PubMed Health⌕ Search

Biomedical subjects

Linda Zangwill

Publications and source records attributed to Linda Zangwill.

5 recordsLinked to original sources

Using unsupervised learning with independent component analysis to identify patterns of glaucomatous visual field defects.

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 SAP field 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 -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.

Artificial Intelligence↗

Efficacy and safety of memantine treatment for reduction of changes associated with experimental glaucoma in monkey, II: Structural measures.

PURPOSE: To determine, using anatomic measurements, whether daily oral dosing with memantine is both safe and effective to reduce the injury associated with experimental glaucoma in primates. METHODS: Argon laser treatment of the anterior chamber angle was used to induce chronic ocular hypertension (COHT) in the right eyes of 18 macaque monkeys. Nine animals were daily orally dosed with 4 mg/kg memantine while the other nine animals received vehicle only. Measurements of intraocular pressure (IOP) from both eyes of all animals were made at regular intervals. Appearance of the optic nerve head, retinal vessels, and surrounding retina was documented with stereo fundus photographs obtained at multiple time points throughout the study. Measurements of optic nerve head topography were obtained from confocal laser scans made from animals with the highest IOPs at approximately 3, 5, and 10 months after elevation of IOP. At approximately 16 months after IOP elevation, animals were killed and histologic counts of cells in the retinal ganglion cell (RGC) layer were made. RESULTS: Histologic measurements showed that, for animals with moderate elevation of IOP, memantine treatment was associated with an enhanced survival of RGCs in the inferior retina. Measurements of optic nerve head topography showed less IOP-induced change in memantine-treated animals. This effect was seen in measurements of both the cup and the neuroretinal rim. A comparison of these same histologic and morphologic measurements in normotensive eyes from the two treatment groups showed that memantine treatment was not associated with any significant effects on these eyes. CONCLUSIONS: Histologic measurements of RGC survival as well as tomographic measurements of nerve head topography show that systemic treatment with memantine, a compound which does not lower intraocular pressure, is both safe and effective to reduce changes associated with experimental glaucoma.

Administration, Oral↗

Evaluating several sources of variability for standard and SWAP visual fields in glaucoma patients, suspects, and normals.

PURPOSE: To quantify factors affecting test-retest variability of threshold measurements over a series of 3 serial visual fields (VF). DESIGN: Prospective comparative observational study. PARTICIPANTS: Forty-one normals, 10 suspects and 35 stable glaucoma patients. METHODS: All subjects performed 3 standard and 3 short-wavelength automated perimetry (SWAP) VFs. At each VF location, severity (defined as age-corrected total deviation) and test-retest variability (TRV), defined as the standard deviation of 3 serial threshold values, were calculated. A multiple regression model (constructed separately for standard VF and SWAP) incorporated 13 factors: severity, location, eccentricity, study group, diagnosis, superior versus inferior hemifield, nasal versus temporal hemifield, one-versus-two thresholds, age, mean pupil size, pupil size variability, between-subject variation, and residual variation. MAIN OUTCOME MEASURES: Variability in threshold sensitivity VF values. RESULTS: Mean TRV (+/- standard deviation) for normal, suspect and glaucoma eyes, respectively, was: 1.28 +/- 0.87, 1.53 +/- 1.04 and 2.20 +/- 1.79 dB for standard VF, and 1.87 +/- 1.35, 1.86 +/- 1.24 and 2.68 +/- 1.85 dB for SWAP. The contribution of each factor to the model for standard VF and SWAP (SWAP in parentheses) were: severity 15.5% (6.9%); location 2.7% (4.1%); eccentricity 1.1% (0.64%); diagnosis 2.9% (5.9%); "superior versus inferior" hemifield 0.17% (1.7%); "nasal versus temporal" hemifield 0.06% (0.02%); one-versus-two thresholds 0.04% (0.16%); age 0.1% (0.06%); mean pupil size 0.59% (0.1%); pupil size variability 3.2% (2.8%); between-subject 8.0% (13.5%) and residual variation 61.0% (66.6%). Excluding between-subject and residual variation, the 11-factor model was able to account for less than one third of the variability seen in both standard VF and SWAP. CONCLUSIONS: Severity of defect and between subject variation exerted the largest effect on TRV. However, even if all 11 factors could be adjusted for, it would reduce the magnitude of TRV by only 30%. More work is needed to reduce the remaining variability inherent in psychophysical testing and to better understand the intrinsic physiological variability present both in healthy and diseased eyes. It is possible that a larger number of VFs used for the calculation of TRV might further reduce the magnitude of the remaining variability found in this study.

