Skip to main content
. Author manuscript; available in PMC: 2015 Feb 11.
Published in final edited form as: Multimodal Brain Image Anal (2012). 2012;7509:115–133. doi: 10.1007/978-3-642-33530-3_10

Table 2.

Summary of the proposed multimodal label fusion framework

  1. Compute the most likely image intensity parameters:
    • 0. Initialize qx(n) = 1/N, Σl = limα →∞ αId (equivalent to majority voting).
    • 1. Update q with Equation 7 until convergence.
    • 2. Update means and variances with Equation 9.
    • 3. Update the bias field using the derivatives in Equation 10.
    • 4. Go to 1 until convergence.

  2. Compute the most likely segmentation for each voxel using Equation 11 and the latest estimate of q from step I.1.