Detailed statistical comparison of CD41 cluster distributions within two clots. First level analysis is displayed in the images (a-c).

(a) shows the detailed analysis of the full samples (both samples reduced to 60 000 points) density (top) and curvature (bottom). In this case, density and curvature distributions of clusters for each cluster dimensions were compared at one time using the KS-/WX-test (blue/red). The results indicate a similarity of samples for all cluster dimensions. As an indicator, an aggregated p-value is determined using the weighting t-Norm-AND. The p-value (black dashed line) remains mostly above the critical p-value area (orange bar) proving the similarity hypothesis. The corresponding similarities are sim_M = 0.57/0.91 for the density and curvature distributions respectively. Remarkably, the density distribution shows a lower similarity compared to the curvature. A closer look at the comparison of the density distributions shows, that for small cluster dimensions between 10 nm and 150 nm as well as for large cluster dimensions ranging from approx. 1 400 nm– 1 800 nm, the distributions are within the critical p-value area. (b) depicts the aggregated p-value (density and curvature) calculated with the weighting t-Norm-AND (blue) and the confidence interval (black dashed line). A sim_M = 0.71 is determined. For such large samples, a faster bootstrap-resampling is used. (d) depicts 100 curves–representing relative density of 100-resamples each including 2 000 points (green)–used for the bootstrap based comparison of the cluster/curvature density distribution for both samples. In black, the averaged curve is represented. The detailed analysis (first level statistics) of the bootstrapped samples (d) was performed using similar statistics as used in (e). (c) shows the detailed (first level statistic) comparison of the distributions of cluster density (top) and curvature (bottom) for all given cluster dimensions (sizes). For this analysis, the KS/WX- tests (blue/red) as well as the t-Norm-AND values are plotted (black). Additionally, the boundaries of the confidence interval are displayed (magenta). The bootstrap derived data shows a strong similarity between the samples, sim_M = 0.91 and 0.95 for cluster density and curvature distributions was determined. (e-i) depict the second level analysis. (e) shows the p-values determined by the mean-cross analysis of the average density (left)/curvature (right) values determined via bootstrapping. Herein p-values of the mean-cross comparison (blue) and lower/upper confidential boundaries (black dashed line) are shown. In general, as expected a huge similarity between the samples is visible, sim_M = 0.92 for the mean cluster density distribution and 0.97 for the mean curvature distribution was determined. (h) shows the aggregated p-values for the cluster mean cross comparison of curvature and the mean cross of density distributions (blue). The aggregated p-values show a high similarities; sim_M = 0.91 and sim_L = 1. (f) depicts the calculated K/-H-Ripley’s-function values for the bootstrapped data (green) for all cluster dimensions. (g) depicts the comparison of the K-/H-Ripley function values (K-left, H-right). The p-values of the cross comparison (red), mean-cross comparison (blue) and lower/upper confidential bounds (black dashed line) are shown. This cross-comparison of the cluster-data also indicates a similarity of the samples for Ripley’s function analysis. The aggregated p-value of the results of the K-/H-Ripley functions comparison (i), indicates a slight dissimilarity of the results for cluster dimensions (sizes) between 200 nm– 600 nm.

(TIF)