Abstract
Purpose :
To evaluate the potential lateral resolution of a commercial optical coherence tomography (OCT) system and to optimize this capability.
Methods :
The lateral point spread (PSF) of a Heidelberg Spectralis was evaluated by a variety of methods and was been modelled using Zemax software to investigate its optical lateral resolution.
Results :
The spot size on the retina for the Heidelberg SPECTRALIS was quoted in older literature to be 14 micrometers, and this was assumed to represent the lateral resolution. Measurement of the lateral PSF using diamond dust and the validated target, evaluation of a USAF target and analysis of speckle size all independently supported estimates of the optical lateral resolution of about 6 micrometers in eyes with good optical quality, which approaches the diffraction limit for pupil diameter of 2 mm. Zemax modeling corroborated these results. Considering the Nyquist sampling guidelines, an isotropic scan spacing of 2 micrometers was performed in volume scans in patients under an IRB approved protocol. These scans show improvement in detail visible in the retina, as compared with the conventional ‘high-resolution’ scan protocol (6 µm spacing) of the Heidelberg Spectralis.
Conclusions :
Although a calculation of the spot size of the illumination beam was reported in the past, the actual lateral resolution, as inferred from the lateral PSF of the instrument, is much better. The clinically used scan parameters undersample the potential lateral resolution of the instrument. Improved image quality could be obtained by a denser scan protocol consistent with the Nyquist theorem. This set of considerations, the underestimate of the lateral resolution combined with undersampling, may well apply to all commercial clinical OCT instruments. Increasing the sampling density would be a motivation to increase the scanning speed to make these higher resolution images practical in the clinic.
This is a 2021 Imaging in the Eye Conference abstract.