Confocal Microscopy List

confocal laser scanning mode mean vs sum

Classic

List

Threaded

14 messages Options

SUBSCRIBE CONFOCALMICROSCOPY Sara Smith

confocal laser scanning mode mean vs sum

Dear confocal listers,

I've got a few questions about our Zeiss LSM510 system. And I am wondering if you could help me out with this.

Under the scan control panel in the software, there are two methods to do the scanning : mean and sum.

I knew averaging improves the image by increasing the signal to noise ratio.

Then what is the advantage of adding up the pixel value of all scan? Is this designed for weak signal image acquisition only?

By adding the value of each scan, you get amplification of your "real signal" not the noise.

Am I right about the "sum method"?

Another question is about the Line or Fram mode for averaging. People claim that fram averaging helps reduce photobleaching.

If I use one direction scan for a fixed sample and my system is a laser point scanning confocal microscope.

Would these two method apply the same photobleaching to my sample? In another word, would these two method have the same acquire time?

Any feedback would be appreciated.

Best,

Goodhouse, Joseph G.

Re: confocal laser scanning mode mean vs sum

Yi,

Yes summation is a way to collect weak signals but you will also be summing the noise. Not as much depending on how much of is random. You can filter it post collection to improve the S/N ratio. Another way to collect weak signals is to scan very slow to get good S/N ratio. This often works better. Uni-direction line scanning will not bleach more than frame mode. This you can easily test.

Joe Goodhouse
Confocal Core Lab Manager
Dept. of Molecular Biology
Princeton University

Washington Road

Princeton, NJ. 08544-1014
609-258-5432

Visit us at http://www.molbio1.princeton.edu/facility/confocal/

From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of yee sarah
Sent: Wednesday, March 03, 2010 11:09 AM
To: [hidden email]
Subject: confocal laser scanning mode mean vs sum

Dear confocal listers,

I've got a few questions about our Zeiss LSM510 system. And I am wondering if you could help me out with this.

Under the scan control panel in the software, there are two methods to do the scanning : mean and sum.

I knew averaging improves the image by increasing the signal to noise ratio.

Then what is the advantage of adding up the pixel value of all scan? Is this designed for weak signal image acquisition only?

By adding the value of each scan, you get amplification of your "real signal" not the noise.

Am I right about the "sum method"?

Another question is about the Line or Fram mode for averaging. People claim that fram averaging helps reduce photobleaching.

If I use one direction scan for a fixed sample and my system is a laser point scanning confocal microscope.

Would these two method apply the same photobleaching to my sample? In another word, would these two method have the same acquire time?

Any feedback would be appreciated.

Best,

Glen MacDonald-2

Re: confocal laser scanning mode mean vs sum

In reply to this post by SUBSCRIBE CONFOCALMICROSCOPY Sara Smith

On Mar 3, 2010, at 8:09 AM, yee sarah wrote:

> Dear confocal listers,
>
> I've got a few questions about our Zeiss LSM510 system. And I am wondering if you could help me out with this.
>
> Under the scan control panel in the software, there are two methods to do the scanning : mean and sum.
> .......
>
> Another question is about the Line or Fram mode for averaging. People claim that fram averaging helps reduce photobleaching.
> If I use one direction scan for a fixed sample and my system is a laser point scanning confocal microscope.
Averaging or summing only reduces photobleaching if it allows you to avoid increasing the dwell time and/or laser intensity to achieve the same SNR.
Line operations may be fractionally faster that frame operations but both accomplish the same thing. Line averaging is usually preferred for live specimens since you will have less motion artifact between pixels.
>
> Would these two method apply the same photobleaching to my sample? In another word, would these two method have the same acquire time?
>
> Any feedback would be appreciated.
Regards,
Glen

Glen MacDonald
Core for Communication Research
Virginia Merrill Bloedel Hearing Research Center
Box 357923
University of Washington
Seattle, WA 98195-7923 USA
(206) 616-4156
[hidden email]

Gert van Cappellen

Re: confocal laser scanning mode mean vs sum

In reply to this post by Goodhouse, Joseph G.

A few years ago, I have done some test with a LSM510 and found out that the S/N ratio improved more from line averaging than from slowing down the scan speed. Slowing down the scan speed might also give more photobleaching although I didn't measure this in my situation.

Gert van Cappellen
Optical Imaging Centre Erasmus MC
http://www.erasmusmc.nl/oic

Goodhouse, Joseph G. schreef:

Yi,

Yes summation is a way to collect weak signals but you will also be summing the noise. Not as much depending on how much of is random. You can filter it post collection to improve the S/N ratio. Another way to collect weak signals is to scan very slow to get good S/N ratio. This often works better. Uni-direction line scanning will not bleach more than frame mode. This you can easily test.

