[Insight-users] ITK for fluorescence microscopy - BioImageXD

Mon May 11 20:15:06 EDT 2009

hi guys,

Here is our implementation of Fast ICA in ITK.
The corresponding paper and matlab code can be found here (http://www.cis.hut.fi/projects/ica/fastica/ 
).

Our implementation was first motivated by the need of a fast umixing  
code that would be open and Free (this code is release under BSD  
license).
We must also admit that it is the fruit of a "dare thing" between two  
of the main developers here after an infamous System Biology dep. beer  
hour. I will try to tell the tale from memory for posterity:

a. gelas: I can implement any algorithm in less than a day [in ITK].
k. mosaliganti: I can be faster than anyone in matlab.

The result was kind of a draw, but now we have a nice algorithm. :-)

Let me know if you have any suggestion or remark on the code.

alex.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: FastICA.zip
Type: application/zip
Size: 15685 bytes
Desc: not available
URL: <http://www.itk.org/pipermail/insight-users/attachments/20090511/085d6d1d/attachment-0001.zip>
-------------- next part --------------

On May 11, 2009, at 3:16 PM, Gouaillard, Alexandre wrote:

> hi badri,
>
> I'm so sorry, this is the unmixing code I was speaking long time ago  
> and that I should have sent you already.
>
> We made some experiments and although it is working very well on  
> most, it is underperforming in the specific application our lab is  
> interested in.
>
> We were hoping to find time to optimize it before releasing it, but  
> we have since lost that hope. I will send you a tarball containing  
> the mathlab code and the corresponding ITK implementation for you to  
> try *right away*. We would be interested in feedback on how it  
> compares to the unmixing code you are using today. I guess that if  
> we could match the result, it would be of interest for  you to use  
> this code, and to remove one external library dependency, right?
>
> We plan to contribute this to ITK, but we won't be on time for the  
> 3.14 release (in a few weeks). We preferred to focus on transferring  
> the segmentation and tracking algorithms first.
>
> regards,
>
> alex.
>
>
> -----Original Message-----
> From: Badri Roysam [mailto:roysam at ecse.rpi.edu]
> Sent: Monday, May 11, 2009 2:31 PM
> To: Gouaillard, Alexandre
> Cc: naraya3 at rpi.edu
> Subject: Re: [Insight-users] ITK for fluorescence microscopy -  
> BioImageXD
>
> HI Alex, I am interested to learn more about your unmixing method.  
> We'd love
> to try it out for our immunology live-cell imaging data.
>
>
> Badri Roysam
> Professor, Department of Electrical, Computer and Systems Engineering
> Associate Director, NSF Center for Subsurface Sensing & Imaging  
> Systems (CenSSIS ERC)
> Co-Director, Rensselaer Center for Open Source Software
> Rensselaer Polytechnic Institute
> 110 8th Street, Troy, New York 12180-3590, USA.
> Office(JEC 7010): 518-276-8067, Assistant: 518-276-8525, Lab(JEC  
> 6308): 518-276-8207, Fax: 518-276-8715
> Email: roysam at ecse.rpi.edu, Web: http://www.ecse.rpi.edu/~roysam
>
>
>
> ----- Original Message -----
> From: alex gouaillard [mailto:Alexandre_Gouaillard at hms.harvard.edu]
> To: Badri Roysam [mailto:roysam at ecse.rpi.edu]
> Cc: Daniel James White [mailto:white at mpi-cbg.de], Prashanth  
> Ravindran [mailto:prashanth at 100ximaging.com], ITK [mailto:insight-users at itk.org 
> ]
> Subject: Re: [Insight-users] ITK for fluorescence microscopy -  
> BioImageXD
>
>
>> hi badri,
>>
>> that's great news. I was sure at one point you would, but I did not
>> know when, hence the question mark in my e-mail.
>>
>> Farsight seems to have a lot of activity recently and that's great
>> news for everybody working on cell segmentation. Wiki begins to look
>> good.
>>
>> alex.
>>
>>
>>
>> On May 11, 2009, at 10:14 AM, Badri Roysam wrote:
>>
>>> Dear All, The farsight toolkit SVN will become open this month. We
>>> are working on finalizing the
>>> license model and taking care of various little details.
>>>
>>>
>>> Badri Roysam
>>> Professor, Department of Electrical, Computer and Systems  
>>> Engineering
>>> Associate Director, NSF Center for Subsurface Sensing & Imaging
>>> Systems (CenSSIS ERC)
>>> Co-Director, Rensselaer Center for Open Source Software
>>> Rensselaer Polytechnic Institute
>>> 110 8th Street, Troy, New York 12180-3590, USA.
>>> Office(JEC 7010): 518-276-8067, Assistant: 518-276-8525, Lab(JEC
>>> 6308): 518-276-8207, Fax: 518-276-8715
>>> Email: roysam at ecse.rpi.edu, Web: http://www.ecse.rpi.edu/~roysam
>>>
>>>
>>>
>>> ----- Original Message -----
>>> From: alex gouaillard [mailto:alexandre_gouaillard at hms.harvard.edu]
>>> To: Daniel James White [mailto:white at mpi-cbg.de], Prashanth
>>> Ravindran [mailto:prashanth at 100ximaging.com]
>>> Cc: ITK [mailto:insight-users at itk.org]
>>> Subject: Re: [Insight-users] ITK for fluorescence microscopy -
