Aslam Tariq M, Patton Niall, Rose Christopher J
Princess Alexandra Eye Pavilion, Chalmers St, Edinburgh, UK.
BMC Ophthalmol. 2006 Aug 23;6:30. doi: 10.1186/1471-2415-6-30.
This paper presents and tests a comprehensive computerised system of analysis of digital images of posterior capsule opacification (PCO). It updates and expands significantly on a previous presentation to include facilities for selecting user defined central areas and for registering and subsequent merging of images for artefact removal. Also, the program is compiled and thus eliminates the need for specialised additional software. The system is referred to in this paper as the open-access systematic capsule assessment (OSCA). The system is designed to be evidence based, objective and openly available, improving on current systems of analysis.
Principal features of the OSCA system of analysis are discussed. Flash artefacts are automatically located in two PCO images and the images merged to produce a composite free from these artefacts. For this to be possible the second image has to be manipulated with a registration technique to bring it into alignment with the first. Further image processing and analysis steps use a location-sensitive entropy based texture analysis of PCO. Validity of measuring PCO progression of the whole new system is assessed along with visual significance of scores. Reliability of the system is assessed.
Analysis of PCO by the system shows ability to detect early progression of PCO, as well as detection of more visually significant PCO. Images with no clinical PCO produce very low scores in the analysis. Reliability of the system of analysis is demonstrated.
This system of PCO analysis is evidence-based, objective and clinically useful. It incorporates flash detection and removal as well as location sensitive texture analysis. It provides features and benefits not previously available to most researchers or clinicians. Substantial evidence is provided for this system's validity and reliability.
本文介绍并测试了一个用于分析后囊膜混浊(PCO)数字图像的综合计算机系统。它在之前的展示基础上进行了重大更新和扩展,包括选择用户定义的中心区域以及对图像进行配准和后续合并以去除伪影的功能。此外,该程序已编译完成,因此无需专门的额外软件。本文将该系统称为开放获取系统囊膜评估(OSCA)。该系统旨在基于证据、客观且开放可用,对当前的分析系统进行了改进。
讨论了OSCA分析系统的主要特征。在两张PCO图像中自动定位闪光伪影,并将图像合并以生成一张无这些伪影的合成图像。为实现这一点,必须使用配准技术对第二张图像进行处理,使其与第一张图像对齐。进一步的图像处理和分析步骤使用基于位置敏感熵的PCO纹理分析。评估了整个新系统测量PCO进展的有效性以及分数的视觉显著性。评估了该系统的可靠性。
该系统对PCO的分析显示出能够检测PCO的早期进展以及检测出更具视觉显著性的PCO。在分析中,无临床PCO的图像得分非常低。证明了分析系统的可靠性。
这个PCO分析系统基于证据、客观且在临床上有用。它结合了闪光检测和去除以及位置敏感纹理分析。它提供了大多数研究人员或临床医生以前无法获得的功能和优势。为该系统的有效性和可靠性提供了大量证据。