Department of Ophthalmology, Daegu Veterans Health Service Medical Center, Daegu, Korea.
Department of Ophthalmology, Yeungnam University College of Medicine, 170 Hyunchung-ro, Nam-gu, 42815, Daegu, Korea.
BMC Ophthalmol. 2021 Mar 20;21(1):141. doi: 10.1186/s12886-021-01900-y.
Ultra-widefiled (UWF) retinal images include significant distortion when they are projected onto a two-dimensional surface for viewing. Therefore, many clinical studies that require quantitative analysis of fundus images have used stereographic projection algorithm, three-dimensional fundus image was mapped to a two-dimensional stereographic plane by projecting all relevant pixels onto a plane through the equator of the eye. However, even with this impressive algorithm, refractive error itself might affect the size and quality of images theoretically. The purpose of this study is to investigate the effects of refractive power on retinal area measurements (quantification) using UWF retinal imaging (Optos California; Dunfermline, Scotland, UK).
A prospective, interventional study comprised 50 healthy eyes. UWF images were acquired first without the use of a soft contact lens (CL) and then repeated with six CLs (+ 9D, +6D, +3D, -3D, -6D, and - 9D). Using stereographically projected UWF images, the optic disc was outlined by 15-17 points and quantified in metric units. We divided the subjects into three groups according to axial length: Groups A (22-24 mm), B (24-26 mm), and C (≥ 26 mm). The primary outcome was percentage change before and after use of the CLs. Secondary outcome was proportion of subjects with magnification effects, maximal changes > 10 %.
The study population was 6, 28, and 16 eyes in each group. Overall changes for the measured area were not significantly different in the whole study population. Group C had a larger proportion of magnification effects compared to Groups A and B (50.0 %, 0 %, and 3.6 %, P = 0.020). Measured area with plus lenses was significantly higher in Group C (P < 0.001).
The use of CLs might affect quantification of eyes with long axial length when using UWF images. Ophthalmologists should consider refractive error when measuring area in long eyes.
超广角（UWF）视网膜图像在投影到二维表面进行观察时会产生显著的失真。因此，许多需要对眼底图像进行定量分析的临床研究都使用了立体投影算法，即将所有相关像素通过眼球赤道投射到一个平面上，从而将三维眼底图像映射到二维立体平面上。然而，即使有了如此令人印象深刻的算法，折射误差本身也可能在理论上影响图像的大小和质量。本研究旨在使用超广角视网膜成像（英国苏格兰邓弗姆林的 Optos California）探讨屈光力对视网膜面积测量（量化）的影响。
一项前瞻性、干预性研究纳入了 50 只健康眼。首先在不使用软性隐形眼镜（CL）的情况下获取超广角图像，然后重复使用 6 个 CL（+9D、+6D、+3D、-3D、-6D 和-9D）。使用立体投影超广角图像，通过 15-17 个点勾勒视盘，并以公制单位进行量化。我们根据眼轴长度将受试者分为三组：A 组（22-24mm）、B 组（24-26mm）和 C 组（≥26mm）。主要结局为使用 CL 前后的百分比变化。次要结局为具有放大效果的受试者比例，最大变化>10%。
每组的研究人群分别为 6、28 和 16 只眼。整个研究人群的测量面积总体变化无显著差异。与 A 组和 B 组相比，C 组的放大效果比例更大（50.0%、0%和 3.6%，P=0.020）。C 组使用正透镜时测量面积显著升高（P<0.001）。
使用 CL 可能会影响长眼轴的超广角图像的量化。眼科医生在测量长眼的面积时应考虑屈光误差。
