Detection%20of%20Anatomical%20Landmarks

1
Detection of Anatomical Landmarks
Georgetown University Medical Center Friday
October 6, 2006

• Bruno Jedynak
• Camille Izard

2
Anatomical Landmarks

• Manually defined points in the anatomy (
  geometric landmarks)
• !! Landmarker consistency, variability between
  exerts
• Used as is to analyze shapes
• Used as control point for image
  segmentation/registration

3
Landmarking the hippocampus from Brain MRI
4
Manual landmarking of the Hippocampus
5
Automatic landmarking

• Given a set of manually landmarked images
• Goal build a system that can landmark new images
• The system must adapt to different kind,
  different number of landmarks

6
Automatic landmarking Example

• Given 38 images expertly landmarked. K landmarks
  per image
• Goal landmark new images
• Mean error per new image
• Or expert evaluation

7
Stochastic modeling

• Build a likelihood function
• Learn
• For each new image, compute

8
Landmarks are points
Define
9
Template matching paradigm

• Identify landmarks with a deformation of the 3d
  space.
• Examples of deformations
• Affine
• Splines
• Diffeomorphisms

10
Spline model

• Define
• Identify
• Such that

11
Forward model

• Brain MRI gray-values are modeled as a mixture
  of Gaussians distributions.
• There are 6 components in the mixture CSF,GM,
  WM, CSF-GM, GM-WM, VeryWhite (Skull, blood
  vessels, )

12
Forward Model
13
Tissue Probability Map
csf csf-gm gm gm-wm wm outliers
HoH 0 0.04 0.90 0.06 0 0
14
Estimating the tissue probability map

• Learn the photometry of each image
• Register each image on the template
• Use the E.M. algo. for mixture of Gaussians to
  estimate

15
Automatic landmarking of a new image

• Learn the photometry parameters
• Use gradient ascent to maximize

16
Results
17
Results
18
Results
19
Current work

• Estimating the std. dev. of the Kernels
• Add control points to generate more complex
  deformations (K1)
• Test on schizophrenic and other brains