Purchase this article with an account.
D. W. Wong, J. Liu, Z. Zhang, J. Lim, N. Tan, T. Wong, T. Aung, M. Sandar, H. Li, S. Lu; Performance of an Automatic Software to Determine the Cup-to-Disc Ratio From Digital Fundus Photographs for Glaucoma Diagnosis. Invest. Ophthalmol. Vis. Sci. 2009;50(13):311.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To determine the performance and validity of an automated cup-to-disc ratio detection software from digital retinal photographs.
We tested the performance of ARGALI (Automatic cup-to-disc Ratio measurement system for Glaucoma detection and AnaLysIs) software, an intelligent imaging system to detect the optic cup and disc and its ratio (CDR) automatically from digital fundus images. The system was evaluated using a sample set of 98 fundus images taken from normative individuals aged 40-80 years from the Singapore Malay Eye Study (SiMES). The system was compared to clinical determination of the CDR based on assessment from a senior glaucoma specialist.
The mean difference between the ARGALI CDR and the clinical CDR was less than 0.10 CDR. Of the 98 images, 85 (86.7%) had an ARGALI versus clinical CDR difference of less than 0.15, and 91 (92.8%) had a difference of less than 0.2 CDR, a figure previously reported as the inter-observer variability for a clinical CDR assessment.
Based on the ARGALI system, 92.8% of the automatically evaluated images were within reported clinical inter-observer variability. These data support further development of ARGALI into an automated imaging tool for the early detection of glaucomatous optic neuropathy.
This PDF is available to Subscribers Only