usually give deeper insight into the structure of the data space. We have applied MOG, MGG, LDA, QDA, and Parzen window to the classification of glaucoma from SAP. Performance of the various classifiers was compared by the areas under their receiver operating characteristic curves and by sensitivities (true-positive rates) at chosen specificities (true-negative rates). The machine-learning-type classifiers showed improved performance over the best indexes from STATPAC. Forward-selection and backward-elimination methodology further improved the classification rate and also has the potential to reduce testing time by diminishing the number of visual-field location measurements.

Melanoma of the iris is a rare condition compared to posterior ocular tumors and in this case report we present a 51-year-old female patient with diffuse iris melanoma. Traditional COMS (Collaborative Ocular Melanoma Study) plaques are used at our institution for radiation therapy, so a novel modification of the traditional plaque was required to allow better conformance with placement on the cornea. The usual silastic insert was machined to dimensions in compliance with the cornea, placed without incident, and treatment delivered with excellent patient tolerance of the modified plaque.

Refractive error (RE) is a complex, multifactorial disorder characterized by a mismatch between the optical power of the eye and its axial length that causes object images to be focused off the retina. The two major subtypes of RE are myopia (nearsightedness) and hyperopia (farsightedness), which represent opposite ends of the distribution of the quantitative measure of spherical refraction. We performed a fixed effects meta-analysis of genome-wide association results of myopia and hyperopia from 9 studies of European-derived populations: AREDS, KORA, FES, OGP-Talana, MESA, RSI, RSII, RSIII and ERF. One genome-wide significant region was observed for myopia, corresponding to a previously identified myopia locus on 8q12 (p = 1.25×10-8), which has been reported by Kiefer et al. as significantly associated with myopia age at onset and Verhoeven et al. as significantly associated to mean spherical-equivalent (MSE) refractive error. We observed two genome-wide significant associations with hyperopia. These regions overlapped with loci on 15q14 (minimum p value = 9.11×10-11) and 8q12 (minimum p value 1.82×10-11) previously reported for MSE and myopia age at onset. We also used an intermarker linkage- disequilibrium-based method for calculating the effective number of tests in targeted regional replication analyses. We analyzed myopia (which represents the closest phenotype in our data to the one used by Kiefer et al.) and showed replication of 10 additional loci associated with myopia previously reported by Kiefer et al. This is the first replication of these loci using myopia as the trait under analysis. "Replication-level" association was also seen between hyperopia and 12 of Kiefer et al.'s published loci. For the loci that show evidence of association to both myopia and hyperopia, the estimated effect of the risk alleles were in opposite directions for the two traits. This suggests that these loci are important contributors to variation of refractive error across the distribution.

Machine learning classifiers were employed to detect glaucomatous progression using longitudinal series of structural data extracted from retinal nerve fiber layer thickness measurements and visual functional data recorded from standard automated perimetry tests. Using the collected data, a longitudinal feature vector was created for each patient's eye by computing the norm 1 difference vector of the data at the baseline and at each follow-up visit. The longitudinal features from each patient's eye were then fed to the machine learning classifier to classify each eye as stable or progressed over time. This study was performed using several machine learning classifiers including Bayesian, Lazy, Meta, and Tree, composing different families. Combinations of structural and functional features were selected and ranked to determine the relative effectiveness of each feature. Finally, the outcomes of the classifiers were assessed by several performance metrics and the effectiveness of structural and functional features were analyzed.

Glaucoma is a potentially blinding optic neuropathy that results in a decrease in visual sensitivity. Visual field abnormalities (decreased visual sensitivity on psychophysical tests) are the primary means of glaucoma diagnosis. One form of visual field testing is Frequency Doubling Technology (FDT) that tests sensitivity at 52 points within the visual field. Like other psychophysical tests used in clinical practice, FDT results yield specific patterns of defect indicative of the disease. We used Gaussian Mixture Model with Expectation Maximization (GEM), (EM is used to estimate the model parameters) to automatically separate FDT data into clusters of normal and abnormal eyes. Principal component analysis (PCA) was used to decompose each cluster into different axes (patterns). FDT measurements were obtained from 1,190 eyes with normal FDT results and 786 eyes with abnormal (i.e., glaucomatous) FDT results, recruited from a university-based, longitudinal, multi-center, clinical study on glaucoma. The GEM input was the 52-point FDT threshold sensitivities for all eyes. The optimal GEM model separated the FDT fields into 3 clusters. Cluster 1 contained 94% normal fields (94% specificity) and clusters 2 and 3 combined, contained 77% abnormal fields (77% sensitivity). For clusters 1, 2 and 3 the optimal number of PCA-identified axes were 2, 2 and 5, respectively. GEM with PCA successfully separated FDT fields from healthy and glaucoma eyes and identified familiar glaucomatous patterns of loss.