Joe Goodhouse
Confocal Core Lab Manager
Dept. of Molecular Biology
Princeton University

Washington Road

Princeton, NJ. 08544-1014
609-258-5432

Visit us at http://www.molbio1.princeton.edu/facility/confocal/

From: Confocal Microscopy List [[hidden email]] On Behalf Of yee sarah
Sent: Wednesday, March 03, 2010 11:09 AM
To: [hidden email]
Subject: confocal laser scanning mode mean vs sum

Dear confocal listers,

I've got a few questions about our Zeiss LSM510 system. And I am wondering if you could help me out with this.

Under the scan control panel in the software, there are two methods to do the scanning : mean and sum.

I knew averaging improves the image by increasing the signal to noise ratio.

Then what is the advantage of adding up the pixel value of all scan? Is this designed for weak signal image acquisition only?

By adding the value of each scan, you get amplification of your "real signal" not the noise.

Am I right about the "sum method"?

Another question is about the Line or Fram mode for averaging. People claim that fram averaging helps reduce photobleaching.

If I use one direction scan for a fixed sample and my system is a laser point scanning confocal microscope.

Would these two method apply the same photobleaching to my sample? In another word, would these two method have the same acquire time?

Any feedback would be appreciated.

Best,

Yi

SUBSCRIBE CONFOCALMICROSCOPY Sara Smith

Re: confocal laser scanning mode mean vs sum

In reply to this post by Goodhouse, Joseph G.

Hi Joe,

Thank you for your reply.

I am very interested in knowing how to filter the noise sum post collection.

Would you please explain more about this?

I think it will be very helpful for weak signal acquisition.

Best,

On Wed, Mar 3, 2010 at 12:23 PM, Goodhouse, Joseph G. <[hidden email]> wrote:

Yi,

Yes summation is a way to collect weak signals but you will also be summing the noise. Not as much depending on how much of is random. You can filter it post collection to improve the S/N ratio. Another way to collect weak signals is to scan very slow to get good S/N ratio. This often works better. Uni-direction line scanning will not bleach more than frame mode. This you can easily test.

Joe Goodhouse
Confocal Core Lab Manager
Dept. of Molecular Biology
Princeton University

Washington Road

Princeton, NJ. 08544-1014
609-258-5432

Visit us at http://www.molbio1.princeton.edu/facility/confocal/

From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of yee sarah
Sent: Wednesday, March 03, 2010 11:09 AM

To: [hidden email]
Subject: confocal laser scanning mode mean vs sum

Dear confocal listers,

I've got a few questions about our Zeiss LSM510 system. And I am wondering if you could help me out with this.

Under the scan control panel in the software, there are two methods to do the scanning : mean and sum.

I knew averaging improves the image by increasing the signal to noise ratio.

Then what is the advantage of adding up the pixel value of all scan? Is this designed for weak signal image acquisition only?

By adding the value of each scan, you get amplification of your "real signal" not the noise.

Am I right about the "sum method"?

Another question is about the Line or Fram mode for averaging. People claim that fram averaging helps reduce photobleaching.

If I use one direction scan for a fixed sample and my system is a laser point scanning confocal microscope.

Would these two method apply the same photobleaching to my sample? In another word, would these two method have the same acquire time?

Any feedback would be appreciated.

Best,

Yi

SUBSCRIBE CONFOCALMICROSCOPY Sara Smith

Re: confocal laser scanning mode mean vs sum

In reply to this post by Gert van Cappellen

I've tried using slow speed to improve S/N ratio. It did help, however it also introduced photobleach caused by longer dwell time. In my case, I used argon laser (quite powerful) for GFP imaging.

I think it might work better on synthetic probe than fluorescence protein.

On Wed, Mar 3, 2010 at 3:04 PM, Gert van Cappellen <[hidden email]> wrote:

A few years ago, I have done some test with a LSM510 and found out that the S/N ratio improved more from line averaging than from slowing down the scan speed. Slowing down the scan speed might also give more photobleaching although I didn't measure this in my situation.

Gert van Cappellen
Optical Imaging Centre Erasmus MC
http://www.erasmusmc.nl/oic

Goodhouse, Joseph G. schreef:

Yi,

Yes summation is a way to collect weak signals but you will also be summing the noise. Not as much depending on how much of is random. You can filter it post collection to improve the S/N ratio. Another way to collect weak signals is to scan very slow to get good S/N ratio. This often works better. Uni-direction line scanning will not bleach more than frame mode. This you can easily test.

