Spinning Disk Exposure Times

[Claire Brown]

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Just a note to those of you using spinning disk confocal microscopes.

I have recently been testing out many different spinning disk confocal
microscopes and one of my worries was artifacts in the images due to

Variable sampling in different pixels due to the spinning nature of the
disk. Most of the commercial companies talk about the importance of making
sure

Your camera and your disk are synchronized. However, what I have found using
a uniform sample (Chroma Technology green fluorescent plastic slide) is that
it is also important to synchronize your exposure time. We found with a disk
spinning at 2500 rpms and an exposure time of 10 ms on an EM-CCD camera we
see lines in the images due to uneven sampling of pixels within the image.
However, if we went with 8, 12 or 16 ms (any multiple of 4 will be on the
same frequency as 2500 rpms or 400 us/spin) these line artifacts disappeared
because our exposure time was in sync with the disk and the camera.

 

These artifacts are not apparent by eye when a heterogeneous cellular sample
is in place because they are very subtle, but they will certainly be
important for quantitative imaging. So it is very important to use the
appropriate exposure times.

 

Sincerely,

 

Claire

 

 

_________________________________________________________________

Claire M. Brown, PhD

Life Sciences Complex Imaging Facility Director

McGill University Department of Biochemistry

http://www.lifesciencescomplex.mcgill.ca/imaging

 

[Farid Jalali]

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Thank you Claire for this information. I have reported to Olympus a few times, with images, in the past year and a half or so regarding disk artifacts. I am happily using their Disk Scanning Unit (DSU) on an IX81 frame. Although I rarely get use the spinning disk at such low exposures, even with an EM-CCD, I can actually see the horizontal and vertical slit appertures that make up the Olympus DSU spinning disk. Its made more much more apparent during acquisition by exaggerating the image contrast. After acquisition and 3D deconvolution, the artifact becomes more robust and very obvious when setting thresholds for image segmentation; the disk lines can actually be problematic when trying to segment nuclei for example.  Our disk operates at about 3000rpm, and most of the time its not a problem. Shifting the speed to 5000 rpm only made things worse. I do not think I have been given notice of a fix as yet.

Cheers
Farid

On Tue, Apr 29, 2008 at 4:24 PM, Claire Brown <[hidden email]> wrote:

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Just a note to those of you using spinning disk confocal microscopes.

I have recently been testing out many different spinning disk confocal
microscopes and one of my worries was artifacts in the images due to

Variable sampling in different pixels due to the spinning nature of the
disk. Most of the commercial companies talk about the importance of making
sure

Your camera and your disk are synchronized. However, what I have found using
a uniform sample (Chroma Technology green fluorescent plastic slide) is that
it is also important to synchronize your exposure time. We found with a disk
spinning at 2500 rpms and an exposure time of 10 ms on an EM-CCD camera we
see lines in the images due to uneven sampling of pixels within the image.
However, if we went with 8, 12 or 16 ms (any multiple of 4 will be on the
same frequency as 2500 rpms or 400 us/spin) these line artifacts disappeared
because our exposure time was in sync with the disk and the camera.

 

These artifacts are not apparent by eye when a heterogeneous cellular sample
is in place because they are very subtle, but they will certainly be
important for quantitative imaging. So it is very important to use the
appropriate exposure times.

 

Sincerely,

 

Claire

 

 

_________________________________________________________________

Claire M. Brown, PhD

Life Sciences Complex Imaging Facility Director

McGill University Department of Biochemistry

http://www.lifesciencescomplex.mcgill.ca/imaging

 

--
Farid Jalali MSc
Senior Research Technician/ Lab Manager
Dr. Robert Bristow Lab
Applied Molecular Oncology
Princess Margaret Hospital
Toronto, Canada
416-946-4501 X4351 (Princess Margaret Hospital)
416-581-7754 STTARR at MaRS Building
416-581-7791 STTARR Microscopy Suite

[Steve Baxter-2]

Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

[Commercial Response]

Hi,

With the CSU / DSU approaches, it is vital to synchronise the exposure  
time of the camera to a whole number of disk segments. If this is not  
done then you will see curved scanline artefacts on the image  
(particularly at short exposure times). The UltraVIEW does this by  
synchronising the CSU disk speed, the camera exposure and the laser  
exposure together. This has to be done accurately, any jitter in the  
camera exposure time or mismatch in the disk speed will result in  
scanlines appearing again.

