Purchase this article with an account.
James Tian, Brenton Keller, Hoan Ngo, Nicole M Fuerst, S. Tammy Hsu, Ryan P McNabb, Cynthia A Toth, Joseph A. Izatt, Anthony N Kuo; Development of microscope-integrated optical coherence tomography (MIOCT) generated topography maps for corneal astigmatic axis detection. Invest. Ophthalmol. Vis. Sci. 2018;59(9):1347. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Corneal surgery can result in significant post-operative astigmatism that is difficult to assess and address intraoperatively. MIOCT, a high resolution intrasurgical imaging method, can allow for generation of topography data to assess intraoperative astigmatism and potentially improve surgical outcomes. Here, we show initial results of MIOCT based topography with display of steep and flat axes in both cornea phantoms and an ex vivo cornea.
Plastic ellipsoid phantom corneas were 3D printed with known astigmatisms of 2.8, 5, 10, and 22.5 diopters (D). All 4 were imaged by MIOCT with the major axis placed at 0-90° in steps of 22.5°. An ex vivo cornea in an artificial anterior chamber was imaged after limbal relaxing incisions (LRIs) at 0° and 180° and then after suture closure as shown in Fig 1, row 2&3. The ex vivo cornea was imaged at 3 positions 45° apart.Volumes were taken with a 100kHz 1050nm swept source MIOCT (4.63mm (depth) and 10x10mm (width x length) with a sampling of 800x300x100 pixels).The epithelium was automatically segmented in all volumes; manual correction in the ex vivo cornea volume was done as needed. The corneal surfaces were dewarped, centered, and fit to a 6th order Zernike polynomial in MATLAB to calculate curvatures per ANSI standards. The primary outcome was the comparison of the calculated to reference angles of the axes of astigmatism, analyzed with a t-test.
The mean differences between the reference and calculated angles for the phantoms were 1.2° (p=0.40) for the 22.5D astigmatic phantom, 1.8° (p=0.24) for the 10 D phantom, 2° (p=0.34) for the 5D phantom, and 7° (p=0.30) for the 2.8D phantom.In the ex vivo cornea, MIOCT generated topography showed flattening in the axis of the paired LRIs and subsequent steeping in the axis of the compression sutures. Rotating the cornea to 3 positions 45° apart, MIOCT topography angles differed from the reference angles by a mean of 4.7° (p=0.17).
MIOCT generated topography maps showed steep axes not significantly different from reference values in phantoms and ex vivo corneas. With improvements in calibration and processing speed, this has potential for clinical use in corneal transplantations and other corneal astigmatism management.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
Fig 1. Microscope view, MIOCT topography map, and overlay in 1. phantoms and ex vivo cornea with 2. LRI and 3. suture closure.
This PDF is available to Subscribers Only