Joe Goodhouse
Confocal Core Lab Manager
Dept. of Molecular Biology
Princeton University

Washington Road

Princeton, NJ. 08544-1014
609-258-5432

Visit us at http://www.molbio1.princeton.edu/facility/confocal/

From: Confocal Microscopy List [[hidden email]] On Behalf Of yee sarah
Sent: Wednesday, March 03, 2010 11:09 AM
To: [hidden email]
Subject: confocal laser scanning mode mean vs sum

Dear confocal listers,

I've got a few questions about our Zeiss LSM510 system. And I am wondering if you could help me out with this.

Under the scan control panel in the software, there are two methods to do the scanning : mean and sum.

I knew averaging improves the image by increasing the signal to noise ratio.

Then what is the advantage of adding up the pixel value of all scan? Is this designed for weak signal image acquisition only?

By adding the value of each scan, you get amplification of your "real signal" not the noise.

Am I right about the "sum method"?

Another question is about the Line or Fram mode for averaging. People claim that fram averaging helps reduce photobleaching.

If I use one direction scan for a fixed sample and my system is a laser point scanning confocal microscope.

Would these two method apply the same photobleaching to my sample? In another word, would these two method have the same acquire time?

Any feedback would be appreciated.

Best,

Yi

Goodhouse, Joseph G.

Re: confocal laser scanning mode mean vs sum

In reply to this post by SUBSCRIBE CONFOCALMICROSCOPY Sara Smith

Yee,

You can run a fine filter or a 3 x 3 kernel median filter over the images. You will find this under the Process tab. If you have a low expression GFP and are working with fixed samples I suggest using an anti-GFP antibody which will enhance the signal greatly. I do not suggest real slow scan speeds if you are optically sectioning a lot as you definitely will photo bleach.

Joe Goodhouse
Confocal Core Lab Manager
Dept. of Molecular Biology
Princeton University

Washington Road

Princeton, NJ. 08544-1014
609-258-5432

Visit us at http://www.molbio1.princeton.edu/facility/confocal/

From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of yee sarah
Sent: Wednesday, March 03, 2010 4:09 PM
To: [hidden email]
Subject: Re: confocal laser scanning mode mean vs sum

Hi Joe,

Thank you for your reply.

I am very interested in knowing how to filter the noise sum post collection.

Would you please explain more about this?

I think it will be very helpful for weak signal acquisition.

Best,

On Wed, Mar 3, 2010 at 12:23 PM, Goodhouse, Joseph G. <[hidden email]> wrote:

Yi,

Joe Goodhouse
Confocal Core Lab Manager
Dept. of Molecular Biology
Princeton University

Washington Road

Princeton, NJ. 08544-1014
609-258-5432

Visit us at http://www.molbio1.princeton.edu/facility/confocal/

From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of yee sarah
Sent: Wednesday, March 03, 2010 11:09 AM

To: [hidden email]

Subject: confocal laser scanning mode mean vs sum

Dear confocal listers,

I've got a few questions about our Zeiss LSM510 system. And I am wondering if you could help me out with this.

Under the scan control panel in the software, there are two methods to do the scanning : mean and sum.

I knew averaging improves the image by increasing the signal to noise ratio.

Then what is the advantage of adding up the pixel value of all scan? Is this designed for weak signal image acquisition only?

By adding the value of each scan, you get amplification of your "real signal" not the noise.

Am I right about the "sum method"?

Another question is about the Line or Fram mode for averaging. People claim that fram averaging helps reduce photobleaching.

If I use one direction scan for a fixed sample and my system is a laser point scanning confocal microscope.

Would these two method apply the same photobleaching to my sample? In another word, would these two method have the same acquire time?

Any feedback would be appreciated.

Best,

Dale Callaham

Re: confocal laser scanning mode mean vs sum

Biorad had a Kalman filter collection mode that was very effective. I
forget the formula, and I think it was handled in hardware on those
older machines, but it did work well. The formula was in the MRC-600
manual. Does anyone know if there is any equivalent method - maybe
through a macro? - available on the Zeiss 510?

