[ome-users] Flexibility in MATLAB Chains

[Tom Macura]

Hi Melissa,

Doing "binary calculations prior to the tracking algorithm" doesn't  
require anything special. Create a regular analysis chain with three  
modules (same as you itemized): 1. Original 5D; 2. Binary Image; 3.  
Tracking Module. In the Chain XML, make nodes for modules 1,2,3 and  
make the following node connections:
1->2
2->3
1->3

The OME Analysis Engine will realize that the Binary Image module has  
to be executed before the Tracking Module and act accordingly.

"We want do Binary calculations offline".
Please expand on what exactly you are planning. We have functionality  
in place for importing external numerical calculations into OME via  
spreadsheets. But what you seem to need is a way for importing into  
OME images that represent external pixel computations. We don't have  
such functionality ATM and I'm not planning on developing it myself.

> however, we also want the flexibility at stage 3 to
> query whether the binary image has been created, and if it hasnt, to
> create it.  Is there a way to do this?  I know that the pixel sets are
> related, but how do we specifically ask whether a certain set (binary)
> relating to a gloabl set (original), has been created?
It's possible to do this using the Perl API but not in Module/Chain  
XML. Our workflows don't support conditionals or iteration.

Tom

On Oct 23, 2006, at 5:12 PM, Tom Macura wrote:

> Melissa (debartolomeo at ece.ucsb.edu) wrote:
>
> I want to take a given image (in this case, a Z-stack of microtubule
> images) and run it through an analysis chain as below:
>
> 1. Original 5D ->
> 2. Binary Image (creates new PixelSet) ->
> 3. Microtubule-specific tracking algorithm (which requires BOTH the
> Original Image and the Binary Image) ->
> (4.) output PixelSets
>
> we want to do the binary calculations offline, or, at least prior  
> to the
> tracking alorithm.  however, we also want the flexibility at stage  
> 3 to
> query whether the binary image has been created, and if it hasnt, to
> create it.  Is there a way to do this?  I know that the pixel sets are
> related, but how do we specifically ask whether a certain set (binary)
> relating to a gloabl set (original), has been created?
>
> If you can help (I'm sure it's just that I'm not understanding  
> something
> as fully as I could) that would be great.  Have a good day!
> _______________________________________________
> ome-users mailing list
> ome-users at lists.openmicroscopy.org.uk
> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users