A Quantitative Analysis System for UHR-OCT

Abstract

 

Purpose:

 

Creating a quantitative 3D analysis system for UHR-OCT, which would include three features. Firstly, a 3D measuring tool which measures the distance between two user defined points, was to be designed. Secondly, a 3D retinal thickness-mapping tool was to be implemented. Thirdly, a sub-routine which is able to perform a 3D segmentation of the data stack and convert surfaces into "solid" object triangle meshes, was to be programmed.

 

Methods:

 

Using B-mode scan data sets, consisting of 60 cross sections, from a 3D UHR-OCT with an axial resolution of 2-3 µm and the raytracing package Cinema 4D XL 9.102 (Maxon Computer Inc., Friedrichsburg, Germany), several procedures for 3D UHR-OCT rendering were developed. A subroutine was designed, which lets the user define the 3D positions of two callipers which each have three axes. The distance between the two callipers is displayed on the screen in real time. A second subroutine was designed which displays a 3D reconstruction of the volume stack with a colour coding indicating each voxel’s distance from the retinal pigment epithelium layer. Using Cinema 4D's "vectorizer" capability, a third sub-routine was designed which recognizes areas of similar intensity in each B-mode scan and converts this information into splines. All splines from each of the 60 B-mode scans are subsequently joined into a triangle mesh.

 

Results:

 

The measuring tool delivers reliable measurements in all three axes. The retinal thickness mapping system offers new insight into the spatial dimensions of retinal pathologies. The segmentation system effectively recognizes extra- as well as intra-retinal structures, but cannot yet differentiate subtle differences in neighbouring retinal layers.

 

Conclusions:

 

It was possible to create several subroutines, which significantly increase the amount of information that be can be gathered from a UHR-OCT examination.  

 

Keywords: retina • imaging/image analysis: clinical • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)