Purpose: The variational Bayesian independent component analysis-mixture model (VIM), an unsupervised machine-learning classifier, was used to automatically separate Matrix Frequency Doubling Technology (FDT) perimetry data into clusters of healthy and glaucomatous eyes, and to identify axes representing statistically independent patterns of defect in the glaucoma clusters. Methods: FDT measurements were obtained from 1,190 eyes with normal FDT results and 786 eyes with abnormal FDT results from the UCSD-based Diagnostic Innovations in Glaucoma Study (DIGS) and African Descent and Glaucoma Evaluation Study (ADAGES). For all eyes, VIM input was 52 threshold test points from the 24-2 test pattern, plus age. Results: FDT mean deviation was -1.00 dB (S.D. = 2.80 dB) and -5.57 dB (S.D. = 5.09 dB) in FDT-normal eyes and FDT-abnormal eyes, respectively (p

Age-related macular degeneration (AMD) is a leading cause of visual impairment in aging populations in industrialized countries. Here we investigated whether the genotype of vascular endothelial growth factor A (VEGFA) gene is associated with response to anti-VEGF therapy. 223 eyes with neovascular AMD were treated with intravitreal anti-VEGF therapy. Responders were defined as patients who had an improvement in best corrected visual acuity (BCVA) of at least 5 letters or one line on the EDTRS visual acuity chart along with resolution of intraretinal or subretinal fluid over 12 months. Patients who did not meet the definition of responders were classified as poor-responders. The vision of responders (n = 148) improved while the vision of poor-responders (n = 75) worsened (P

Purpose We demonstrate foveal hypoplasia by optical coherence tomography (OCT). Design Observational case report. Methods Foveal hypoplasia was demonstrated in a 79-year-old man with decreased vision since childhood using fundus photography, fluorescein angiography, and OCT scans through four meridians of the central foveal. Results The OCT showed extension of all neurosensory retinal layers through the area in which the fovea would normally be located. Conclusion Optical coherence tomography can be used to demonstrate foveal hypoplasia.

PURPOSE. We evaluated Progression of Patterns (POP) for its ability to identify progression of glaucomatous visual field (VF) defects. METHODS. POP uses variational Bayesian independent component mixture model (VIM), a machine learning classifier (MLC) developed previously. VIM separated Swedish Interactive Thresholding Algorithm (SITA) VFs from a set of 2,085 normal and glaucomatous eyes into nine axes (VF patterns): seven glaucomatous. Stable glaucoma was simulated in a second set of 55 patient eyes with five VFs each, collected within four weeks. A third set of 628 eyes with 4,186 VFs (mean 6 SD of 6.7 6 1.7 VFs over 4.0 6 1.4 years) was tested for progression. Tested eyes were placed into suspect and glaucoma categories at baseline, based on VFs and disk stereoscopic photographs; a subset of eyes had stereophotographic evidence of progressive glaucomatous optic neuropathy (PGON). Each sequence of fields was projected along seven VIM glaucoma axes. Linear regression (LR) slopes generated from projections onto each axis yielded a degree of confidence (DOC) that there was progression. At 95% specificity, progression cutoffs were established for POP, visual field index (VFI), and mean deviation (MD). Guided progression analysis (GPA) was also compared. RESULTS. POP identified a statistically similar number of eyes (P > 0.05) as progressing compared with VFI, MD, and GPA in suspects (3.8%, 2.7%, 5.6%, and 2.9%, respectively), and more eyes than GPA (P ¼ 0.01) in glaucoma (16.0%, 15.3%, 12.0%, and 7.3%, respectively), and more eyes than GPA (P ¼ 0.05) in PGON eyes (26.3%, 23.7%, 27.6%, and 14.5%, respectively). CONCLUSIONS. POP, with its display of DOC of progression and its identification of progressing VF defect pattern, adds to the information available to the clinician for detecting VF progression. (Invest Ophthalmol Vis Sci. 2012;53:6557‐6567) DOI:10.1167/iovs.11-8363

