99 per individual were assessed. For automatic analysis (ACCMetrics20 and NFA FIJI21), all good quality images were selected, as more measurements often result in more reliable results combined with a lower risk of selection bias. An example of a representative image analysis is shown in Figure 2. The original image is shown in Figure 2A. CCMetrics (Figure 2B) and NeuronJ (Figure 2C) were affected by manual actions. NFA FIJI (Figure 2D) was based on completely different algorithms and showed a skeletonized image. ACCMetrics analysis (Figure 2E) is calculated completely automatically. Figure 2 Examples of nerve fiber length calculations. A: the original 400x400 µm image of a patient with sarcoidosis and SFN symptoms. B: manual analysis performed with CCMetrics (CNFL = 25.8 mm/mm2). C: semiautomatic analysis with NeuronJ (CNFL = 31.74 mm/mm2). D: automatic analysis NFA FIJI (NFA = 19,650 µm2/mm2). E: automatic analysis performed with ACCMetrics (CNFL = 20.46 mm/mm2). First, Kruskal Wallis tests were applied to analyze differences in CNFL between four groups for the different software methods. Pearson’s chi-square logistic regression was then used to calculate correlation coefficients and 95% confidence interval for intra-class correlation coefficient. BlandAltman plots were used to determine the agreement between the same parameter calculated with different software techniques. Finally, the quadratic relationship between CNFL and CNFA was reproduced based on linear regression and the passing Bablock technique as described by Brines et al.21 6 104 6
RkJQdWJsaXNoZXIy MjY0ODMw