Zeiss LSM file defeats LSM Reader & Toolbox

[Francis Burton]

Hi,

I'm trying to read Zeiss LSM files directly into ImageJ rather than having
to use the free Zeiss Image Browser software to export TIFFs first. These
files are produced from linescan experiments (laser whizzes back and forth
along a line, resulting in an image representing x vs time). There are two
12 bit channels, one for each wavelength used, and the resulting 'images'
are subsequently ratioed to produce a single linescan image.

Unfortunately, none of the available free software that I have tried so far
seems to be able to handle these files. LSM Reader reports "Unsupported LSM
scantype: 2", and if I rename the file from .LSM to .TIF, TiffDecoder tells
me that "ImageJ cannot open 48-bit LZW compressed TIFFs".

(I also tried the MATLAB function tiffread.m mentioned at
http://en.wikipedia.org/wiki/LSM_(Zeiss) but that didn't work either and
in any case I really want to use ImageJ not MATLAB!)

Would it be straightforward to modify LSM Reader to be able to handle these
files? Presumably a LZW decoder is now included. Is it simply a matter of
interpreting the resulting decompressed bytes as pairs of 12 bit chunks?

Cheers,
Francis

--
Dr Francis L Burton,          |  [hidden email]
West Medical Building,        |
University of Glasgow,        |  Tel [phone]
Glasgow, G12 8QQ, Scotland.   |  Fax [phone]

[Michael Weber-4]

Francis,

why do you use the Zeiss compression? This only makes things more difficult.

Michael


Francis Burton wrote:

> Hi,
>
> I'm trying to read Zeiss LSM files directly into ImageJ rather than having
> to use the free Zeiss Image Browser software to export TIFFs first. These
> files are produced from linescan experiments (laser whizzes back and forth
> along a line, resulting in an image representing x vs time). There are two
> 12 bit channels, one for each wavelength used, and the resulting 'images'
> are subsequently ratioed to produce a single linescan image.
>
> Unfortunately, none of the available free software that I have tried so far
> seems to be able to handle these files. LSM Reader reports "Unsupported LSM
> scantype: 2", and if I rename the file from .LSM to .TIF, TiffDecoder tells
> me that "ImageJ cannot open 48-bit LZW compressed TIFFs".
>
> (I also tried the MATLAB function tiffread.m mentioned at
> http://en.wikipedia.org/wiki/LSM_(Zeiss) but that didn't work either and
> in any case I really want to use ImageJ not MATLAB!)
>
> Would it be straightforward to modify LSM Reader to be able to handle these
> files? Presumably a LZW decoder is now included. Is it simply a matter of
> interpreting the resulting decompressed bytes as pairs of 12 bit chunks?
>
> Cheers,
> Francis

[Francis Burton]

Michael,

Whether I like it or not, I have to work with the files that my colleagues
produce. I could ask them to turn off compression (if there's a switch) but
they may not do that if the increase in file size makes the files unwieldy
e.g. they can't burn one experiment onto a DVD disc.

Francis

At 10:53 04/03/08 +0100, you wrote:

>Francis,
>
>why do you use the Zeiss compression? This only makes things more difficult.
>
>Michael
>
>
>Francis Burton wrote:
>> Hi,
>>
>> I'm trying to read Zeiss LSM files directly into ImageJ rather than having
>> to use the free Zeiss Image Browser software to export TIFFs first. These
>> files are produced from linescan experiments (laser whizzes back and forth
>> along a line, resulting in an image representing x vs time). There are two
>> 12 bit channels, one for each wavelength used, and the resulting 'images'
>> are subsequently ratioed to produce a single linescan image.
>>
>> Unfortunately, none of the available free software that I have tried so far
>> seems to be able to handle these files. LSM Reader reports "Unsupported LSM
>> scantype: 2", and if I rename the file from .LSM to .TIF, TiffDecoder tells
>> me that "ImageJ cannot open 48-bit LZW compressed TIFFs".
>>
>> (I also tried the MATLAB function tiffread.m mentioned at
>> http://en.wikipedia.org/wiki/LSM_(Zeiss) but that didn't work either and
>> in any case I really want to use ImageJ not MATLAB!)
>>
>> Would it be straightforward to modify LSM Reader to be able to handle these
>> files? Presumably a LZW decoder is now included. Is it simply a matter of
>> interpreting the resulting decompressed bytes as pairs of 12 bit chunks?
>>
>> Cheers,
>> Francis
>
>

[ctrueden]

Hi Francis,

Did you try Bio-Formats? http://www.loci.wisc.edu/ome/formats.html

-Curtis

On Tue, Mar 4, 2008 at 3:47 AM, Francis Burton <[hidden email]>
wrote:

> Hi,
>
> I'm trying to read Zeiss LSM files directly into ImageJ rather than having
> to use the free Zeiss Image Browser software to export TIFFs first. These
> files are produced from linescan experiments (laser whizzes back and forth
> along a line, resulting in an image representing x vs time). There are two
> 12 bit channels, one for each wavelength used, and the resulting 'images'
> are subsequently ratioed to produce a single linescan image.
>
> Unfortunately, none of the available free software that I have tried so
> far
> seems to be able to handle these files. LSM Reader reports "Unsupported
> LSM
> scantype: 2", and if I rename the file from .LSM to .TIF, TiffDecoder
> tells
> me that "ImageJ cannot open 48-bit LZW compressed TIFFs".
>
> (I also tried the MATLAB function tiffread.m mentioned at
> http://en.wikipedia.org/wiki/LSM_(Zeiss)<http://en.wikipedia.org/wiki/LSM_%28Zeiss%29>but that didn't work either and
> in any case I really want to use ImageJ not MATLAB!)
>
> Would it be straightforward to modify LSM Reader to be able to handle
> these
> files? Presumably a LZW decoder is now included. Is it simply a matter of
> interpreting the resulting decompressed bytes as pairs of 12 bit chunks?
>
> Cheers,
> Francis
>
> --
> Dr Francis L Burton,          |  [hidden email]
> West Medical Building,        |
> University of Glasgow,        |  Tel [phone]
> Glasgow, G12 8QQ, Scotland.   |  Fax [phone]
>