Many clinical entities may be associated with the development of peripheral retinal neovascularization. In this paper, we review the mechanisms of normal and abnormal angiogenesis in the retina. Specific disease entities associated with peripheral proliferative retinopathies are discussed. These include vascular disease with ischemia, inflammatory diseases with possible ischemia and a variety of miscellaneous causes, including hereditary diseases and tumors. Basic principles for the clinical evaluation of patients with retinal neovascularization are described. Finally, the treatments for retinal neovascularization, including cryopexy and local and panretinal photocoagulation are reviewed, and techniques and possible mechanisms of the beneficial results of treatment are described

Objectives To evaluate the independent and joint effects of genetic factors and environmental variables on advanced forms of age-related macular degeneration (AMD), including geographic atrophy and choroidal neovascularization, and to develop a predictive model with genetic and environmental factors included. Methods Demographic information, including age at onset, smoking status, and body mass index, was collected for 1844 participants. Genotypes were evaluated for 8 variants in 5 genes related to AMD. Unconditional logistic regression analyses were performed to generate a risk predictive model. Results All genetic variants showed a strong association with AMD. Multivariate odds ratios were 3.52 (95% confidence interval, 2.08-5.94) for complement factor H, CFH rs1061170 CC, 4.21 (2.30-7.70) for CFH rs2274700 CC, 0.46 (0.27-0.80) for C2 rs9332739 CC/CG, 0.44 (0.30-0.66) for CFB rs641153 TT/CT, 10.99 (6.04-19.97) for HTRA1/LOC387715 rs10490924 TT, and 2.66 (1.43-4.96) for C3 rs2230199 GG. Smoking was independently associated with advanced AMD after controlling for age, sex, body mass index, and all genetic variants. Conclusion CFH confers more risk to the bilaterality of geographic atrophy, whereas HTRA1/LOC387715 contributes more to the bilaterality of choroidal neovascularization. C3 confers more risk for geographic atrophy than choroidal neovascularization. Risk models with combined genetic and environmental factors have notable discrimination power. Clinical Relevance Early detection and risk prediction of AMD could help to improve the prognosis of AMD and to reduce the outcome of blindness. Targeting high-risk individuals for surveillance and clinical interventions may help reduce disease burden.

OBJECTIVE To examine the persistence of the original treatment effects 10 years after the Diabetes Control and Complications Trial (DCCT) in the follow-up Epidemiology of Diabetes Interventions and Complications (EDIC) study. In the DCCT, intensive therapy aimed at near-normal glycemia reduced the risk of microvascular complications of type 1 diabetes mellitus compared with conventional therapy. METHODS Retinopathy was evaluated by fundus photography in 1211 subjects at EDIC year 10. Further 3-step progression on the Early Treatment Diabetic Retinopathy Study scale from DCCT closeout was the primary outcome. RESULTS After 10 years of EDIC follow-up, there was no significant difference in mean glycated hemoglobin levels (8.07% vs 7.98%) between the original treatment groups. Nevertheless, compared with the former conventional treatment group, the former intensive group had significantly lower incidences from DCCT close of further retinopathy progression and proliferative retinopathy or worse (hazard reductions, 53%-56%; P < .001). The risk (hazard) reductions at 10 years of EDIC were attenuated compared with the 70% to 71% over the first 4 years of EDIC (P < .001). The persistent beneficial effects of former intensive therapy were largely explained by the difference in glycated hemoglobin levels during DCCT. CONCLUSION The persistent difference in diabetic retinopathy between former intensive and conventional therapy ("metabolic memory") continues for at least 10 years but may be waning. TRIAL REGISTRATION (clinicaltrials.gov) Identifiers: NCT00360815 and NCT00360893.

