Abstract
Purpose: :
Spectral Domain Optical Coherence Tomography is a high-speed, high-resolution, non invasive method that allows visualization of structures in the eye. Application of SD-OCT technology in routine clinical practice is relatively new and its clinical interpretation along with other traditional imaging methods is being studied. Hence it is of high value to correlate the OCT image with modalities such as Fundus Photography, Fluorescein Angiography (FA), and Fundus Autofluorescence imaging (FAF). Automated registration algorithms have been developed to register the reconstructed OCT fundus image to these other modalities.
Methods: :
Existing algorithms were used to segment the Internal Limiting Membrane (ILM) and Retinal Pigment Epithelium (RPE) layers from the Cirrus HD-OCT data set. These segmentations were then used to obtain an en face view of the OCT volume which enhances the contrast of the blood vessels compared to the background. The blood vessels were segmented from the en face OCT image and the fundus image and an algorithm based on aligning the blood vessel segmentations was used to register the images.
Results: :
The registration transforms the OCT fundus image and the other imaging modalities into the same co-ordinate frame. This allows correlating lesions amongst these different modalities, and allows for the cross-sectional analysis of pathological features visualized on traditional planar imaging. Registration of images provides additional diagnostic information over what can be achieved from each imaging modality independently. For example, the overlay of derived OCT thickness maps on the fundus images gives valuable clinical information for diagnosis, assessment of disease progression, quantitative monitoring of treatment effect, or surgical planning for focal laser treatment.
Conclusions: :
The described method helps to automatically register en face images from OCT data cubes to images obtained from other ophthalmic imaging modalities such as color fundus, FA and FAF. This capability can form a valuable additional diagnostic resource for the clinician and researcher by enabling accurate quantitative and qualitative comparisons between OCT and other imaging modalities.
Keywords: imaging/image analysis: clinical • retina • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)