Skip to main content

Table 4 Comparing the mean and median as ML estimators within CalcConsEmbed

From: Consensus embedding: theory, algorithms and application to segmentation and classification of biomedical data

Noise Inhomogeneity ϕ A c c ( X ̃ G E M e d ) ϕ A c c ( X ̃ G E M e a n )
  0% 66.86 ± 2.89 66.89 ± 2.91
0% 20% 61.65 ± 4.58 65.34 ± 4.12
  40% 64.28 ± 5.93 63.39 ± 6.51
  0% 80.62 ± 1.03 80.45 ± 1.07
1% 20% 73.07 ± 8.97 77.81 ± 0.96
  40% 66.46 ± 9.80 70.56 ± 7.15
  0% 85.38 ± 0.75 85.53 ± 0.84
3% 20% 84.61 ± 0.81 84.49 ± 0.76
  40% 79.19 ± 7.56 81.37 ± 1.39
  0% 89.68 ± 1.36 90.85 ± 1.32
5% 20% 86.81 ± 1.38 87.01 ± 1.83
  40% 81.67 ± 1.51 81.82 ± 1.32
  0% 87.81 ± 0.73 86.17 ± 6.11
7% 20% 86.07 ± 1.05 82.73 ± 8.23
  40% 81.53 ± 1.57 81.72 ± 1.47
  0% 75.51 ± 14.35 74.32 ± 16.11
9% 20% 78.18 ± 9.86 73.63 ± 12.75
  40% 78.18 ± 9.86 73.63 ± 12.75
  1. Pixel-level WM detection accuracy and standard error averaged over 10 MNI brain images and for 18 combinations of noise and inhomogeneity (180 experiments) with each of the 2 ML estimators considered in CalcConsEmbed: (1) median ( Ψ ( X ̃ G E M e d ) ) , (2) ( Ψ ( X ̃ G E M e a n ) ) .