Machine learning is concerned with the design and development of algorithms and techniques that allow computers to "learn" patterns in data using iterative processes. Such processes can be supervised (guided by a priori group membership information) or unsupervised (guided by patterns within the data). Machine learning classifiers (MLC) are unconstrained by statistical assumptions and therefore are adaptable to complex data. Recent applications of MLC techniques to the detection and monitoring of glaucoma by analysis of visual field and optical imaging data suggest that these methods can provide improvement over currently available techniques. This article provides some background about the classification task in glaucoma and the structure and evaluation of MLCs, and it reviews MLC techniques as they have been applied to visual function and optical imaging in glaucoma research.

Aims: To compare the non-decanted (standard) 4 mg versus the decanted 20 mg intravitreal triamcinolone acetonide (IVTA) injections and to assess their effect on intraocular pressure (IOP). Methods: We retrospectively reviewed the records of 92 consecutive eyes, which received an intravitreal injection of either dose of triamcinolone acetonide, at a single retina centre. The change in IOP (elevation of at least 5 mm Hg from baseline or above 21 mm Hg) was analysed with a multivariate logistic analysis. The mean follow-up period in both groups was 27 weeks. A subgroup analysis comparing vitrectomised to non-vitrectomised eyes in both groups was also performed. Results: Of the 92 eyes, 46% (23 of 51) in the 4 mg group versus 30% (12 of 41) in the 20 mg group had an IOP >21 mm of Hg (p = 0.14) after a mean follow-up period of 27 weeks. The vitrectomised eyes (3 of 24) in the 20 mg group had a significantly lower rate of IVTA induced IOP elevation than the non-vitrectomised eyes (9 of 17) (p = 0.013). The IOP elevation occurred significantly earlier in the 4 mg group (vitrectomised eyes 27 (SD 43) days and non-vitrectomised eyes 61 (52) days) than in the 20 mg group (vitrectomised eyes 104 (56) days and non-vitrectomised eyes 119 (82) days), independent of the vitreous status (vitrectomised p = 0.05 and non-vitrectomised p = 0.04). The mean value of initial high IOP in the non-vitrectomised eyes was higher in the 4 mg group than in the corresponding 20 mg group (p = 0.048). Conclusion: Decanted 20 mg IVTA may not pose a significantly greater risk of IOP elevation than the 4 mg non-decanted IVTA.

A 50-year-old man presented with uncontrolled erythroclastic glaucoma and recurrent vitreous hemorrhage from a previously irradiated choroidal melanoma. Despite trabeculectomy, intraocular pressure became uncontrolled due to increasing bleeding from the melanoma. The vitreous hemorrhage became voluminous and could not be stopped with multiple vitrectomies with endolaser, transpupillary thermotherapy, and transscleral laser. Endoresection of the regressed tumor and its bleeding vasculature resulted in immediate cessation of the bleeding and control of the intraocular pressure.

Media opacification in eyes filled with silicone oil makes the evaluation of recurrent retinal detachment difficult. Ultrasonography through silicone oil is subject to significant imaging artifacts. We performed magnetic resonance imaging on six patients with unilateral intravitreal silicone oil to determine if the technique would detect detached retina and subretinal oil. All patients had undergone pars plana vitrectomy with silicone oil injection for proliferative vitreoretinopathy; five patients had encircling solid silicone scleral buckles. In five patients the media were clear, and ophthalmoscopic findings were correlated with magnetic resonance findings. Four patients had recurrence of inferior retinal detachment; magnetic resonance imaging demonstrated subretinal oil in three of these patients. One patient had a concentric, shallow, anterior retinal detachment; magnetic resonance scanning demonstrated a globular hyperintensity suggestive of subretinal oil. In the sixth patient, who had an opaque cornea, magnetic resonance imaging suggested that the retina was attached preoperatively; this was confirmed at subsequent surgery. A chemical shift artifact was helpful in defining the contour of retinal detachments and the presence of subretinal oil by outlining the silicone oil within the eye.

