1 Instituto de Biomecánica de Valencia, Universitat Politècnica de València, Valencia, Spain.
2 Departament de Cirurgia, Universitat de València, Hospital Universitari i Politècnic la Fe, Valencia, Spain.
Am J Rhinol Allergy. 2019 Mar;33(2):121-128. doi: 10.1177/1945892418811324. Epub 2018 Nov 20.
The respiratory epithelium is frequently infected by the respiratory syncytial virus, resulting in inflammation, a reduction in cilia activity and an increase in the production of mucus.
In this study, an automatic method has been proposed to characterize the ciliary motility from cell cultures by means of a motility index using a dense optical flow algorithm. This method allows us to determine the ciliary beat frequency (CBF) together with a ciliary motility index of the cells in the cultures. The object of this analysis is to automatically distinguish between normal and infected cells in a culture.
The method was applied in 2 stages. It was concluded from the first stage that the CBF is not a good enough indicator to discriminate between the control and infected cultures. However, the ciliary motility index does succeed in discriminating between the control and infected cultures using the t test with a value t = 6.46 and P < .001. In the second stage, it has been shown that the ciliary motility index did not differ significantly between patients, and the analysis of variance test gives α = 0.05, F = 1.61, P = .20. A threshold for this index has been determined using a receiver operating characteristics analysis that gives an area under the curve of 0.93.
We have obtained a ciliary motility index that is able to discriminate between control and infected cultures after the eighth postinfection day. After infection, there is a rapid cilia loss of the cells and the measured CBF corresponds to the remaining noninfected cells. This is why the CBF does not discriminate between the control and the infected cells.
呼吸道合胞病毒经常感染呼吸道上皮细胞，导致炎症、纤毛活动减少和黏液分泌增加。
本研究提出了一种自动方法，通过使用密集光流算法的运动指数来表征细胞培养物中的纤毛运动。该方法允许我们确定纤毛摆动频率（CBF）以及培养物中细胞的纤毛运动指数。该分析的目的是自动区分培养物中的正常细胞和感染细胞。
该方法分两个阶段应用。第一阶段的结论是，CBF 不足以区分对照和感染培养物。然而，使用 t 检验，纤毛运动指数成功地区分了对照和感染培养物，t 值为 6.46，P 值<.001。在第二阶段，已经表明纤毛运动指数在患者之间没有显著差异，方差分析检验给出α=0.05，F=1.61，P=0.20。使用接收者操作特性分析确定了该指数的阈值，曲线下面积为 0.93。
我们获得了一个纤毛运动指数，能够在感染后第 8 天区分对照和感染培养物。感染后，细胞的纤毛迅速丧失，测量的 CBF 对应于未感染的剩余细胞。这就是为什么 CBF 不能区分对照和感染细胞的原因。
