Linear Discriminant Functions to Improve the Glaucoma Probability Score Analysis to Detect Glaucomatous Optic Nerve Heads

Exploring the Heidelberg Retinal Tomograph 3 Diagnostic Accuracy across Disc Sizes and Glaucoma Stages: A Multicenter Study

Purpose: To evaluate the agreement of glaucomatous structural defects of the ganglion cell complex (GCC) detected with the spectral domain optical coherence tomography (sdOCT) with the optic nerve head alterations detected with the Heidelberg retina tomography (HRT), of glaucoma patients with ocular hypertension or open angle glaucoma.
 Material and Methods: Ninety patients eyes with structural glaucomatous defects were enrolled. All of them underwent imaging examination of GCC with sdOCT and the optic disk with HRT. The Cohen's kappa coefficient of agreement was used.
 Results: The agreement between the optic disc and GCC using the parameters of the programs analysis of the HRT, the moorfields regression analysis (MRA) and glaucoma probability score (GPS) was not significant. Instead between MRA and GPS a good agreement was calculated. Significant agreements were found between MRA and GPS on one hand and GCC on the other, considering location and length of the glaucomatous damage, while non significant agreements were found between GPS and GCC for the location and the length of the glaucomatous structural defect.
 Conclusions: There is no significance (Please explain further if you are referring to significance in terms of the difference, similarity or agreement) between HRT and sdOCT for the detection of the glaucomatous damage between the optic nerve head and the ganglion cell complex. Instead MRA and GCC detect comparable areas and lengths of the glaucomatous damage. On the other hand GPS records larger deficits relative to MRA and has not a significant agreement with the study of GCC.

To assess the ability of Heidelberg Retina Tomograph (HRT) Moorfields Regression Analysis (MRA) and Glaucoma Probability Score (GPS) classifications at baseline to predict glaucomatous progression in ocular hypertensive eyes. One hundred ninety-eight ocular hypertensive subjects underwent regular HRT and visual field (VF) testing from 1993 to 2001. HRT progression was assessed using linear regression of rim area/time. VF progression was assessed by pointwise linear regression of sensitivity/time. Subjects were classified as progressing or stable at the end of the study period. The relationship between baseline abnormal (outside normal limits combined with borderline classification) MRA and GPS classification and progression status was assessed by odds ratios (ORs). An abnormal superotemporal MRA was the only classification found to be predictive of HRT progression in isolation (OR 3.05, 1.25-7.47). Abnormal global, superotemporal, superonasal, and temporal MRA classifications were all associated with significant ORs for predicting HRT or VF progression (OR range: 1.77-2.54). Abnormal GPS classifications were not predictive of disease behavior. Combined abnormal GPS and MRA classifications were associated with higher ORs than either classification in isolation. Patients with an abnormal MRA and GPS classification at presentation may be at increased risk of HRT or VF change.

To evaluate the intrasession reproducibility of the Heidelberg Retina Tomograph III (HRT III) Glaucoma Probability Score (GPS) and analyze the factors that may affect it. Twenty normal subjects and 20 glaucomatous patients underwent 10 HRT III examinations of 1 randomly selected eye during the same session. The reproducibility of the GPS and its parameters was assessed by calculating the intraclass correlation coefficients (ICCs) and 95% confidence intervals of the study population as a whole, and of the 2 subgroups. The factors influencing individual reproducibility were evaluated by means of regression analysis. The reproducibility of the GPS parameters (at a confidence level of 95%) was almost perfect in the population as a whole (ICCs 0.964-0.856) and in the normal subjects (ICCs 0.976-0.897), but decreased in the glaucoma patients in whom it ranged from almost perfect to substantial (ICCs 0.922-0.705). The parameters were mainly influenced by image quality, but also by diagnosis and age, although the probability of glaucoma did not seem to depend on these variables. GPS is a highly reproducible diagnostic tool, especially in the case of "good" or better quality images as evaluated by the HRT software; moreover, the reproducibility of the GPS parameters was significantly better in normal subjects than in glaucoma patients.

To determine the sensitivity of manufacturer supported normative algorithms in Heidelberg retinal tomography (HRT) and Stratus optical coherence tomography (OCT) in detecting advanced primary open angle glaucoma. A total of 157 subjects with advanced primary open angle glaucoma were recruited. The eye with the more severe glaucoma, as judged by mean deviation on Humphrey visual field, was imaged. The total number of optic disc sectors classified as being outside normal limits on Moorfields Regression Analysis (MRA) from the HRT II and HRT 3 software or from the sector and quadrant averages analysis of the Retinal Nerve Fiber Layer OCT scans was compared using the Wilcoxon signed ranks test. At least one segment in each eye was identified as being outside normal limits by the OCT sector Retinal Nerve Fiber Layer analysis (100% sensitivity). Five eyes were classified as being within normal limits on the OCT quadrant average analysis (96.8% sensitivity). Four eyes were found to be either completely within normal limits or borderline on the HRT II MRA (97.5% sensitivity), of which one was subsequently identified as being outside normal limits by the HRT 3 MRA. The total number of segments classified as being outside normal limits was greater on HRT 3 MRA than HRT II MRA (P < 0.001). Automated optic nerve imaging devices rarely produce a totally normal result in the presence of advanced primary open angle glaucoma. The Stratus OCT Retinal Nerve Fiber Layer clock hour normative database performs better than other modalities studied. These findings are important in considering the utility of these algorithms for community screening.

