November 16, 2006

From Utah Center for Neuroimage Analysis

Present: Tolga, Paul, Samuel, James, Bryan, Chi-Bin, Melissa

Discussed how visualization tools might help in microscopy data analysis.

Discussed what might be causing the jitter in slice-to-slice registration and how to fix it. Possible causes are:

1. the truncation of the polynomial degree while cascading multiple polynomial warps,

2. the blob features might not be capturing a fine enough level of detail, but only a global warp

Paul will generate volumes of slice pairs (so no cascading). The theory is that if these look much better we can conclude that the problem lies in the cascasding.