Skip to main content

Table 1 Performance on simulated images of cell clusters

From: CytoSpectre: a tool for spectral analysis of oriented structures on cellular and subcellular levels

  Error mean ± std Normalized error mean ± std Pearson’s r
Mean orientation 1.5534° ± 3.3967° - -
Circular variance 0.2384 ± 0.2162 0.2384 ± 0.2162 0.4156
Angular std 0.4042 ± 0.2632 0.2858 ± 0.1861 0.4230
Circular skewness −0.0001 ± 0.0312 −0.0002 ± 0.0578 0.4686
Circular kurtosis −0.3930 ± 0.2209 −0.3272 ± 0.1840 0.4980
Wavelength, mean 21.9955 μm ± 5.5363 μm 0.5281 ± 0.1329 0.9532
Wavelength, median 24.4053 μm ± 6.1943 μm 0.5703 ± 0.1447 0.9262
Wavelength, mode 31.7249 μm ± 10.1070 μm 0.7023 ± 0.2237 0.4770
Wavelength, std 2.1006 μm ± 3.6847 μm 0.1104 ± 0.1936 0.4722
  1. Errors (mean ± standard deviation) and Pearson’s linear correlation coefficient between true and estimated values for non-degraded simulated images of the cell cluster dataset (N = 1000) are shown for each parameter. Normalized errors were obtained by dividing the errors by the full range of true values. The normalized error and linear correlation coefficient are not applicable to the mean orientation. Mean orientation error is given as absolute error in degrees