Zhao Wen-Juan, Duan Fang, Li Zhong-Ting, Yang Hua-Jun, Huang Qiang, Wu Kai-Li
Zhongshan Ophthalmic Center, State Key Laboratory of Ophthalmology, Sun Yat-sen University, Guangzhou 510060, Guangdong Province, China.
Int J Ophthalmol. 2014 Feb 18;7(1):71-6. doi: 10.3980/j.issn.2222-3959.2014.01.13. eCollection 2014.
To develop an image-based objective method to precisely evaluate regional ocular bulbar injection.
Six healthy adult volunteers were photographed in four orientations (superior, inferior, nasal and temporal sides) with and without stimulating eye drops. Six line segments (covering 30°) were drawn 4mm away from the limbus on each image using ImageJ software. The graph peaks, which were derived from the areas under the line segments and corresponded to the cross-sectional grey-level of the vessels, were analyzed to obtain peak area, peak height/width (PH/PW), and peak numbers. Different-sized areas were selected to calculate the pixels based on the edge-detection algorithm. Also, conjunctival and superficial scleral vessels were analyzed separately.
This method had a smaller coefficient of variation, especially for PH/PW, in all four orientations. Hyperaemia parameters changed the least after challenging in the superior region. Moreover, 95% of the PH/PW ratios were greater than 0.87 in conjunctival vessels and less than 1.00 in superficial scleral vessels. PH/PW significantly increased in conjunctival vessels and changed less in superficial scleral vessels.
A new method of objectively assessing bulbar injection based on ocular surface images was developed. This method can be used to quantify ocular regional injection and to distinguish the superficial scleral and conjunctival vessels.
开发一种基于图像的客观方法,以精确评估眼球局部注射情况。
对6名健康成年志愿者在使用和不使用刺激性滴眼液的情况下,从四个方向(上方、下方、鼻侧和颞侧)进行拍照。使用ImageJ软件在每张图像上距角膜缘4mm处绘制6条线段(覆盖30°)。分析从线段下方区域得出的、对应血管横截面灰度的图形峰值,以获得峰值面积、峰高/峰宽(PH/PW)和峰值数量。基于边缘检测算法选择不同大小的区域来计算像素。同时,分别分析结膜和浅层巩膜血管。
该方法在所有四个方向上的变异系数较小,尤其是对于PH/PW。充血参数在上方区域刺激后变化最小。此外,结膜血管中95%的PH/PW比值大于0.87,浅层巩膜血管中该比值小于1.00。结膜血管中的PH/PW显著增加,而浅层巩膜血管中的变化较小。
开发了一种基于眼表图像客观评估眼球充血的新方法。该方法可用于量化眼部区域充血情况,并区分浅层巩膜和结膜血管。