Computational and Mathematical Methods in Medicine / 2014 / Article / Tab 1 / Research Article
Crossing Fibers Detection with an Analytical High Order Tensor Decomposition Table 1 Success rate (%) in detecting the number of crossing fiber bundles, results of 100 simulations of noise for each crossing angle and for each SNR level.
Separation angle SNR Method 40 Adecomp-SHOT 100% 100% 100% 100% 100% 100% 100% 100% 100% 68% 3% CP-decomp. 100% 100% 100% 100% 100% 100% 100% 100% 100% 61% 1% Z-eigenvalues 100% 100% 99% 99% 90% 11% 0% 30 Adecomp-SHOT 100% 100% 100% 100% 100% 100% 100% 100% 95% 41% 4% CP-decomp 100% 100% 100% 100% 100% 100% 100% 100% 94% 39% 3% Z-eigenvalues 100% 100% 98% 100% 74% 11% 0% 20 Adecomp-SHOT 99% 99% 99% 99% 98% 99% 98% 99% 77% 33% 8% CP-decomp 100% 100% 100% 100% 100% 100% 100% 99% 74% 34% 4% Z-eigenvalues 99% 100% 100% 99% 72% 24% 0% 10 Adecomp-SHOT 96% 82% 88% 82% 80% 81% 72% 67% 54% 36% 27% CP-decomp 100% 100% 100% 100% 100% 100% 99% 92% 80% 67% 55% Z-eigenvalues 100% 99% 98% 95% 73% 47% 0%