Dale

Goodhouse, Joseph G. wrote:

>
>
> Yee,
>
> You can run a fine filter or a 3 x 3 kernel median filter over the
> images. You will find this under the Process tab. If you have a low
> expression GFP and are working with fixed samples I suggest using an
> anti-GFP antibody which will enhance the signal greatly. I do not
> suggest real slow scan speeds if you are optically sectioning a lot as
> you definitely will photo bleach.
>
> Joe Goodhouse
> Confocal Core Lab Manager
> Dept. of Molecular Biology
> Princeton University
>
> Washington Road
>
> Princeton, NJ. 08544-1014
> 609-258-5432
>
> Visit us at http://www.molbio1.princeton.edu/facility/confocal/
>
> *From:* Confocal Microscopy List
> [mailto:[hidden email]] *On Behalf Of *yee sarah
> *Sent:* Wednesday, March 03, 2010 4:09 PM
> *To:* [hidden email]
> *Subject:* Re: confocal laser scanning mode mean vs sum
>
> Hi Joe,
>
> Thank you for your reply.
>
> I am very interested in knowing how to filter the noise sum post
> collection.
>
> Would you please explain more about this?
>
> I think it will be very helpful for weak signal acquisition.
>
> Best,
>
> Yi
>
> On Wed, Mar 3, 2010 at 12:23 PM, Goodhouse, Joseph G.
> <[hidden email] <mailto:[hidden email]>> wrote:
>
> Yi,
>
> Yes summation is a way to collect weak signals but you will also be
> summing the noise. Not as much depending on how much of is random. You
> can filter it post collection to improve the S/N ratio. Another way to
> collect weak signals is to scan very slow to get good S/N ratio. This
> often works better. Uni-direction line scanning will not bleach more
> than frame mode. This you can easily test.
>
> Joe Goodhouse
> Confocal Core Lab Manager
> Dept. of Molecular Biology
> Princeton University
>
> Washington Road
>
> Princeton, NJ. 08544-1014
> 609-258-5432
>
> Visit us at http://www.molbio1.princeton.edu/facility/confocal/
>
> *From:* Confocal Microscopy List
> [mailto:[hidden email]
> <mailto:[hidden email]>] *On Behalf Of *yee sarah
> *Sent:* Wednesday, March 03, 2010 11:09 AM
>
>
> *To:* [hidden email]
> <mailto:[hidden email]>
>
> *Subject:* confocal laser scanning mode mean vs sum
>
> Dear confocal listers,
>
> I've got a few questions about our Zeiss LSM510 system. And I am
> wondering if you could help me out with this.
>
> Under the scan control panel in the software, there are two methods to
> do the scanning : mean and sum.
>
> I knew averaging improves the image by increasing the signal to noise
> ratio.
>
> Then what is the advantage of adding up the pixel value of all scan? Is
> this designed for weak signal image acquisition only?
>
> By adding the value of each scan, you get amplification of your "real
> signal" not the noise.
>
> Am I right about the "sum method"?
>
> Another question is about the Line or Fram mode for averaging. People
> claim that fram averaging helps reduce photobleaching.
>
> If I use one direction scan for a fixed sample and my system is a laser
> point scanning confocal microscope.
>
> Would these two method apply the same photobleaching to my sample? In
> another word, would these two method have the same acquire time?
>
> Any feedback would be appreciated.
>
> Best,
>
> Yi
>

Sarah Chacko

Re: confocal laser scanning mode mean vs sum

Going back to the original question, wouldn't summing just give the same result as averaging, only multiplied by a constant? Or is it different?

It's something I was wondering about, on my system the modes are called average and integrate, but I think integrate just sums the pixel values.

Sarah

________________________________________
From: Confocal Microscopy List [[hidden email]] On Behalf Of Dale Callaham [[hidden email]]
Sent: 04 March 2010 00:40
To: [hidden email]
Subject: Re: confocal laser scanning mode mean vs sum

Biorad had a Kalman filter collection mode that was very effective. I
forget the formula, and I think it was handled in hardware on those
older machines, but it did work well. The formula was in the MRC-600
manual. Does anyone know if there is any equivalent method - maybe
through a macro? - available on the Zeiss 510?

Dale

Goodhouse, Joseph G. wrote:

Boswell, Carl A - (cboswell)

Re: confocal laser scanning mode mean vs sum

Hi Sarah,

Summation and integration are the same. Each option sums the intensity at
each pixel from all the scans selected. This will bring out the signal from
the noise a bit simply because the noise (different from what I call
background, which is nonspecific staining or autofluorescence) is a random
event, while the signal is constant. This is used when the signal is weak
but still obvious compared to "non-signal".

I suppose this is better than slowing the scan speed because I"m speculating
that increasing pixel dwell time produces more photobleaching than several
fast scans. In other words, the same total dwell time as a single slow
scan, but broken up over several rapid scans, is better. Is there any math
or physics that supports or refutes this logic?

Cheers,
C

Carl A. Boswell, Ph.D.
Molecular and Cellular Biology
University of Arizona
520-954-7053
FAX 520-621-3709
----- Original Message -----
From: "Sarah Chacko" <[hidden email]>
To: <[hidden email]>
Sent: Thursday, March 04, 2010 2:22 AM
Subject: Re: confocal laser scanning mode mean vs sum

Going back to the original question, wouldn't summing just give the same
result as averaging, only multiplied by a constant? Or is it different?

