眼底图像分析系统在原发性开角型青光眼早期诊断中的作用

Effect of fundus image devices for early diagnosis of primary open angle glaucoma

眼底图像分析技术的迅速发展使视神经损害的直接定量检测成为可能，同时因其检测手段客观、结果可靠而日益受到重视。但由于临床医师对眼底图像分析技术如海德堡视网膜断层扫描、相干光断层扫描等在青光眼早期诊断中的意义仍存在认识上的分歧，因此有必要就此问题进行论述。有多数研究结果表明，现有的检测设备不仅在分辨率和重复性方面已经达到临床要求，而且能够较好地区分正常人和青光眼的眼底改变，但现阶段利用视神经结构定量指标来实现对青光眼的早期诊断，其敏感性和特异性尚不尽如人意。其原因是多方面的，如结构指标个体差异较大，缺乏完善的正常人数据库，缺乏达成共识的敏感而特异的诊断指标等。尽管如此，现有的研究趋势表明，利用图像分析技术测量青光眼性视神经损害已显示出令人鼓舞的临床应用前景，临床医师若能充分了解各种图像分析手段的优缺点，这些新的检测技术将成为早期诊断原发性开角型青光眼的有益补充手段。

Evaluation of the optic disc and retinal nerve fiber layer measurements （RNFL） with sterephotography is the current ＂gold standard＂ for the diagnosis of early primary open angle glaucoma （POAG）, but it is still subjective and is dependent too much upon an experienced observer. Substantial variability exists in the interpretation of optic disc change over time. There are consistent evidences that image analysis devices such as optical coherence tomography （OCT） and Heidelberg retina tomograph （ HRT- I1 ） can detect early to moderate glaucoma as well as expert assessment of stereoscopic optic disc and RNFL photographs. Optic disc and RNFL defect can be quantitatively measured by these devices. But the sensitivity and specificity of the measurements obtained with these devices are not good enough for the early diagnosis of POAG at this time due to several reasons, such as individual variety of the structure parameters in normal subjects, lack of powerful normative database, lack of diagnostic parameters which axe good enough both in sensitivity and specificity for detecting early glaucoma damage. However, information obtained from these image analysis devices might be helpful in early POAG diagnosis if clinicians can understand the advantages and limitations of these instruments.