Search tips
Search criteria

Results 1-4 (4)

Clipboard (0)
more »
Year of Publication
Document Types
1.  Scan quality effect on glaucoma discrimination by glaucoma imaging devices 
The British journal of ophthalmology  2009;93(12):1580-1584.
To evaluate, within ocular imaging scans of acceptable quality as determined by manufacturers' guidelines, the effects of image quality on glaucoma discrimination capabilities.
One hundred and four healthy and 75 glaucomatous eyes from the Advanced Imaging in Glaucoma Study (AIGS) were imaged with GDx-VCC, HRT II and StratusOCT. Quality score (QS≥8), pixel standard deviation (SD≤50) and signal strength (SS≥5) were used as quality parameter cut-offs, respectively. GDx nerve fibre indicator (NFI) and HRT Moorfields regression analysis (MRA) classifications and OCT mean retinal nerve fibre layer (RNFL) thickness were used as the discriminatory parameters. Logistic regression models were used to model the dichotomous clinical classification (healthy vs glaucoma) as a function of image-quality parameters and discriminatory parameters.
Quality parameter covariates were statistically non-significant for GDx and HRT but had an inverse effect on OCT in predicting disease (a higher SS had a lower probability of glaucoma). Age was a significant covariate for GDx and HRT, but not OCT, while ethnicity and interaction between the image quality and the institute where scans were acquired were significant covariates in the OCT models.
Scan quality within the range recommended as acceptable by the manufacturer of each imaging device does not affect the glaucoma discriminating ability of GDx or HRT but does affect Stratus OCT glaucoma discrimination.
PMCID: PMC2917916  PMID: 19692363
2.  Heidelberg Retina Tomograph 3 machine learning classifiers for glaucoma detection 
To assess performance of classifiers trained on Heidelberg Retina Tomograph 3 (HRT3) parameters for discriminating between healthy and glaucomatous eyes.
Classifiers were trained using HRT3 parameters from 60 healthy subjects and 140 glaucomatous subjects. The classifiers were trained on all 95 variables and smaller sets created with backward elimination. Seven types of classifiers, including Support Vector Machines with radial basis (SVM-radial), and Recursive Partitioning and Regression Trees (RPART), were trained on the parameters. The area under the ROC curve (AUC) was calculated for classifiers, individual parameters and HRT3 glaucoma probability scores (GPS). Classifier AUCs and leave-one-out accuracy were compared with the highest individual parameter and GPS AUCs and accuracies.
The highest AUC and accuracy for an individual parameter were 0.848 and 0.79, for vertical cup/disc ratio (vC/D). For GPS, global GPS performed best with AUC 0.829 and accuracy 0.78. SVM-radial with all parameters showed significant improvement over global GPS and vC/ D with AUC 0.916 and accuracy 0.85. RPART with all parameters provided significant improvement over global GPS with AUC 0.899 and significant improvement over global GPS and vC/D with accuracy 0.875.
Machine learning classifiers of HRT3 data provide significant enhancement over current methods for detection of glaucoma.
PMCID: PMC2916743  PMID: 18523087
3.  Imaging of the retinal nerve fibre layer for glaucoma 
Glaucoma is a group of diseases characterised by retinal ganglion cell dysfunction and death. Detection of glaucoma and its progression are based on identification of abnormalities or changes in the optic nerve head (ONH) or the retinal nerve fibre layer (RNFL), either functional or structural. This review will focus on the identification of structural abnormalities in the RNFL associated with glaucoma.
A variety of new techniques have been created and developed to move beyond photography, which generally requires subjective interpretation, to quantitative retinal imaging to measure RNFL loss. Scanning laser polarimetry uses polarised light to measure the RNFL birefringence to estimate tissue thickness. Optical coherence tomography (OCT) uses low-coherence light to create high-resolution tomographic images of the retina from backscattered light in order to measure the tissue thickness of the retinal layers and intraretinal structures. Segmentation algorithms are used to measure the thickness of the retinal nerve fibre layer directly from the OCT images. In addition to these clinically available technologies, new techniques are in the research stages. Polarisation-sensitive OCT has been developed that combines the strengths of scanning laser polarimetry with those of OCT. Ultra-fast techniques for OCT have been created for research devices. The continued utilisation of imaging devices into the clinic is refining glaucoma assessment. In the past 20 years glaucoma has gone from a disease diagnosed and followed using highly subjective techniques to one measured quantitatively and increasingly objectively.
PMCID: PMC2907255  PMID: 19028735
4.  Sources of longitudinal variability in optical coherence tomography nerve-fibre layer measurements 
The purpose of this study was to compare the day-to-day reproducibility of optical coherence tomography (OCT; StratusOCT, Carl Zeiss Meditec, Dublin, CA) measurements of retinal nerve-fibre layer (RNFL) measurements at time points 1 year apart.
One eye in each of 11 healthy subjects was examined using the StratusOCT fast RNFL scan protocol. Three fast RNFL scans with signal strength ≥7 were obtained on each of 3 days within a month. This protocol was repeated after 12 months. A linear mixed effects model fitted to the nested data was used to compute the variance components.
The square root of the variance component that was attributed to the differences between subjects was 7.17 μm in 2005 and 7.28 μm in 2006. The square roots of the variance component due to differences between days within a single subject were 1.95 μm and 1.50 μm, respectively, and for within day within a single subject were 2.51 μm and 2.55 μm, respectively. There were no statistically significant differences for any variance component between the two testing occasions.
Measurement error variance remains similar from year to year. Day and scan variance component values obtained in a cohort study may be safely applied for prediction of long-term reproducibility.
PMCID: PMC2743163  PMID: 18523086

Results 1-4 (4)