Purpose:
To demonstrate a novel processing algorithm for extended range Fourier Domain Optical Coherence Tomography (FDOCT) optimized for real–time 2D imaging of the entire human anterior segment in vivo.
Methods:
FDOCT images are corrupted by unresolved positive and negative distances, causing a symmetric image artifact. Consequently, imaging is typically performed with the entire sample to one side of the reference, thus utilizing only half of the sample depth. Full–range imaging, in which positive and negative distances are resolved, can be achieved by indirectly measuring the complex component of the interferometric OCT signal, described in the literature by stepping through multiple 90o phase shifts in the reference arm. We have developed a quadrature projection algorithm for the removal of symmetric artifact in FDOCT imaging which automatically compensates for mis–calibrated phase shifts in 90o stepped interferometry, and also facilitates development of non–90o phase shift techniques. We demonstrate this algorithm on a swept source FDOCT (o = 1310nm, Δ = 89nm) platform using a Michelson–type interferometer constructed from a 3x3 fused fiber coupler which provides three phase shifted images simultaneously with roughly 120o phase separation. Image acquisition and processing is performed using custom software for real time display of extended range cross sectional scans. The line rate was set by the swept source scan rate (6.66kHz), and the number of lines per scan distance was software adjustable.
Results:
The extended range swept–source FDOCT system boasts a sample depth of 6.6 mm with 110dB sensitivity near DC and a roll–off of –7dB at the maximum depth. The symmetric artifact was suppressed by >30dB, and the axial resolution of the system was 9 microns. Full range images were acquired and displayed in real time on a standard computer. Sample motion did not corrupt the full–range images of anterior segment acquired in vivo on human volunteers (Fig. 1).
Conclusions:
A new processing algorithm is demonstrated for high speed full range FDOCT, allowing for cross–sectional imaging of the entire depth of the anterior segment.
Keywords: anterior segment • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • image processing