Diferencia entre revisiones de «GIC-experimental-databases/Whole OASIS deformation feature vectors»

De Grupo de Inteligencia Computacional (GIC)
Sin resumen de edición
Sin resumen de edición
 
(No se muestran 6 ediciones intermedias del mismo usuario)
Línea 2: Línea 2:
Data
Data
</h4>
</h4>
[http://data.sands-social-network-mockup.com/oasis_data FTP]


Username: ftp@sands-social-network-mockup.com
:[http://data.sands-social-network-mockup.com/oasis_data FTP directory with OASIS data]


[http://data.sands-social-network-mockup.com/oasis_data/jacs.hdft5 OASIS Jacobian Determinants]
 
[http://data.sands-social-network-mockup.com/oasis_data/modulatedgm.hdft5 OASIS Modulated Grey Matter]
Direct links to the datasets
[http://data.sands-social-network-mockup.com/oasis_data/geodan.hdft5 OASIS Geodesic Anisotropy]
 
:[http://data.sands-social-network-mockup.com/oasis_data/jacs.hdft5 OASIS Jacobian Determinants]
:[http://data.sands-social-network-mockup.com/oasis_data/modulatedgm.hdft5 OASIS Modulated Grey Matter]
:[http://data.sands-social-network-mockup.com/oasis_data/geodan.hdft5 OASIS Geodesic Anisotropy]




Línea 24: Línea 26:


<h4>
<h4>
Code snippet for reading HDF5 file from Matlab
Code snippet for reading the HDF5 files from Matlab
</h4>
</h4>
<font face="monospace">
<font face="monospace">
Línea 36: Línea 38:
  end
  end
  clear i finfo var varname dsname
  clear i finfo var varname dsname
<font face="monospace">
</font face="monospace">





Revisión actual - 08:58 4 oct 2013

Data

FTP directory with OASIS data


Direct links to the datasets

OASIS Jacobian Determinants
OASIS Modulated Grey Matter
OASIS Geodesic Anisotropy


Reference

- Alexandre Savio, Manuel Graña; Deformation based feature selection for Computer Aided Diagnosis of Alzheimer’s Disease, Expert Systems with Applications, Available online 20 September 2012, Volume 40, Issue 5, April 2013, Pages 1619–1628 [1] draft JCR (2011) 2.203 (Q1 in CSAI)


Software


Code snippet for reading the HDF5 files from Matlab

filename = 'modulatedgm.hdf5'
finfo = h5info(filename);
for i = 1:size(finfo.Datasets,1)
   varname = finfo.Datasets(i).Name
   var     = genvarname(varname);
   dsname  = ['/' varname];
   eval( [var '= h5read(filename, dsname);' ]);
end
clear i finfo var varname dsname


Note

In order to reduce the size of the database, the feature vector elements have been changed from float32 to float16. We haven't tested this data. If you have any problem, please contact us.