Suppr超能文献

增殖性糖尿病视网膜病变的分形几何:对视网膜血管生成诊断及过程的启示

The fractal geometry of proliferative diabetic retinopathy: implications for the diagnosis and the process of retinal vasculogenesis.

作者信息

Daxer A

机构信息

Department of Ophthalmology, University Hospital, Innsbruck, Austria.

出版信息

Curr Eye Res. 1993 Dec;12(12):1103-9. doi: 10.3109/02713689309033508.

Abstract

10 retinal vessel patterns with neovascularisation at or near the optic disk (NVD) from eyes of patients with diabetic retinopathy were compared with vascular patterns from 14 normal eyes. The vascular patterns were taken from low angle fundus photographs. After digitizing, the fractal dimensions were calculated by means of the density-density correlation function method. The fractal dimension was found to be significantly higher for vessel patterns with NVD [D = 1.845 +/- 0.056 (m +/- sd)] as compared with the normal control group (D = 1.708 +/- 0.073) (p < 0.001). The fractal dimension of 1.8 appears to be a cutoff value. Higher values may indicate proliferative changes. Under these conditions the sensitivity of the method for the detection of NVD > or = Grade 3 in the Early Treatment Diabetic Retinopathy Study (ETDRS) grading system is 90%. The presence of such NVD in an eye is a "high risk characteristic" for severe visual loss, which requires panretinal laser treatment. Fractal analysis is therefore a possible new strategy for computer assisted "automated" detection and quantification of proliferative diabetic retinopathy. The fractal dimension of the new vessels suggests possible mechanisms involved in retinal vasculogenesis.

摘要

将10例糖尿病视网膜病变患者眼部视盘或其附近出现新生血管(NVD)的视网膜血管模式与14只正常眼睛的血管模式进行了比较。血管模式取自低角度眼底照片。数字化后,通过密度-密度相关函数法计算分形维数。发现有NVD的血管模式的分形维数[D = 1.845±0.056(平均值±标准差)]显著高于正常对照组(D = 1.708±0.073)(p < 0.001)。1.8的分形维数似乎是一个临界值。更高的值可能表明存在增殖性变化。在这些条件下,该方法在早期糖尿病视网膜病变研究(ETDRS)分级系统中检测NVD≥3级的敏感性为90%。眼中出现此类NVD是严重视力丧失的“高风险特征”,需要进行全视网膜激光治疗。因此,分形分析是计算机辅助“自动化”检测和量化增殖性糖尿病视网膜病变的一种可能的新策略。新生血管的分形维数提示了视网膜血管生成可能涉及的机制。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验