Purpose:
To determine whether enhanced near-infrared (NIR) fundus images using novel imaging algorithms would allow for improved characterization of subretinal pathology such as drusen and geographic atrophy (GA). To evaluate whether this methodology can be used to objectively follow the progression of drusen and GA.
Methods:
Eyes of 10 subjects with subretinal pathology including drusen and GA were studied. Red free, NIR, and color fundus photos were captured for each patient using the Topcon TRC 501X tri-functional fundus camera and the Heidelberg Spectralis OCT/SLO. A novel matrix edge field algorithm was developed to study the NIR images. NIR images were denoised, and the edge map images were extracted. A segmentation algorithm was applied to the enhanced NIR image to facilitate calculation of the area and perimeter of geographic atrophy lesions.
Results:
Enhanced NIR images improved detection of subretinal lesions acquired in the NIR wavelength. Image enhancement and segmentation algorithms are particularly suited to detect drusen and to delineate geographic atrophy. The algorithms were applicable to NIR images acquired by both the Topcon and Heidelberg Spectralis. Comparison of enhanced Topcon and Heidelberg NIR images via manual registration showed good overlay, suggesting good correlations between the two different imaging modalities.
Conclusions:
Enhanced NIR images can provide clinicians and scientists with an additional en face imaging modality for improved characterization of subretinal pathology in subjects with AMD. The segmentation algorithm for geographic atrophy and drusen may be used to follow the progression of these lesions over time.
Keywords: age-related macular degeneration • imaging/image analysis: clinical • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)