Yap, Timothy E; Davis, Benjamin M; Bloom, Philip A; Cordeiro, M Francesca; Normando, Eduardo M; (2022) Glaucoma Rose Plot Analysis: Detecting Early Structural Progression Using Angular Histograms. Ophthalmology Glaucoma 10.1016/j.ogla.2022.06.002. (In press).

Preview

Text Cordeiro_Glaucoma Rose Plot Analysis_Pre-proof.pdf - Accepted Version Download (2MB) | Preview

Abstract

PURPOSE: To evaluate novel 'Rose Plot Analysis' (RPA) in the analysis and presentation of glaucoma structural progression data. DESIGN: A case-control image analysis study using retrospective retinal imaging series SUBJECTS: Subjects with open-angle glaucoma, with at least five registered spectral-domain optical coherence tomography (SD-OCT) scans. METHODS: Glaucoma RPA analysis was developed combining a novel application of angular histograms and dynamic cluster analysis of circumpapillary RNFL (cRNFL) OCT data. RPA plots were created for each eye and for each visit. Significant clusters of progression were indicated in red. Three masked clinicians categorised all RPA plots (progressing/not progressing), in addition to the measuring of significant RPA area. Masked OCT series assessment with linear regression of averaged global and sectoral cRNFL thickness was conducted as the 'clinical imaging standard'. MAIN OUTCOME MEASURES: Inter-observer agreement was compared between RPA and the clinical imaging standard. Discriminative ability was assessed using receiver operating characteristic (ROC) curves. Time to detection of progression was compared using Kaplan-Meier survival analysis, and agreement of RPA with the clinical imaging standard was calculated. RESULTS: 743 scans from 98 eyes were included in the study. Inter-observer agreement was significantly greater when categorising RPA (kappa = 0.86, 95% CI 0.81 - 0.91) compared to OCT image series (kappa = 0.66, 95% CI 0.54 - 0.77). The discriminative power of RPA to differentiate progressing from non-progressing eyes (AUC = 0.97, 95% CI 0.92 - 1.00) was greater than global cRNFL thickness (AUC = 0.71 95% CI 0.59 - 0.82, p<0.0001) and equivalent to sectoral cRNFL regression (AUC = 0.97, 95% CI 0.92 - 1.00). Kaplan-Meier survival analysis showed progression was detected 8.7 months sooner by RPA than global cRNFL linear regression (p<0.0001) in progressing eyes, however not compared with sectoral cRNFL (p=0.06). RPA showed substantial agreement with the presence of significant thinning on sectoral cRNFL linear regression (kappa=0.715, 95% CI 0.578-0.853). CONCLUSIONS: RPA has been shown to provide accurate and intuitive at-a-glance data analysis and presentation that improves inter-observer agreement and may aid early diagnosis of glaucomatous disease progression.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item