Comparison of reflectivity maps and outer retinal topography in retinal disease by 3-D Fourier domain optical coherence tomography

Abstract

We demonstrate and compare two image processing methods for visualization and analysis of three-dimensional optical coherence tomography (OCT) data acquired in eyes with different retinal pathologies. A method of retinal layer segmentation based on a multiple intensity thresholding algorithm was implemented in order to generate simultaneously outer retinal topography maps and reflectivity maps. We compare the applicability of the two methods to the diagnosis of retinal diseases and their progression. The data presented in this contribution were acquired with a high speed (25,000 A-scans/s), high resolution (4.5 microm) spectral OCT prototype instrument operating in the ophthalmology clinic.

Figures

Fig. 1

a. Schematic diagram of the spectral optical coherence tomography instrument. b. Photograph of the prototype SOCT instrument operating in the ophthalmology clinic.

Fig. 2

a. Image processing procedures applied to OCT data acquired in normal retina: I. definition of the first region of interest (ROI); II. automatic generation of a smooth curve corresponding to the shape of the outer retinal contour (ORC); the order of the polynomial fit is adjusted manually; III. flattening and cropping the cross-sectional images with respect to the ORC and definition of the second ROI; IV. automatic segmentation of the entire set of three-dimensional data and identification of lines representing the basal part of the RPE and the IS/OS junction. b. Contour maps representing the distance between the RPE and the outer retinal contour (RPE topography), ORC and IS/OS (IS/OS topography), and the RPE to IS/OS thickness. c. Generation of reflectivity maps: outer retina is divided into two regions (RDZ-I and RDZ-II) indicated by yellow lines. Axial summation of the RDZ-I and RDZ-II generates corresponding reflectivity maps.

Fig. 3

Three-dimensional data acquired in a healthy eye. Left panel: macula (View 1), right panel: optic disk (View 2).

Fig. 4

Comparison of retinal topography mapping with reflectivity mapping in non-exudative age related macular degeneration. Soft drusen in 76 year-old patient are visualized in the cross-sectional image and five SOCT maps. Full three-dimensional dataset is accessible via OSA ISP (View 3).

Fig. 5

Results of a follow up (30 months) imaging of the 76 year-old patient with soft drusen. Drusen growth can be assessed by comparison of OCT maps with the corresponding maps in Fig. 4. Full three-dimensional dataset is accessible via OSA ISP (View 4).

Fig. 6

Comparison of retinal topography mapping with reflectivity mapping: confluent drusen, 61 year-old patient. Full three-dimensional dataset is accessible via OSA ISP (View 5).

Fig. 7

Results of a follow up (12 months) imaging of the 61 year-old patient with confluent drusen. Full three-dimensional dataset is accessible via OSA ISP (View 6).

Fig. 8

Comparison of retinal topography mapping with reflectivity mapping: choroidal neovascularisation in age related macular degeneration, 66 year-old patient. Full three-dimensional dataset is accessible via OSA ISP (View 7).

Fig. 9

Results of a follow up (22 months) imaging of the 66 year-old patient with choroidal neovascularisation in age related macular degeneration. Full three-dimensional dataset is accessible via OSA ISP (View 8).

Fig. 10

Comparison of retinal topography mapping with reflectivity mapping: elevations of neurosensory retina in the chronic central serous chorioretinopathy, 43 year-old patient. Full three-dimensional dataset is accessible via OSA ISP (View 9).

Fig. 11

Comparison of retinal topography mapping with reflectivity mapping: elevations of neurosensory retina in retinal detachment, 17 year-old patient. Full three-dimensional dataset is accessible via OSA ISP (View 10).