table of contents
LIPSIA     Conversion of matlab to Vista and vice versa
mattov

The program 'mattov' converts all data in a matlab mat-File into the vista data format.

mattov -in test.mat -out test.v

In our example, the mat-File 'test.mat' contains the following data (in matlab):

>> whos
  Name      Size                    Bytes  Class

  a        30x40                     9600  double array
  b       100x100                   40000  single array
  c        20x1                        40  int16 array

Grand total is 11220 elements using 49640 bytes

The output vista file contains exactly the same matrices as images. Different matrices appear as different objects in the vista file which can be visualized using 'vxview'.

'mattov' can handle arrays of all data types.

vtomat

The program 'vtomat' converts all images of a vista file into a mat-File. All vista image data types are supported. 'Graph-files' are not supported.

vtomat -in file.v -out file.mat

Using 'vxview', the example 'file.v' looks as follows:

Under Matlab, the mat-file can be loaded. All variables are named 'objX' where 'X' is the object number in the vista file.

>> load file.mat
>> whos
  Name       Size                           Bytes  Class

  obj0     256x256x5                       327680  uint8 array
  obj1      64x64x5                         20480  uint8 array
  obj2      64x64x480                     3932160  int16 array
  obj3      64x64x480                     3932160  int16 array
  obj4      64x64x480                     3932160  int16 array
  
>> colormap gray;
>> imagesc(obj0(:,:,1)');
>>
Converting functional MRI data to mat-file

In Lipsia, the functional MRI data are stored in different images (vista objects). Each slice (row,column,time) is a single 3D-object which appears in the vista object list. Thus, the number of functional slices coincides with the number of 'VShort' images in the vista file.

Normally, 'vtomat' converts all functional images of a vista file as separate matrices in the mat-file. However, it is often useful to obtain a single 4-D image. This can be achieved using the option '-funconly true'. If this option is used, all functional data (images of 'VShort' type) are saved into a matlab matrix named 'fnc'. Non-functional images (anatomical, epiT1, tmaps, zmaps,...) are not converted.

vtomat -in file1.v -out file1.mat -funconly true

In this example, the vista file contains 5 functional slices (64x64) with 480 timesteps. The mat-file 'file1.mat' can be loaded and processed under Matlab as follows:

>> load file1.mat
>> whos
  Name      Size                    Bytes  Class
  fnc       4-D                  19660800  int16 array

Grand total is 9830400 elements using 19660800 bytes

>> size(fnc)
ans =
    64    64   480     5
>> colormap gray;
>> imagesc(fnc(:,:,1,5)');
>>   
Parameters of 'mattov':
-help
Prints usage information.
-in
Matlab input file. Default: (none)
-out
Output vista file. Default: (none)
Parameters of 'vtomat':
-help
Prints usage information.
-in
Input vista file. Default: (none)
-out
Matlab output file. Default: (none)
-funconly [ true | false ]
Functional data only (in one single 4D-object). Default: false


Max Planck Institute for Human Cognitive and Brain Sciences. Further Information: lipsia@cbs.mpg.de
Copyright © 2007 Max Planck Institute for Human Cognitive and Brain Sciences. All rights reserved.