[Philip Ershler]

On Mar 4, 2008, at 2:47 AM, Francis Burton wrote:

> Hi,
>
> I'm trying to read Zeiss LSM files directly into ImageJ rather than  
> having
> to use the free Zeiss Image Browser software to export TIFFs first.  
> These
> files are produced from linescan experiments (laser whizzes back and  
> forth
> along a line, resulting in an image representing x vs time). There  
> are two
> 12 bit channels, one for each wavelength used, and the resulting  
> 'images'
> are subsequently ratioed to produce a single linescan image.
>
> Unfortunately, none of the available free software that I have tried  
> so far
> seems to be able to handle these files. LSM Reader reports  
> "Unsupported LSM
> scantype: 2", and if I rename the file from .LSM to .TIF,  
> TiffDecoder tells
> me that "ImageJ cannot open 48-bit LZW compressed TIFFs".
>
> (I also tried the MATLAB function tiffread.m mentioned at
> http://en.wikipedia.org/wiki/LSM_(Zeiss) but that didn't work either  
> and
> in any case I really want to use ImageJ not MATLAB!)
>
> Would it be straightforward to modify LSM Reader to be able to  
> handle these
> files? Presumably a LZW decoder is now included. Is it simply a  
> matter of
> interpreting the resulting decompressed bytes as pairs of 12 bit  
> chunks?
>

I ran into this same issue. The problem is that when the two gentlemen  
wrote the LSM ToolBox, they had no line scan files (scantype: 2) to test
their software. So rather than push out some software that might give  
unreliable results, they excluded line scan files. After I sent them  
several
examples linescan files, they modified the ToolBox so that it can now  
handle linescan files. I'm not sure if the changed Toolbox is  
available for
download from the ImageJ web site. Check there first and make sure  
that you've got the most recent version of ImageJ and the LSM Toolbox.

Actually, the LSM Reader has been folded into the ImageJ program. This  
means that one can open LSM files right from the File Menu.
If you can't get a version of something that will open your linescan  
files, get back to me and I will send you a copy of what I've got.

HTH,

Phil

[Patrick Pirrotte]

Hi,

I've got one bug I'd like to squash before releasing the LSMToolbox/LSMReader 4.0c.
I'll try to do that still this week and release the new version.

Regards,

Patrick

----- Original Message -----
From: Philip Ershler <[hidden email]>
Date: Tuesday, March 4, 2008 23:02
Subject: Re: Zeiss LSM file defeats LSM Reader & Toolbox
To: [hidden email]

> On Mar 4, 2008, at 2:47 AM, Francis Burton wrote:
>
> > Hi,
> >
> > I'm trying to read Zeiss LSM files directly into ImageJ rather
> than 
> > having
> > to use the free Zeiss Image Browser software to export TIFFs
> first. 
> > These
> > files are produced from linescan experiments (laser whizzes
> back and 
> > forth
> > along a line, resulting in an image representing x vs time).
> There 
> > are two
> > 12 bit channels, one for each wavelength used, and the
> resulting 
> > 'images'
> > are subsequently ratioed to produce a single linescan image.
> >
> > Unfortunately, none of the available free software that I have
> tried 
> > so far
> > seems to be able to handle these files. LSM Reader
> reports 
> > "Unsupported LSM
> > scantype: 2", and if I rename the file from .LSM to
> .TIF, 
> > TiffDecoder tells
> > me that "ImageJ cannot open 48-bit LZW compressed TIFFs".
> >
> > (I also tried the MATLAB function tiffread.m mentioned at
> > http://en.wikipedia.org/wiki/LSM_(Zeiss) but that didn't work
> either 
> > and
> > in any case I really want to use ImageJ not MATLAB!)
> >
> > Would it be straightforward to modify LSM Reader to be able
> to 
> > handle these
> > files? Presumably a LZW decoder is now included. Is it simply
> a 
> > matter of
> > interpreting the resulting decompressed bytes as pairs of 12
> bit 
> > chunks?
> >
>
> I ran into this same issue. The problem is that when the two
> gentlemen 
> wrote the LSM ToolBox, they had no line scan files (scantype: 2)
> to test
> their software. So rather than push out some software that might
> give 
> unreliable results, they excluded line scan files. After I sent
> them 
> several
> examples linescan files, they modified the ToolBox so that it
> can now 
> handle linescan files. I'm not sure if the changed Toolbox
> is 
> available for
> download from the ImageJ web site. Check there first and make
> sure 
> that you've got the most recent version of ImageJ and the LSM Toolbox.
>
> Actually, the LSM Reader has been folded into the ImageJ
> program. This 
> means that one can open LSM files right from the File Menu.
> If you can't get a version of something that will open your
> linescan 
> files, get back to me and I will send you a copy of what I've got.
>
> HTH,
>
> Phil
>