Abstract
Purpose :
To present application of the modified Knox-Thompson phase-based method for extraction of the light-evoked functional responses of human retina imaged with optical coherence tomography (OCT).
Methods :
The human volunteers were imaged in vivo with a clinical-grade OCT retinal imaging system, equipped with a controlled light stimulator, to allow ORG experiments. Before the testing volunteers were dark-adapted for over 5 minutes. The clinical-grade OCTs cannot resolve individual photoreceptors and instead probe a group of cells, resulting in limited phase information due to temporal speckle decorrelation. Implementation of the modified Knox-Thompson method allowed us to successfully extract phase-based ORG signals from the outer segment band (between IS/OS and COST).
Results :
The phase-based ORG signals extracted from the same data sets, although more sensitive to retina motion and more computational demanding than ORG intensity-based methods, provided an order of magnitude higher sensitivity of detecting changes in retina layers positions. The cross-correlation of the phase difference between the measured layers of the object allowed identification of the intervals with strong phase correlation, thus allowing extraction of the ORG signal in a clinical setting.
Conclusions :
Successful implementation of phase-based ORG signal analysis provided an order of magnitude higher sensitivity of detecting changes in retina layer positions. ORG signal analysis using clinical-grade raster scanning OCT systems is now possible, which should open a path to clinically friendly ORG probing.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.