A Study On Ic Disc

1181 Words5 Pages
ic disc is excluded, as not having the vessels attained their status inside the retina yet. Within this area, the six largest veins and the six largest arteries are measured, following an iterative procedure of pairing up the largest vessels with the smallest ones, until a final single number is obtained. All the values are entered in eq.6 and eq.7 for arterioles and venules respectively. The final value for the vein is termed central retinal vein equivalent (CRVE) and the respective final value for the artery is termed central retinal artery equivalent (CRAE). The ratio CRAE/CRVE is known as arterio-venous ratio. Arterioles:Wˆ =0.88∗ (W12+W2) (6) Veins:Wˆ =0.95∗ (W12+W2) (7) where wˆ is the estimate of the parent trunk arteriole or venule…show more content…
For the former, the Shapiro-Wilk test was used, and the null hypothesis that the data are normally distributed was not rejected, regardless of the feature under investigation (p-values ranging from 0.30 to 0.56). Similarly for the sphericity, the Mauchlys test was used, which again failed to reject the null hy- pothesis that the assumption of sphericity is met (p-values ranged from 0.16-0.39). Although ANOVA is robust in marginal violations of normality, it still suffers from sphericity, which if present, causes the test to become unstable i.e. leads to an increase of Type I error;that is, the likelihood of detecting a statistically significant result when there is not one. 3.2.1 Mixed Model Filter As mentioned above, in order to account for the different way that the features are measured, a mixed model factorial/nested design has been developed in MATLAB 2015b version, in which all the local measurements are used in the statistical anal- ysis. As can be seen in fig. 4, in the case of widths and angles, we have multiple measurements within each subject, in a nested formation. That means that all these observations are not independent, and thus that needs to be taken into account. Us- ing this design, each measurement in P1jM1k, where 1 is the first case, e.g. pre-DR group, j the corresponding patient and k the specific measurement, is
Open Document