Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

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

Figure 2

WM detection results on synthetic BrainWeb data. Figure 2: Pixel-level WM detection results visualized for one image from the MNI brain MRI dataset, each row corresponding to a different combination of noise and inhomogeneity: (a)-(e) 1% noise, 0% inhomogeneity, (f)-(j) 3% noise, 20% inhomogeneity, (k)-(o) 7% noise, 40% inhomogeneity. The first column shows the original PD MRI image with the ground truth for WM outlined in red, while the second, third, fourth, and fifth columns show the pixel-level WM classification results for Ψ(F), Ψ(X MDS ), Ψ(X GE ), and Ψ ( X ̃ G E ) , respectively. The red and green colors in (b)-(e), (g)-(j), (l)-(o) denote the GM and WM regions identified in each result image.

Back to article page