Conventional OCT artifacts in the human sclera revealed by polarization-sensitive OCT

Abstract

Purpose : Intensity images by conventional optical coherence tomography (OCT) are known to accompany artifacts depending on the polarization state of the detected light from the object; polarization-sensitive OCT (PS-OCT) is theoretically free of such artifacts and expected to provide more accurate intensity images. The purpose of this study is to examine the clinical relevance of the artifact-free feature of PS-OCT by observing the posterior segment of the human eye.

Methods : In this exploratory study, 34 eyes of 17 patients with age-related macular degeneration, glaucoma, or high myopia, who visited the University of Tokyo Hospital, were included. We used a prototype device of PS-OCT, which acquires all 2x2 Jones-matrix elements with four detectors measuring two orthogonal polarizations from the eye for two incident orthogonal polarizations. Artifact-free intensity images were obtained by averaging four kinds of conventional intensity images output by each detector. Birefringence images of the eye were obtained by analyzing the Jones matrices. The artifacts contained in the four kinds of intensity images were examined by human inspection comparing with the artifact-free intensity images and birefringence images. Additionally, birefringence at the artifact sites was quantified as phase retardation.

Results : Artifacts were observed in at least one of the four intensity images in 26 eyes (76.5%) of 15 patients (88.2%), and in all of the four intensity images In 16 eyes (47.1%) of 11 patients (64.7%). These artifacts were recognizable as vessels, layers, or masses in the sclera. The artifact sites generally had strong birefringence in appearance, with the average value of 0.251 ± 0.085. Significantly stronger birefringence was measured in the sites where artifacts were observed in all of the four intensity images than in the other artifact sites (0.284 ± 0.019, 0.199 ± 0.024, p = 0.0099).

Conclusions : The result provided partial evidence to the theory that tissues with stronger birefringence like the sclera tend to cause OCT artifacts. Since intensity images by conventional OCT are obtained similarly to one of the four intensity images in principle, PS-OCT is considered to provide more accurate in vivo image of human tissues than conventional OCT, thus clinical benefit, especially when observing fibrous tissues including the sclera.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.