Purpose To determine and validate the diagnostic ability of a linear discriminant function(LDF) based on the retinal nerve fiber layer thickness at each of the 12 clockhour positions obtained using optical coherence tomography for discriminating between healthy eyes and eyes with early glaucomatous visual field loss Methods We prospectively selected 62 consecutive healthy individuals and 73 patients with open‐angle glaucoma to calculate the LDF. Another independent prospective sample of 280 healthy eyes and 302 glaucomatous eyes was used to evaluate the diagnostic accuracy of the LDF Results The proposed function was LDF=15.584 – (12‐ o’clock segment thickness_0.032) – (7‐o’clock segment thickness_0.041) – (3‐o’clock segment thickness[nasal side]_0.121). The greatest area under the receiver operating characteristic curve was observed for our LDF in both populations: 0.962 and 0.922. Our LDF and the average thickness yielded sensitivities of 74.5% and 67.8%, respectively, at a fixed specificity of 95% Conclusion The LDF increased the diagnostic ability of the isolated retinal nerve fiber layer thickness at the 12 clock‐hour positions. Compared with optical coherence tomography–provided parameters, our LDF had the highest sensitivities at 85% and 95% fixed specificities to discriminate between healthy and early glaucomatous eyes

The aim of this cross-sectional study was to evaluate the results of a visual field (VF) test for patients with glaucoma and pseudo-fixation loss. These patients exhibit fixation loss (FL) rates >20% with the Humphrey Field Analyzer (HFA); however, actual fixation stabilizes when a head-mounted perimeter (imo) is used. This device is able to adjust the stimulus presentation point by tracking eye movements. We subjected 54 eyes of 54 patients with glaucoma and pseudo-FL to the HFA 30–2 or 24–2 Swedish Interactive Threshold Algorithm -Standard protocol. All patients also underwent the imo 30–2 or 24–2 Ambient Interactive Zipper Estimated Sequential Testing protocol after HFA measurement. We compared HFA and imo reliability indices [including false-positive (FP) responses, false-negative (FN) responses, and FL rate], global indices [including mean deviation (MD), visual field index (VFI), and pattern standard deviation (PSD)], and retinal sensitivity for each test point. There were no significant differences in MD, VFI, and PSD between HFA and imo, and these measures were strongly correlated (r > 0.96, p < 0.01). There were no significant differences in FP and FN between both devices, while FL measured with HFA (27.5%) was significantly reduced when measured with imo (13.2%) (p < 0.01). There was no correlation in FL and FN between both devices, and a weak correlation for FP (r = 0.29, p = 0.04). At each test point, retinal sensitivity averaged 1.7 dB higher with HFA, compared with imo (p < 0.01). There was no significant variability in global indices in patients with pseudo-FL. The FP response rate might have influenced measures of FL in patients with glaucoma and pseudo-FL.

Structural and Functional Assessment in HIV-Infected Patients Using Optical Coherence Tomography and Frequency-Doubling Technology Perimetry

To develop a diagnostic setup with classification rules for combined analysis of morphology [Heidelberg Retina Tomograph (HRT)] and function [frequency doubling technology (FDT) perimetry] measurements. We used 2 independent case-control studies from the Erlangen eye department as learning and test data for automated classification using random forests. One eye of 334 open angle glaucoma patients and 254 controls entered the study. All individuals underwent HRT scanning tomography of the optic disc, FDT screening, conventional perimetry, and evaluation of fundus photographs. Random forests were learned on individuals of the Erlangen glaucoma registry (102 preperimetric patients, 130 perimetric patients, 161 controls). The classification performances of random forests and built-in classifiers were examined by receiver operator characteristic analysis on an independent second cohort of individuals (47 preperimetric patients, 55 perimetric patients, 93 controls). HRT measurements had a higher diagnostic power for early glaucomas and FDT perimetry for glaucoma patients with visual field loss. A combination of all parameters using automated classification was superior to single tests in comparison to the diagnostic instrument with the higher diagnostic power in the respective group. Highest sensitivities at a fixed specificity (95%) in the patients of the present test population were: HRT=32%, FDT=19%, combined analysis=47% in preperimetric patients and HRT=76%, FDT=89%, combined analysis=96% in perimetric patients. The feasibility of machine learning for medical diagnostic assistance could be demonstrated in patients from 2 independent study populations. A predictive model using automated classification is able to combine the advantages of morphology and function, resulting in a higher diagnostic power for glaucoma detection.

