Difference between revisions of "NA"

From NAMIC Wiki
Jump to: navigation, search
m (Update from Wiki)
m (Update from Wiki)
Line 1: Line 1:
== Base Classes ==
+
== Testing Methods for Statistical Shape Analysis ==
  
'''SphericalWaveletObject'''<dimension of signal ,type of signal>::itkObject
+
=== Description ===
  
: Functions:
+
3 Main correction methods in local statistical shape analysis for the multiple comparison problem exists.
:: -get/setSubdivisionLevel() : sets the level of recursive, spherical icosahedron subdivision
 
:: -getTemplateMesh() [sphere] : returns the subdivided icosahdron mesh with lower levels preceding higher levels in the point list.
 
:: -get/setSignalVector (does not compute, only storage)
 
:: -get/setCoefficientVector (does not compute, only storage)
 
:: -IsSignalVectorAvailable() whether the signal information is available
 
:: -IsCoefficientVector() whether the coefficients are available
 
  
: :: -getIndicesSetAtResolution() //this the K set, used for the forward/backward filters
+
# Bonferroni correction (simple, worst case, conservative)
:: -getDifferenceIndicesSetAtResolution() //this is the M set, used for the forward/backward filters
+
# Minimum statistic using Non-parametric permutation tests (less conservative than Bonferroni, full control of family wise error rate (FWER, Type I error, false positives), still pessimistic)
: Member Variables:
+
# False Discovery Rate (FDR, control of false discovery rate, i.e. control of false positive in only those features that reject Null Hypothesis, less conservative)
:: Private:
 
::: coeffs:itkVector<itkVector<double>> (k-th vector stores the coefficients for dimension k of signal) [Nxd], where N is the number of points in the template mesh, and d is the dimenstionality of the signal
 
::: signal:itkVector<itkVector<type of signal>> (k-th vector stores the signal at dimension k) [Nxd], where N is the number of points in the template mesh, and d is the dimenstionality of the signal
 
::: A,B,C,K,M …(what type of structure?)
 
::: itkMesh: TemplateMesh
 
  
<br />'''SurfaceWaveletsObject'''<dim,type>::'''SphericalWaveletsObject'''<dim,type>
+
Currently the UNC shape analysis pipeline incorporates the permutation test based correction technique. This project will
  
: Functions:
+
* extend the shape analysis pipeline to incorporate Bonferroni and FDR
:: -set/getSphericalParametrizedMesh() [internally we remesh]
+
* change the current code to use the ITK statistics framework
:: -get/setSurfaceSignal
 
  
== Filters ==
+
=== Current Status ===
  
'''SphericalWaveletsSignalToCoefficientsFilter'''<nowiki>::processingObject </nowiki>
+
* permutation tests are working
 +
* both bonferroni and FDR will be rather simple to implement, the bulk of work will be associated with transfering the code into the ITK framework
 +
* [[Image:2006_06_PW_StatAnal.ppt|Image:2006 06 PW StatAnal.ppt]]
  
: Functions:
+
=== Members ===
:: -setInput(sphericalWaveletObject)
 
:: -getOutput(sphericalWaveletObject)
 
  
<br />'''SphericalWaveletsCoefficentsToSignalFilter'''<nowiki>::processingObject </nowiki>
+
* Martin Styner (UNC)
 
 
: Functions:
 
:: -setInput(sphericalWaveletObject)
 
:: -getOutput(sphericalWaveletObject)
 
 
 
<br />
 
 
 
== Example of use for shape analysis ==
 
 
 
<br />
 
 
 
: SphericalWaveletObject swo<1,float>
 
: swo.setSubdivision(5)
 
: SphericalWaveletObject::surfaceSignalType s
 
: swo.setSignal(s);
 
 
 
: mesh = swo.getTemplateMesh(); //can visualize
 
 
 
: SphericalWaveletSignalToCoefficientsFilter swf
 
: swf.SetInput(swo);
 
: swf.Update();
 
: swg.GetOutput(swo);
 
 
 
== Links ==
 
 
 
* [[NA-MIC/Projects/Structural/Shape_Analysis/Spherical_Wavelets_in_ITK|Spherical Wavelets in ITK]]
 
* [[NA-MIC/Projects/Structural/Shape_Analysis/3D_Shape_Analysis_Using_Spherical_Wavelets|3D Shape Analysis Using Spherical Wavelets]]
 

Revision as of 14:03, 18 December 2006

Home < NA

Testing Methods for Statistical Shape Analysis

Description

3 Main correction methods in local statistical shape analysis for the multiple comparison problem exists.

  1. Bonferroni correction (simple, worst case, conservative)
  2. Minimum statistic using Non-parametric permutation tests (less conservative than Bonferroni, full control of family wise error rate (FWER, Type I error, false positives), still pessimistic)
  3. False Discovery Rate (FDR, control of false discovery rate, i.e. control of false positive in only those features that reject Null Hypothesis, less conservative)

Currently the UNC shape analysis pipeline incorporates the permutation test based correction technique. This project will

  • extend the shape analysis pipeline to incorporate Bonferroni and FDR
  • change the current code to use the ITK statistics framework

Current Status

  • permutation tests are working
  • both bonferroni and FDR will be rather simple to implement, the bulk of work will be associated with transfering the code into the ITK framework
  • File:2006 06 PW StatAnal.ppt

Members

  • Martin Styner (UNC)