Adult↗

Correction for corneal polarization axis improves the discriminating power of scanning laser polarimetry.

PURPOSE: Corneal polarization axis (CPA) has been reported to affect retardation measurements obtained with scanning laser polarimetry (SLP). The purpose of this investigation was to prospectively determine whether correction for CPA improves the discriminating power of SLP for detection of mild-to-moderate glaucoma. DESIGN: Cross-sectional analysis of normal and glaucomatous eyes. METHODS: We constructed a noninvasive slit-lamp-mounted device incorporating two crossed linear polarizers and an optical retarder to measure the slow axis of corneal polarization. Complete ocular examination, standard automated perimetry, SLP imaging, and CPA measurements were performed on normal and glaucomatous eyes. One eye/subject was enrolled; if both eyes of a patient were eligible for the study, the right eye was selected. For each of the 13 SLP parameters, logistic regression was used to determine if including CPA in the model influenced the ability to discriminate between normal and glaucomatous eyes. RESULTS: Forty-three normal eyes (average visual field mean defect, -0.53 +/- 1.4 dB) and 33 glaucomatous eyes (average visual field mean defect, -5.93 +/- 6.5 dB) were enrolled. CPA was significantly correlated with summary retardation parameters (average thickness and integral values) in normal (r = 0.72-0.83, P <.001 for all values) and glaucomatous eyes (r = 0.43-0.62, P =.013 to <.001). Including CPA in the model improved the ability to discriminate between normal and glaucomatous eyes for five retardation parameters quantifying retinal nerve fiber layer (RNFL) thickness (range of P values: 0.045-0.001). For inferior average thickness, area under the receiver operating characteristic (ROC) curve increased significantly (P =.002) from 0.70 to 0.78 after accounting for CPA; with a sensitivity set at 80% specificity improved from 33% to 72%. Correlations between visual field corrected pattern standard deviation and average thickness, ellipse average, superior average, and inferior average significantly increased (range of P values,.018-.001) after adjustment for CPA (r = -0.35 and -0.45, -0.38 and -0.47, -0.46 and -0.57, and -0.42 and -0.49, respectively). CONCLUSIONS: Correction for CPA significantly increases the correlation between retinal nerve fiber layer structural damage and visual function and significantly improves the discriminating power of SLP for detection of mild-to-moderate glaucoma.

Adult↗

Infrequent confirmation of visual field progression.

OBJECTIVE: To evaluate the effects of the repeatability criteria on the detection of change in visual fields by six progression algorithms used in standard automated perimetry. DESIGN: Retrospective, observational case series PARTICIPANTS: Fifty-one glaucoma patients, each with multiple visual fields performed between May 1990 and December 1998, were included. METHODS: Each patient's set of visual fields were analyzed using the glaucoma change probability, the Early Manifest Glaucoma Trial (EMGT) algorithm, a modified glaucoma change probability score, a modified EMGT score, the Advanced Glaucoma Intervention Study algorithm, and the Collaborative Initial Glaucoma Treatment Study algorithm. MAIN OUTCOME MEASURES: The effects of repeatability on the detection of field change, the level of agreement among algorithms, as well as the number of eyes identified as changed with each algorithm, were assessed. RESULTS: Mean follow-up was 34 months (range, 12-87 months). The average percentage of eyes with change based on three consecutive follow-up fields was 8.2% (4.0%-12.5%). However, of those showing change on the initial follow-up, this change from baseline was observed in subsequent examinations on average in 23% (18%-33%), depending on the algorithm. When change was based on just one field, four of the six algorithms noted a significantly greater number of eyes with change. The algorithms, however, did not differ significantly when confirmation of field change required two versus three consecutive follow-up visual fields. CONCLUSIONS: Although current algorithms may help identify change, there are inconsistencies among them. We found that requiring repeatable change from baseline significantly reduces the number of changed eyes identified with each subsequent follow-up field. Identification and confirmation of change in visual fields plays an important role in helping to identify true glaucoma progression; however, the specific methods to do so have yet to be determined.

Adult↗