McLaren J W, Patel S V, Nau C B, Bourne W M
Department of Ophthalmology, Mayo Clinic College of Medicine, Rochester, MN 55905, USA.
J Microsc. 2008 Jan;229(Pt 1):21-31. doi: 10.1111/j.1365-2818.2007.01870.x.
Cell density in the corneal stroma is typically determined by counting the number of bright objects, presumably keratocyte nuclei, in images from clinical confocal microscopy. We present a program that identifies bright objects and counts those that most likely represent cells. Selection variables were determined from 125 normal corneas with cell densities that had been assessed manually. The program was tested on 17 corneas of patients before and at several intervals to 5 years after laser in situ keratomileusis (LASIK) surgery. In these corneas, which showed a decrease in cell density after surgery, the program identified cells as well as human observers did.
角膜基质中的细胞密度通常是通过在临床共聚焦显微镜图像中计数明亮物体(推测为角膜细胞细胞核)的数量来确定的。我们提出了一个程序,该程序可以识别明亮物体并对最有可能代表细胞的物体进行计数。选择变量是从125个细胞密度已通过手动评估的正常角膜中确定的。该程序在17例接受准分子原位角膜磨镶术(LASIK)手术的患者的角膜上进行了术前及术后数年(至5年)几个时间点的测试。在这些术后细胞密度降低的角膜中,该程序识别细胞的能力与人类观察者相当。