To compare the ability of 24-2 frequency-doubling perimetry (FDP-Matrix) with standard automated perimetry with the Swedish interactive threshold algorithm (SAP-SITA) in detection of visual function abnormalities in patients with glaucomatous-appearing optic discs (GAOD). This observational case-control study included 80 patients with GAOD and 54 control subjects diagnosed by masked assessment of optic disc stereoscopic photographs. Abnormal visual function at SAP-SITA and FDP-Matrix testing required consistent abnormalities in 2 visual field examinations, determined using the glaucoma hemifield test outside 99% normal limits, pattern standard deviation outside 95% normal limits, or 3 contiguous points in the pattern deviation probability plot outside 95% normal limits (at least 1 P<1%) within the same hemifield. The FDP-Matrix and SAP-SITA detected abnormal visual function in 51% and 44%, respectively, of GAOD eyes (P = .26), and both perimetry techniques identified 11% of healthy eyes as abnormal. Agreement between FDP-Matrix and SAP-SITA was moderate (kappa = 0.49), as only 35% of GAOD eyes and 2% of healthy eyes had both visual field test results flagged as abnormal. The FDP-Matrix detected abnormal visual function in more eyes with GAOD than did SAP-SITA, although this difference was not significant. Each visual field test tended to identify different subsets of eyes with GAOD as abnormal. Combination of these perimetry techniques may improve the detection of visual function abnormalities in patients with glaucoma.

IntroductionNeurofibromatosis type 1 (NF1) is a multisystemic disease caused by the mutation of Nf1 gene located on chromosome 17q11.2. The mutation determines the loss of function of the protein neurofibromin with consequent uncontrolled cellular proliferation. Patients are characterized by a wide range of dermatological, neurological, and ophthalmological symptoms.PurposeThe aim of the study was to evaluate, through pattern visual evoked potentials (p-VEPs) and frequency doubling technology (FDT) Matrix perimetry, the objective and psychophysical functionality of the optic pathways in a group of NF1 patient.MethodsThe study group consisted of 26 patients affected by NF1 and 17 healthy controls. Each patient underwent a complete ophthalmological examination, p-VEPs with the evaluation of amplitude and latency of the P100 wave, and FDT perimetry, with the evaluation of central sensitivity (CS), mean deviation (MD), pattern standard deviation (PSD) and glaucoma hemifield test (GHT).ResultsNF1 patients showed a statistically significant alteration in the transmission of visual impulse. P-VEPs results highlighted a reduced amplitude and an increased latency of the P100 wave, suggesting an involvement of the visual pathway. Visual field analysis showed a significant reduction in all the observed parameters as well (CS, MD, PSD, and GHT).ConclusionThe present study showed, in NF1 patients, a qualitative and quantitative alteration in the conduction of stimuli through the visual pathways. The observed alterations are present, although, only at a subclinical level. None of the patients included in the study showed any manifest visual deficit nor had any concomitant pathology that might have affected the outcome of the study. In conclusion, electrophysiological exams and computer perimetry may take part, alongside a wider array of exams, in the differential diagnosis and later monitoring of NF1.

24-2 SITA Standard versus 24-2 SITA Faster in Perimetry-Naive Normal Subjects

Retinal nerve fiber layer thickness measurements with scanning laser polarimetry predict glaucomatous visual field loss

Relevance Technologies and research programs are constantly evolving. The latest models of computer perimeters allow for qualitative and quantitative assessment of photosensitivity defects and calculation of perimetric indices [1, 2]. To assess changes in the visual field in computer perimetry, the Single Field Analysis (SFA) protocol for Humphrey Field Analyzer (HFA) and Seven-in-One for Octopus are most often used. The forms of these programs contain the maximum amount of information on one page. In addition to the main data, the protocol contains study quality indicators, the number of false responses, and data on fixation loss. Perimetry results are presented in the form of maps of numerical deviation from the age norm, maps of defect probability, and perimetric indices [3]. All results are compared with the age norm. Perimetric indices are included in the HFA and Octopus perimeters to assess photosensitivity defects in digital values. MD (Mean Deviation) is the value of the average deviation of photosensitivity from the age norm, reflecting the overall loss of photosensitivity. PSD (Pattern Standard Deviation), or sLV (square root of Loss Variance), is the standard deviation of the pattern. It corresponds to the average value of the defect depth without taking into account MD. VFI (Visual Field Index) in HFA is an index that evaluates the general state of retinal sensitivity in percent. It allows to exclude the influence of clouded optical media. The Bebie curve in Octopus is a cumulative curve of sensitivity values ​​of all points, built in descending order of the curve threshold values, indicating local defects [4]. Static perimetry is more functional and effective in comparison with kinetic, therefore most glaucoma classifications are based on computer perimetry. To facilitate the interpretation of perimetry data, a classification by perimetric indices was proposed. In domestic ophthalmology, the classification of A.P. Nesterov and A.Ya. Bunin [5] is still used. It takes into account the results of perimetry and the state of the optic disc. Four stages of glaucoma are distinguished: initial (I), developed (II), advanced (III) and terminal (IV). Classifications of glaucoma based on a combination of the results of computer perimetry and optical coherence tomography are emerging, which can improve the situation in early diagnostics of glaucoma. Conclusion The use of modern computer perimeters makes it possible to conduct a qualitative assessment of photosensitivity defects. Key words: static perimetry; glaucoma; HFA; visual field

Comparison of Detectability of Visual Field Abnormality by Frequency Doubling Technology in Primary Open-angle Glaucoma and Normal-tension Glaucoma