Table 5. Image binarization protocol in Sonoda et al. vs current study.
Sonoda et al. (2014)30 | Sonoda et al. (2015)31 | Current study | |
---|---|---|---|
Study sample size | 15 eyes of 15 subjects | 180 eyes of 180 subjects | 345 eyes of 345 subjects |
Choroidal area measured | 1.5 mm | 7.5 mm | 1.5 mm |
Location of measurement | Centered on fovea, 1500 um | Entire raster scan, 7500 um | Centered on fovea (1500 um) due to the segmental nature of choroidal blood supply as described by Hayreh et al.37 |
Pre-selection of vessels | 3 choroidal vessels with lumens >100 mm were randomly selected and the average reflectivity of these areas was determined by the software | Used autolocal threshold techniques to allow binarization of smaller choroidal vessels or choriocapillaris. | |
Brightness adjustment | Average brightness was set as the minimum value | Brightness was not adjusted as it would reduce the contrast between luminal and stromal areas and possibly affect the autolocal threshold. | |
Order of binarization | Region of interest selected prior to image binarization | To get a clear view of the choroid-scleral interface, image binarization was performed prior to area selection. | |
Image segmentation time | ~5 minutes per image | ~1 minute per image |