If anyone is interested in further technical detail of this, there is  
some in this PerkinElmer patent:

http://v3.espacenet.com/textdoc?DB=EPODOC&IDX=DE60312717T

Cheers,

Steve

On 29 Apr 2008, at 23:09, Farid Jalali wrote:

> Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal 
>  Thank you Claire for this information. I have reported to Olympus a  
> few times, with images, in the past year and a half or so regarding  
> disk artifacts. I am happily using their Disk Scanning Unit (DSU) on  
> an IX81 frame. Although I rarely get use the spinning disk at such  
> low exposures, even with an EM-CCD, I can actually see the  
> horizontal and vertical slit appertures that make up the Olympus DSU  
> spinning disk. Its made more much more apparent during acquisition  
> by exaggerating the image contrast. After acquisition and 3D  
> deconvolution, the artifact becomes more robust and very obvious  
> when setting thresholds for image segmentation; the disk lines can  
> actually be problematic when trying to segment nuclei for example.  
> Our disk operates at about 3000rpm, and most of the time its not a  
> problem. Shifting the speed to 5000 rpm only made things worse. I do  
> not think I have been given notice of a fix as yet.
>
> Cheers
> Farid
>
> On Tue, Apr 29, 2008 at 4:24 PM, Claire Brown  
> <[hidden email]> wrote:
> Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> Just a note to those of you using spinning disk confocal microscopes.
>
> I have recently been testing out many different spinning disk confocal
> microscopes and one of my worries was artifacts in the images due to
>
> Variable sampling in different pixels due to the spinning nature of  
> the
> disk. Most of the commercial companies talk about the importance of  
> making
> sure
>
> Your camera and your disk are synchronized. However, what I have  
> found using
> a uniform sample (Chroma Technology green fluorescent plastic slide)  
> is that
> it is also important to synchronize your exposure time. We found  
> with a disk
> spinning at 2500 rpms and an exposure time of 10 ms on an EM-CCD  
> camera we
> see lines in the images due to uneven sampling of pixels within the  
> image.
> However, if we went with 8, 12 or 16 ms (any multiple of 4 will be  
> on the
> same frequency as 2500 rpms or 400 us/spin) these line artifacts  
> disappeared
> because our exposure time was in sync with the disk and the camera.
>
>
> These artifacts are not apparent by eye when a heterogeneous  
> cellular sample
> is in place because they are very subtle, but they will certainly be
> important for quantitative imaging. So it is very important to use the
> appropriate exposure times.
>
>
> Sincerely,
>
>
> Claire
>
>
>
> _________________________________________________________________
> Claire M. Brown, PhD
> Life Sciences Complex Imaging Facility Director
> McGill University Department of Biochemistry
> http://www.lifesciencescomplex.mcgill.ca/imaging
>
>
>
>
> --
> Farid Jalali MSc
> Senior Research Technician/ Lab Manager
> Dr. Robert Bristow Lab
> Applied Molecular Oncology
> Princess Margaret Hospital
> Toronto, Canada
> 416-946-4501 X4351 (Princess Margaret Hospital)
> 416-581-7754 STTARR at MaRS Building
> 416-581-7791 STTARR Microscopy Suite

Steve Baxter
R&D Leader
Improvision, a PerkinElmer Company
[hidden email]
[hidden email]
[phone]

Image Processing and Vision Company Limited (trading as Improvision)  
is a company registered in England and Wales with company number  
2505778.  VAT number GB536188722.