PURPOSE. Clustering by unsupervised learning with machine learning classifiers was shown to segment clusters of patterns in standard automated perimetry (SAP) for glaucoma in previous publications. In this study, unsupervised learning by independent component analysis decomposed SAPfield patterns into axes, and the information represented by these axes was evaluated. METHODS. SAP fields were used that were obtained with the Humphrey Visual Field Analyzer (Carl Zeiss Meditec, Dublin, CA) from 189 normal eyes and 156 eyes with glaucomatous optic neuropathy (GON) determined by masked review with stereoscopic optic disc photographs. The variational Bayesian independent component analysis mixture model (vB-ICA-mm) partitioned the SAP fields into the most informative number of clusters. Simultaneously, the model learned an optimal number of maximally independent axes for each cluster. RESULTS. The most informative number of clusters in the SAP set was two. vB-ICA-mm placed 68.6% of the eyes with GON in a cluster labeled G and 98.4% of the eyes with normal optic discs in a cluster labeled N. Cluster G optimally contained six axes. Post hoc analysis of patterns generated at 1S D and2 SD from the cluster G mean on the six axes revealed defects similar to those identified by experts as indicative of glaucoma. SAP fields associated with an axis showed increasing severity, as they were located farther in the positive direction from the cluster G mean. CONCLUSIONS. vB-ICA-mm represented the SAP fields with patterns that were meaningful for glaucoma experts. This process also captured severity in the patterns uncovered. These findings should validate vB-ICA-mm as a data-mining technique for new and unfamiliar complex tests. (Invest Ophthalmol Vis Sci. 2005;46:3676‐3683) DOI:10.1167/iovs.04-1167 T his study focuses on the problem faced by physicians in interpreting tests for the diagnosis and management of diseases. The discipline of glaucoma diagnosis is particularly rich with data-intense tests that are currently being applied or developed to determine the status of glaucoma in an eye. Diagnosing and staging glaucoma both necessitate the use of several tests, including structural and functional assessment of the optic nerve. One of these, visual field testing, has been used by experts for several decades as the prime indicator for both diagnosis of glaucoma and the detection of progression. This wide expertise can be used to evaluate the ability of machine learning classifiers to find information that can be valuable in clinical practice. The hypothesis proposed is that machine learning classifiers can learn useful information about visual field patterns. This information can be validated against the prevailing experience of experts. Such validation would give credence to the application of the same process to new and unfamiliar tests. Visual fields have been used to diagnose glaucoma since

Purpose. To determine whether a variational Bayesian independent component analysis mixture model (vB-ICA-mm), a form of unsupervised machine learning, can be used to identify and quantify areas of progression in standard automated perimetry fields. Methods. In an earlier study, it was shown that a model using vB-ICA-mm can separate normal fields from fields with six different patterns of visual field loss related to glaucomatous optic neuropathy (GON) along maximally independent axes. In the present study, an independent group of 191 patient eyes (66 with ocular hypertension (OHT), 12 with suspected glaucoma by field, 61 with suspected glaucoma by disc, and 52 with glaucoma) with five or more standard visual fields under observation for a mean of 6.24 ± 2.65 years and 8.11 ± 2.42 visual fields were evaluated with the vB-ICA-mm. In addition, eyes with progressive GON (PGON) were identified (n = 39). Each participant had a series of fields tested, with each field entered independently and placed along the axes of the previously developed model. This allowed change in one pattern of visual field defect (along one axis) to be assessed relative to results other areas of that same field (no change along other axes). Progression was based on a slope falling outside the 5th and the 95th percentile limits of all slopes, with at least two axes not showing such a deviation in a given individual’s series of fields. Fields were also scored using Advanced Glaucoma Intervention Study (AGIS) and the Early Manifest Glaucoma Treatment Trial (EMGT) criteria. Results. Thirty-two of 191 eyes progressed on vB-ICA-mm by this definition. Of the 32, 22 had field loss at baseline, 7 had only GON, 3 were OHTs and 12 were from the 39 eyes (31%) with PGON. The vB-ICA-mm identified a higher percentage of progressing eyes in each diagnostic category than did AGIS or and the EMGT. Conclusions. The vB-ICA-mm can quantitatively identify progression in eyes with glaucoma by evaluating change in one or more patterns of the visual field loss while other areas or patterns remain stable. This may enable each eye to contribute to the determination of whether change is caused by true progression or by variability.

