Koprowski Robert, Ambrósio Renato, Reisdorf Sven
Department of Biomedical Computer Systems, University of Silesia, Faculty of Computer Science and Materials Science, Institute of Computer Science, ul. Będzińska 39, Sosnowiec, 41-200, Poland.
Department of Ophthalmology of the Federal University of São Paulo & Rio de Janeiro Corneal Tomography and Biomechanics Study Group of the Instituto de Olhos Renato Ambrósio, Rua Conde de Bonfim 211 / 712, Rio de Janeiro, RJ - CEP: 20520-050, Brasil.
J Biophotonics. 2015 Nov;8(11-12):968-78. doi: 10.1002/jbio.201400137. Epub 2015 Jan 26.
The paper presents an original analysis method of corneal deformation images from the ultra-high-speed Scheimpflug camera (Corvis ST tonometer). Particular attention was paid to deformation frequencies exceeding 100 Hz and their reproducibility in healthy subjects examined repeatedly. A total of 4200 images with a resolution of 200 × 576 pixels were recorded. The data derived from 3 consecutive measurements from 10 volunteers with normal corneas. A new image analysis algorithm, written in Matlab with the use of the Image Processing package, adaptive image filtering, morphological analysis methods and fast Fourier transform, was proposed. The following results were obtained: (1) reproducibility of the eyeball reaction in healthy subjects with precision of 10%, (2) corneal vibrations with a frequency of 369 ± 65 Hz (3) and amplitude of 7.86 ± 1.28 µm, (4) the phase shift within two parts of the cornea of the same subject of about 150°. The result of image sequence analysis for one subject and deformations with a corneal frequency response above 100 Hz.
本文提出了一种来自超高速Scheimpflug相机(Corvis ST眼压计)的角膜变形图像的原创分析方法。特别关注了超过100Hz的变形频率及其在反复检查的健康受试者中的可重复性。总共记录了4200张分辨率为200×576像素的图像。数据来自10名角膜正常志愿者的连续3次测量。提出了一种用Matlab编写的新图像分析算法,该算法使用了图像处理包、自适应图像滤波、形态分析方法和快速傅里叶变换。获得了以下结果:(1)健康受试者眼球反应的可重复性,精度为10%;(2)频率为369±65Hz的角膜振动;(3)振幅为7.86±1.28µm;(4)同一受试者角膜两部分之间的相移约为150°。一名受试者的图像序列分析结果以及角膜频率响应高于100Hz的变形情况。