Skip to main content
. Author manuscript; available in PMC: 2011 Jun 7.
Published in final edited form as: Proc SPIE Int Soc Opt Eng. 2011 Jan 1;7962:796206. doi: 10.1117/12.877884

Figure 2.

Figure 2

Estimated truth denoted by stars using EM ML estimation as in classic STAPLE (left) comparing to estimated truth using EM MAP estimation with data adaptive prior (right). In ML approach, the estimated distribution pattern seems to be correct, but is shifted to the upper right. In MAP approach, the bias gets properly estimated and everything is converging to a reasonable result.