Fig. 6 | Scientific Reports

Fig. 6

From: Looking outside the box with a pathology aware AI approach for analyzing OCT retinal images in Stargardt disease

Fig. 6

(a) B-scan image; (b) ground truth segmentation; (c) prediction using the standard method; (d) prediction using the pathology-aware method. (e-h) Corresponding zoomed-in views of (a-d), respectively. The retinal sublayers IR, ONL, PR-IS, PR-OS, and RPE are colored from top to bottom with yellow, dark blue, red, green, and cyan. The cyan box indicates the severely affected region where the segmentation is not performed. The white box shows the zoomed-in area. The arrows point to the areas where the standard method does not perform well.

Back to article page