Purchase this article with an account.
S.M. Meuer, S.E. Moss, R. Klein, B.E. K. Klein; Detecting age–related macular degeneration in digital images vs film images . Invest. Ophthalmol. Vis. Sci. 2004;45(13):3046.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: To compare gradings of lesions associated with age–related macular degeneration (AMD) from digital and stereoscopic film images. Methods: Sixty two people were recruited to have retinal images of the disc and the macula captured in both eyes through nonpharmacologically dilated pupils (NPD) and pharmacologically dilated pupils (PD) using a 45° digital camera (6.3 megapixels). In addition, 30° stereoscopic retinal film images were taken through pharmacologically dilated pupils of the same eyes (N=124). All images were graded for drusen size, type and area, pigmentary abnormalities, geographic atrophy and exudative lesions by masked graders using the Wisconsin Age–Related Maculopathy Grading Scheme. Exact agreement and unweighted kappa scores were calculated for gradings resulting from digital and film images. Results: When categorizing AMD as none, early and late, agreement was best between digital (PD) and film images both taken through a dilated pupil (exact agreement (EA) = 91%, kappa = 0.85, standard error (SE) = 0.05). Agreement for grades from NPD images compared to dilated film images was also good (EA = 80%, kappa = 0.69, SE = 0.06) as was agreement for the grades from digital images captured both ways (NPD vs PD – EA = 85%, kappa = 0.77, SE = 0.05).
This PDF is available to Subscribers Only