John Pascual | PubMed Health

PubMed2004

Patterns of glaucomatous visual field progression identified by three progression criteria.

PURPOSE: To determine typical patterns of repeatable glaucomatous visual field progression. DESIGN: Retrospective analysis of data obtained from two prospective studies. METHODS: Included were 72 eyes of 72 patients tested up to six times over 2 years, and 40 eyes of 40 patients followed annually for up to 12 years. Each patient had two abnormal baseline visual fields, abnormal optic nerves, and serial fields. Progression was identified using three methods: by glaucoma change probability using total deviation (GCP-TD) and pattern deviation (GCP-PD) plots and by a clinical criteria. Progression was categorized as deepening or expansion of an existing scotoma, or a new scotoma. RESULTS: The percentage of eyes repeatably progressed ranged from 17% to 27%. The most common pattern of progression was a deepening of an existing scotoma in the annual group, followed by expansion. With two follow-ups required, percentages for deepening only were 20% (clinical classifier). A combination of expansion and deepening was most common for the GCP criteria: 15% (GCP-TD classifier), and 10% (GCP-PD classifier) for the annual group. For the semiannual group, deepening was most common with the clinical criteria (11% of eyes), and deepening with expansion was most common by GCP criteria (14%, GCP-TD and GCP-PD). No eyes showed repeatable new scotomas. CONCLUSIONS: Glaucomatous visual fields progress in the area of the visual field where baseline testing showed an existing scotoma. Follow-up testing might be improved by concentrating on already defective locations and using sparser test patterns or screening algorithms in normal areas of the visual field.

Diagnostic Techniques, Ophthalmological↗

PubMed2004

Using unsupervised learning with variational bayesian mixture of factor analysis to identify patterns of glaucomatous visual field defects.

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 clinical experience.

Algorithms↗