>>> BioImageXD
>>>
>>>
>>>> My bad,
>>>>
>>>> BioImageXD is also a wonderful application that has a lot of nice
>>>> features. They also developed an LSM reader which is the most
>>>> complete
>>>> LSM reader freely available in the ITK / VTK community. They also
>>>> went
>>>> the extra mile to make this LSM reader available under BSD, and  
>>>> we're
>>>> using it right now. By the way Dan, we are going to release a  
>>>> channel
>>>> unmixing (FastICA) code in ITK soon. You might be interested in
>>>> integrating it.
>>>>
>>>> Prashanth, That answer your first question: YES ITK is used and
>>>> helpful for microscopy image analysis. I guess you have four  
>>>> projects
>>>> here with four different approaches but all very complementary in
>>>> spirit
>>>>
>>>> FarSight:  multiple languages (C, C++, IDL, ...)  and libraries  
>>>> (ITK,
>>>> VTK, ...) wrapped in python. Interface to bio formats and Omero.
>>>> Framework, not application. No public access (yet?) to svn. No
>>>> information on license.
>>>>
>>>> BioImageXD: mainly C/C++ and python. Use ITK / VTK / wxWindow /
>>>> wxPython. Intenral support for specific formats (zeiss,  
>>>> leica, ...).
>>>> Application, but i suppose python module could be used  
>>>> independently?
>>>> License GPL which prevents me from reading the code.
>>>>
>>>> gaethan:  core in C / C++, wrapping in python or Java. Use bio-
>>>> formats
>>>> from python (and maybe java, with imageJ ?). No application that  
>>>> I'm
>>>> aware of. Most is BSD and transfered to ITK.
>>>>
>>>> gofigure: core in C++. Uses ITK / VTK / Qt. Application with plugin
>>>> mechanism. Support Database Back-end. Should support bio-formats  
>>>> and
>>>> Omero in a close future. BSD, and most features are transfered to
>>>> ITK / VTK.
>>>>
>>>> My understanding is that you start from the hardware part, and  
>>>> you're
>>>> already using a lot of java, and ImageJ. I have no experience with
>>>> Java here, but I guess you could ask some question to Gaethan about
>>>> it.
>>>>
>>>> alex.
>>>>
>>>>
>>>>
>>>>
>>>> On May 11, 2009, at 3:52 AM, Daniel James White wrote:
>>>>
>>>>> Hi  Prashanth ,
>>>>>
>>>>> see
>>>>> www.bioimagexd.net
>>>>>
>>>>> it works very well for us.
>>>>>
>>>>> If you want to start a project from scratch, then good luck!
>>>>> If you want to help us extend the functionality of BXD in
>>>>> collaboration with us,
>>>>> then you would be very welcome to join the team!
>>>>>
>>>>> Its very easy to add ITK functions to ITK,
>>>>> as you can see how we did ti already in the source code,
>>>>> and its only a few lines to add for each new class (depending on  
>>>>> how
>>>>> complex the parameters for it etc are)
>>>>>
>>>>> We are also doing work here at MPI-CBG on segmentation of bright
>>>>> field, phase, DIC etc images.
>>>>>
>>>>> cheers
>>>>>
>>>>> Dan and the BXD team
>>>>>
>>>>> On May 9, 2009, at 6:00 PM, insight-users-request at itk.org wrote:
>>>>>
>>>>>>
>>>>>>
>>>>>> Dear ITK Users-
>>>>>>
>>>>>> I am considering using ITK as the basis for a set of tools for
>>>>>> analysis of fluorescence and bright-field images. I am curious to
>>>>>> know if others have tried/ are trying it, and if so what their
>>>>>> feedback is?
>>>>>>
>>>>>> Thanks,
>>>>>> --Prashanth
>>>>>
>>>>> Dr. Daniel James White BSc. (Hons.) PhD
>>>>> Senior Microscopist / Image Visualisation, Processing and Analysis
>>>>> Light Microscopy and Image Processing Facilities
>>>>> Max Planck Institute of Molecular Cell Biology and Genetics
>>>>> Pfotenhauerstrasse 108
>>>>> 01307 DRESDEN
>>>>> Germany
>>>>>
>>>>>
>>>>> New Mobile Number!!!
>>>>>
>>>>> +49 (0)15114966933 (German Mobile)
>>>>> +49  (0)351 210 2627 (Work phone at MPI-CBG)
>>>>> +49  (0)351 210 1078 (Fax MPI-CBG LMF)
>>>>>
>>>>> http://www.bioimagexd.net
>>>>> http://www.chalkie.org.uk
>>>>> dan at chalkie.org.uk
>>>>> ( white at mpi-cbg.de )
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> _____________________________________
>>>>> Powered by www.kitware.com
>>>>>
>>>>> Visit other Kitware open-source projects at
>>>>> http://www.kitware.com/opensource/opensource.html
>>>>>
>>>>> Please keep messages on-topic and check the ITK FAQ at:
>>>> http://www.itk.org/Wiki/ITK_FAQ
>>>>>
>>>>> Follow this link to subscribe/unsubscribe:
>>>>> http://www.itk.org/mailman/listinfo/insight-users
>>>>
>>>> _____________________________________
>>>> Powered by www.kitware.com
>>>>
>>>> Visit other Kitware open-source projects at
>>>> http://www.kitware.com/opensource/opensource.html
>>>>
>>>> Please keep messages on-topic and check the ITK FAQ at:
>>>> http://www.itk.org/Wiki/ITK_FAQ
>>>>
>>>> Follow this link to subscribe/unsubscribe:
>>>> http://www.itk.org/mailman/listinfo/insight-users
>>>>
>>
>>
> _____________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ITK FAQ at: http://www.itk.org/Wiki/ITK_FAQ
>
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-users