Purchase this article with an account.
Christopher Schuetze, Matthias Bolz, Ramzi Sayegh, Bernhard Baumann, Michael Pircher, Erich Götzinger, Christoph K. Hitzenberger, Ursula Schmidt-Erfurth; Proof Of Principle Of Lesion Size Detection In Geographic Atrophy Using Polarization-sensitive Spectral Domain Optical Coherence Tomography And Correlation To Conventional Imaging Techniques. Invest. Ophthalmol. Vis. Sci. 2011;52(14):1244.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To investigate the reproducibility of automated lesion size detection in patients with geographic atrophy (GA) using polarization-sensitive spectral domain OCT (PS-SD-OCT) and to compare findings with conventional imaging techniques such as scanning laser ophthalmoscopy (SLO) and fundus autofluorescence (FAF) imaging.
10 eyes of 7 patients with GA were examined by PS-SD-OCT providing selective identification of the retinal pigment epithelium (RPE) layer. A novel segmentation algorithm was applied to automatically detect and quantify areas of RPE atrophy. Five examinations per eye were performed at a one day visit. The reproducibility of the segmentation algorithm was evaluated. Measured lesion sizes were correlated with manually assessed areas of GA in SLO and FAF imaging in order to validate the clinical applicability of PS-SD-OCT in GA analysis.
Mean GA lesion size of all patients measured by PS-SD-OCT was 4.71mm2 (±2.97). Mean percentage of standard deviation in all eyes in relation to mean values of all atrophic RPE lesions investigated was 4.33% (±2.13). The mean variation of RPE lesion size as validated from the repeatability measurements was 0.48 mm², implying a solid reproducibility of GA lesion size evaluation in PS-SD-OCT. GA area in SLO measured 4.49 mm2 (±2,71). Mean lesion size in SLO images was 4.7% smaller than determined by PS-SD-OCT, revealing a reliable correlation. Mean GA area of all eyes assessed by FAF was 4.85mm2 (±: 2.73) and correlated excellent to automatically detected PS-SD-OCT images with a relative deviation of 3.0% regarding lesion size quantification.
Evaluation of PS-SD-OCT demonstrated high reproducibility of lesion size determination in patients with GA. Results correlated well with conventional techniques such as SLO and FAF imaging. PS-SD-OCT may therefore be a valuable and specific imaging modality for automated determination of GA lesion size in scientific studies and clinical practice.
This PDF is available to Subscribers Only