PURPOSE. To classify healthy and glaucomatous eyes using relevance vector machine (RVM) and support vector machine (SVM) learning classifiers trained on retinal nerve fiber layer (RNFL) thickness measurements obtained by scanning laser polarimetry (SLP). METHODS. Seventy-two eyes of 72 healthy control subjects (average age 64.3 8.8 years, visual field mean deviation 0.71 1.2 dB) and 92 eyes of 92 patients with glaucoma (average age 66.9 8.9 years, visual field mean deviation 5.32 4.0 dB) were imaged with SLP with variable corneal compensation (GDx VCC; Laser Diagnostic Technologies, San Diego, CA). RVM and SVM learning classifiers were trained and tested on SLP-determined RNFL thickness measurements from 14 standard parameters and 64 sectors (approximately 5.6° each) obtained in the circumpapillary area under the instrument-defined measurement ellipse (total 78 parameters). Tenfold cross-validation was used to train and test RVM and SVM classifiers on unique subsets of the full 164-eye data set and areas under the receiver operating characteristic (AUROC) curve for the classification of eyes in the test set were generated. AUROC curve results from RVM and SVM were compared to those for 14 SLP software-generated global and regional RNFL thickness parameters. Also reported was the AUROC curve for the GDx VCC software-generated nerve fiber indicator (NFI). RESULTS. The AUROC curves for RVM and SVM were 0.90 and 0.91, respectively, and increased to 0.93 and 0.94 when the training sets were optimized with sequential forward and backward selection (resulting in reduced dimensional data sets). AUROC curves for optimized RVM and SVM were significantly larger than those for all individual SLP parameters. The AUROC curve for the NFI was 0.87. CONCLUSIONS. Results from RVM and SVM trained on SLP RNFL thickness measurements are similar and provide accurate classification of glaucomatous and healthy eyes. RVM may be preferable to SVM, because it provides a Bayesian-derived probability of glaucoma as an output. These results suggest that these machine learning classifiers show good potential for glaucoma diagnosis. (Invest Ophthalmol Vis Sci. 2005;46: 1322‐1329) DOI:10.1167/iovs.04-1122

PURPOSE. To determine whether an unsupervised machine learning classifier can identify patterns of visual field loss in standard visual fields consistent with typical patterns learned by decades of human experience. METHODS. Standard perimetry thresholds for 52 locations plus age from one eye of each of 156 patients with glaucomatous optic neuropathy (GON) and 189 eyes of healthy subjects were clustered with an unsupervised machine classifier, variational Bayesian mixture of factor analysis (vbMFA). RESULTS. The vbMFA formed five distinct clusters. Cluster 5 held 186 of 189 fields from normal eyes plus 46 from eyes with GON. These fields were then judged within normal limits by several traditional methods. Each of the other four clusters could be described by the pattern of loss found within it. Cluster 1 (71 GON 3 normal optic discs) included early, localized defects. A purely diffuse component was rare. Cluster 2 (26 GON) exhibited primarily deep superior hemifield defects, and cluster 3 (10 GON) held deep inferior hemifield defects only or in combination with lesser superior field defects. Cluster 4 (6 GON) showed deep defects in both hemifields. In other words, visual fields within a given cluster had similar patterns of loss that differed from the predominant pattern found in other clusters. The classifier separated the data based solely on the patterns of loss within the fields, without being guided by the diagnosis, placing 98.4% of the healthy eyes within the same cluster and spreading 70.5% of the eyes with GON across the other four clusters, in good agreement with a glaucoma expert and pattern standard deviation. CONCLUSIONS. Without training-based diagnosis (unsupervised learning), the vbMFA identified four important patterns of field loss in eyes with GON in a manner consistent with years of

