[Insight-users] Origin of Analyze Format data

[Hans Johnson]

Tanya,

If I remember correctly from conversations 5 years ago, the originator field
was a place for the investigator name when they were collecting data from
the DSR at the mayo clinic.

The ITK implementation follows the definition provided by the original
Analyze developers at the Mayo Clinic.  There are also many other
corruptions of this 348 byte header file for different meanings
(SPM/FSL/Mayo/MRIcro/ and many other lesser known variations).  The lack of
common consensus of what these 348 bytes definitively mean was a major
driving force for implementing the new NIFTI file format.

If at all possible, I would avoid the Analyze format in favor of formats
that have less community divergence in interpretation.

Regards,
Hans




On 1/28/07 9:52 AM, "Luis Ibanez" <luis.ibanez at kitware.com> wrote:

> 
> Hi Tanya,
> 
> I'm not sure what the "originator" field is supposed
> to represent in the Analyze header.
> 
> 
> The specification of the Analyze header
> 
>       http://www.mayo.edu/bir/PDF/ANALYZE75.pdf
> 
> just list this field as a "10 chars".
> 
> 
> While the following table actually leads to think
> that the 10 chars have been used as 5 signed short
> (16 bits) integers:
> 
> http://www.grahamwideman.com/gw/brain/analyze/formatdoc.htm
> 
> 
> Hopefully a more knowledgeable person on the Analyze
> format could give us some hints regarding the intended
> use of this field.
> 
> 
> 
>      Regads,
> 
> 
>         Luis
> 
> 
> --------------
> Tanya wrote:
>> Thank you very much for your reply, Luis,
>>  
>> By the way, what is "originator" in data_history of Analyze format?
>>  
>> Thanks in advance,
>> Tanya
>> 
>> */Luis Ibanez <luis.ibanez at kitware.com>/* wrote:
>> 
>> 
>>     Hi Tanya,
>> 
>>     Since Analyze does not store the origin of the dataset,
>>     you may have to go back to your image provider.
>> 
>>     Chances are that there is an origin DICOM dataset
>>     sitting in some directory, or in a PACS system.
>> 
>>     When you read the analyze file with ITK the origin
>>     will simply be set to the default of (0,0,0).
>> 
>>     You will still be able to register this images.
>> 
>>     The only difference is that you have to be careful
>>     with the interpretation of the resulting transform.
>> 
>>     That is, the transform will be good enough for
>>     doing things like resampling the image, and
>>     performing visualization, but it will not be
>>     good enough for driving an image guided surgery
>>     system.
>> 
>> 
>> 
>>     Regards,
>> 
>> 
>>     Luis
>> 
>> 
>>     ------------
>>     Tanya wrote:
>>> Dear Insight-Users,
>>> 
>>> Where can we find the Origin of an Analyze Format data?
>>> 
>>> If Analyze does not save Origin of the data, then how can ITK
>>     register
>>> the image, since registration should be done in physical
>>     coordinate, and
>>> to convert from image coordinate to physical coordinate we should
>>     now
>>> the origin and the spacing?
>>> 
>>> Answers would be appreciated.
>>> 
>>> TIA,
>>> Tanya.
>>> 
>>> 
>> 
>> 
>> ------------------------------------------------------------------------
>> Don't pick lemons.
>> See all the new 2007 cars
>> <http://autos.yahoo.com/new_cars.html;_ylc=X3oDMTE0OGRsc3F2BF9TAzk3MTA3MDc2BH
>> NlYwNtYWlsdGFncwRzbGsDbmV3Y2Fycw-->
>> at Yahoo! Autos.
>> <http://autos.yahoo.com/new_cars.html;_ylc=X3oDMTE0OGRsc3F2BF9TAzk3MTA3MDc2BH
>> NlYwNtYWlsdGFncwRzbGsDbmV3Y2Fycw-->
>>