Purchase this article with an account.
R. W. Sedeek, E. Redenbo, K. Chandra, S. S. Park; The Effect of Utilizing Image Registration Software on Central Macular Thickness Reproducibility Using Commercially Available Spectral Domain Optical Coherence Tomography. Invest. Ophthalmol. Vis. Sci. 2010;51(13):299.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To determine the effect of the new software features of commercially available spectral-domain optical coherence tomography (SD-OCT), in particular the three-point image registration of Cirrus OCT and the eye tracking of Heidelberg OCT, on reproducibility of central macular thickness (CMT) measurement in eyes with normal and pathologic maculae.
Prospective observational study of eyes with normal and pathologic maculae imaged with the same SD-OCT (Cirrus or Heidelberg) multiple times on the same day by the same operator. Macular pathologies included diabetic macular edema, geographic atrophy, cystoid macular edema, and exudative age-related macular degeneration. CMT was determined from each scan using the central 1mm circle of the ETDRS macular thickness map. For Cirrus OCT (v4.5), 20 eyes of 10 subjects with normal macula and 20 eyes of 10 subjects with macular pathology were scanned twice. CMT of the two Cirrus images were compared with and without the three-point image registration software. For Heidelberg OCT, 24 eyes of 13 subjects with normal macula and 12 eyes of 9 subjects with macular pathology were imaged three times, the second image acquired without eye tracking and the third image acquired with eye tracking. Coefficient of reproducibility of CMT (CoR) defined as 1-(difference in CMT between two scans/average CMT of two scans) x 100 was obtained for each eye with and without image registration and compared using the two-tailed Student t-test.
For eyes imaged using Cirrus OCT, the average CMT was 249.1±26.5µm and 314.6±167.4µm for normal and pathologic eyes, respectively. For both normal and pathologic eyes, the CoR was significantly higher with the three-point image registration than without image registration (99.14±0.75 vs. 98.59±1.27, p=0.04 for normal eyes; 95.41±7.05 vs. 94.26±6.56, p=0.02 for pathologic eyes). For eyes imaged with Heidelberg OCT, the average CMT was 273.6±19.9µm and 340.9±113.9µm for normal and pathologic eyes, respectively. For both groups, the CoR was significantly higher with eye tracking than without eye tracking (99.32±0.48 vs. 98.77±0.85; p=0.003 for normal eyes; 98.86±1.60 vs. 94.97±5.14; p=0.01 for pathologic eyes).
The new image registration software, in particular the three-point image registration for Cirrus OCT and the eye tracking for Heidelberg SD-OCT, significantly improves the reproducibility of CMT in eyes with normal and pathologic maculae.
This PDF is available to Subscribers Only