Objective The authors performed a study to determine the effectiveness and safety of silicone oil as a substitute for gas to fill the vitreous cavity to treat macular holes. Design Multicenter, nonrandomized, interventional trial. Participants Thirty-seven consecutive patients chose vitrectomy with silicone tamponade instead of gas to treat 40 eyes with stage-2 to stage-4 idiopathic age-related macular holes. Stage-2 holes constituted 40% of the holes, and stage-3 and stage-4 holes made up 60%. Intervention All eyes were treated with vitrectomy, manual detachment of the posterior vitreous face (not done for stage-4 holes), autologous serum instillation, and silicone fill of the vitreous cavity. After insertion of the oil, the patients resumed normal activity with no restriction of head or eye position except to avoid faceup position. The oil was removed after approximately 6 weeks. Main outcome measures The authors considered the seal of the macular hole and the preoperative and postoperative logarithm of the minimum angle of resolution (logMAR) visions the most significant measures for comparison to other studies. Results Eighty percent of all holes and 86% of holes not treated previously were sealed with a single silicone tamponade of the vitreous cavity. The logMAR value of visual acuity improved an average of 0.26 (2.6 lines) to 0.61 (20/81) for all eyes and 0.34 (3.4 lines) to 0.52 (20/66) when the macular hole sealed. Completeness of fill of the vitreous cavity with silicone affected seal of the macular hole. Three of eight eyes in which open holes developed after oil removal had less than 90% fill of the vitreous cavity by silicone. Sixty-nine percent of lenses increased opacity one grade or were removed after silicone tamponade. There were no significant adverse effects arising from silicone tamponade. Conclusions Silicone oil tamponade of macular holes is effective and safe. Silicone may be optimal for the treatment of macular holes in persons who must travel, who cannot maintain facedown positioning, or who have monocular vision. The most important factor in the successful closure of the macular hole was the completeness of fill of the vitreous cavity with silicone oil.

PURPOSE We previously reported the use of clustering by unsupervised learning with machine learning classifiers to segment clusters of patterns in standard automated perimetry (SAP) for glaucoma. In this study, the process of unsupervised learning by independent component analysis decomposed SAP field patterns into axes, and the information represented by these axes was evaluated. METHODS SAP fields were obtained with the Humphrey Visual Field Analyzer on 189 normal eyes and 156 eyes with glaucomatous optic neuropathy (GON) determined by masked review with stereoscopic optic disc photos. The variational Bayesian independent component analysis mixture model (vB-ICA-mm) partitioned the SAP fields into the most informative number of clusters. Simultaneously, it learned an optimal number of maximally independent axes for each cluster. RESULTS The most informative number of clusters was two. vB-ICA-mm placed 68.6% of the SAP fields from eyes with GON in a cluster labeled G and 98.4% of the fields from eyes with normal optic discs in a cluster labeled N. Cluster G optimally contained six axes. Post hoc analysis of patterns generated at -1 SD and +2 SD from the cluster G mean on the six axes revealed defects similar to those identified by experts as indicative of glaucoma. SAP fields associated with an axis showed increasing severity as they were located farther in the positive direction from the cluster G mean. CONCLUSIONS vB-ICA-mm represented the SAP fields with patterns that were meaningful for glaucoma experts. This process also captured severity in the patterns uncovered. These findings should validate vB-ICA-mm as a data mining technique for new and unfamiliar complex tests.

The goal of this study was to determine if glaucomatous progression in suspect eyes can be predicted from baseline confocal scanning laser ophthalmoscope (CSLO) and standard automated perimetry (SAP) measurements analyzed with relevance vector machine (RVM) classifiers.Two hundred sixty-four eyes of 193 participants were included. All eyes had normal SAP results at baseline with five or more SAP tests over time. Eyes were labeled progressed (n = 47) or stable (n = 217) during follow-up based on SAP Guided Progression Analysis or serial stereophotograph assessment. Baseline CSLO-measured topographic parameters (n = 117) and baseline total deviation values from the 24-2 SAP test-grid (n = 52) were selected from each eye. Ten-fold cross-validation was used to train and test RVMs using the CSLO and SAP features. Receiver operating characteristic (ROC) curve areas were calculated using full and optimized feature sets. ROC curve results from RVM analyses of CSLO, SAP, and CSLO and SAP combined were compared to CSLO and SAP global indices (Glaucoma Probability Score, mean deviation and pattern standard deviation).The areas under the ROC curves (AUROCs) for RVMs trained on optimized feature sets of CSLO parameters, SAP parameters, and CSLO and SAP parameters combined were 0.640, 0.762, and 0.805, respectively. AUROCs for CSLO Glaucoma Probability Score, SAP mean deviation (MD), and SAP pattern standard deviation (PSD) were 0.517, 0.513, and 0.620, respectively. No CSLO or SAP global indices discriminated between baseline measurements from progressed and stable eyes better than chance.In our sample, RVM analyses of baseline CSLO and SAP measurements could identify eyes that showed future glaucomatous progression with a higher accuracy than the CSLO and SAP global indices. (ClinicalTrials.gov numbers, NCT00221897, NCT00221923.).