Difference between revisions of "2011 Summer Project Week Image Manifold Learning with Spectral Embedding and Laplacian Eigenmaps"

From NAMIC Wiki
Jump to: navigation, search
Line 2: Line 2:
 
<gallery>
 
<gallery>
 
Image:PW-MIT2011.png|[[2011_Summer_Project_Week#Projects|Projects List]]
 
Image:PW-MIT2011.png|[[2011_Summer_Project_Week#Projects|Projects List]]
 +
Image:adni_embedding.png|Learned embedding from part of ADNI data
 
</gallery>
 
</gallery>
 
+
[[File:[[File:Example.jpg]][[File:[[File:Example.jpg]][[File:[[File:Example.jpg]]]]]]]]
 
'''Image Manifold Learning with Spectral Embedding and Laplacian Eigenmaps'''
 
'''Image Manifold Learning with Spectral Embedding and Laplacian Eigenmaps'''
  
Line 31: Line 32:
 
<h3>Progress</h3>
 
<h3>Progress</h3>
  
 +
#:Applied baseline method to part of the ADNI dataset
 +
#:Obtained HD data for future analysis
 
<!-- Fill this out before Friday's summary presentations - list what you did and how well it worked. -->
 
<!-- Fill this out before Friday's summary presentations - list what you did and how well it worked. -->
  

Revision as of 13:50, 24 June 2011

Home < 2011 Summer Project Week Image Manifold Learning with Spectral Embedding and Laplacian Eigenmaps

[[File:Example.jpg[[File:Example.jpg[[File:Example.jpg]]]]]] Image Manifold Learning with Spectral Embedding and Laplacian Eigenmaps

Key Investigators

  • MIT: Ramesh Sridharan
  • MIT: Polina Golland


Objective

  1. Use image manifold learning to better understand pathology (e.g. Alzheimer's, Huntington's) in brain images.

Approach, Plan

  1. We want to learn better embeddings of brain images (to better perform classification, segmentation/registration, etc). We will use a modification of spectral embedding techniques that allows us to incorporate constraints. For example, when longitudinal data involving progression of some pathology is available, we would like to incorporate our knowledge about the temporal relationship by constraining the longitudinal images to line up.
    This is a relatively new project, and we are interested in hearing about potential applications and useful directions for the work.

Progress

  1. Applied baseline method to part of the ADNI dataset
    Obtained HD data for future analysis


References

Belkin, M; Niyogi, P. Laplacian Eigenmaps for Dimensionality Reduction and Data Representation

Delivery Mechanism