It's something I was wondering about, on my system the modes are called
average and integrate, but I think integrate just sums the pixel values.

Sarah

________________________________________
From: Confocal Microscopy List [[hidden email]] On Behalf
Of Dale Callaham [[hidden email]]
Sent: 04 March 2010 00:40
To: [hidden email]
Subject: Re: confocal laser scanning mode mean vs sum

Biorad had a Kalman filter collection mode that was very effective. I
forget the formula, and I think it was handled in hardware on those
older machines, but it did work well. The formula was in the MRC-600
manual. Does anyone know if there is any equivalent method - maybe
through a macro? - available on the Zeiss 510?

Dale

Goodhouse, Joseph G. wrote:

Lingqing Zhang

Re: confocal laser scanning mode mean vs sum

In reply to this post by Sarah Chacko

Sarah,

I think you were right. Sum and Average Mode will give us the same S/N ratio. Just absolute values are different. In a 8 bit data format we can not summing up too many scans, has to watch out signal saturation.

Lingqing
**********************************************
Lingqing Zhang, PhD
Nonlinear Optical Imaging Lab, Manager
LM/EM Imaging Core Facility
West Virginia University
Phone: 304-293-5253
Email: [hidden email]

On Thu, Mar 4, 2010 at 4:22 AM, Sarah Chacko <[hidden email]> wrote:

Going back to the original question, wouldn't summing just give the same result as averaging, only multiplied by a constant? Or is it different?

It's something I was wondering about, on my system the modes are called average and integrate, but I think integrate just sums the pixel values.

Sarah

________________________________________
From: Confocal Microscopy List [[hidden email]] On Behalf Of Dale Callaham [[hidden email]]
Sent: 04 March 2010 00:40

To: [hidden email]
Subject: Re: confocal laser scanning mode mean vs sum

Biorad had a Kalman filter collection mode that was very effective. I
forget the formula, and I think it was handled in hardware on those
older machines, but it did work well. The formula was in the MRC-600
manual. Does anyone know if there is any equivalent method - maybe
through a macro? - available on the Zeiss 510?

Dale

Goodhouse, Joseph G. wrote:
>
>
> Yee,
>
> You can run a fine filter or a 3 x 3 kernel median filter over the
> images. You will find this under the Process tab. If you have a low
> expression GFP and are working with fixed samples I suggest using an
> anti-GFP antibody which will enhance the signal greatly. I do not
> suggest real slow scan speeds if you are optically sectioning a lot as
> you definitely will photo bleach.
>
> Joe Goodhouse
> Confocal Core Lab Manager
> Dept. of Molecular Biology
> Princeton University
>
> Washington Road
>
> Princeton, NJ. 08544-1014
> 609-258-5432
>
> Visit us at http://www.molbio1.princeton.edu/facility/confocal/
>
> *From:* Confocal Microscopy List
> [mailto:[hidden email]] *On Behalf Of *yee sarah
> *Sent:* Wednesday, March 03, 2010 4:09 PM
> *To:* [hidden email]
> *Subject:* Re: confocal laser scanning mode mean vs sum
>
> Hi Joe,
>
> Thank you for your reply.
>
> I am very interested in knowing how to filter the noise sum post
> collection.
>
> Would you please explain more about this?
>
> I think it will be very helpful for weak signal acquisition.
>
> Best,
>
> Yi
>
> On Wed, Mar 3, 2010 at 12:23 PM, Goodhouse, Joseph G.
> <[hidden email] <mailto:[hidden email]>> wrote:
>
> Yi,
>
> Yes summation is a way to collect weak signals but you will also be
> summing the noise. Not as much depending on how much of is random. You
> can filter it post collection to improve the S/N ratio. Another way to
> collect weak signals is to scan very slow to get good S/N ratio. This
> often works better. Uni-direction line scanning will not bleach more
> than frame mode. This you can easily test.
>
> Joe Goodhouse
> Confocal Core Lab Manager
> Dept. of Molecular Biology
> Princeton University
>
> Washington Road
>
> Princeton, NJ. 08544-1014
> 609-258-5432
>
> Visit us at http://www.molbio1.princeton.edu/facility/confocal/
>
> *From:* Confocal Microscopy List
> [mailto:[hidden email]
> <mailto:[hidden email]>] *On Behalf Of *yee sarah
> *Sent:* Wednesday, March 03, 2010 11:09 AM
>
>
> *To:* [hidden email]
> <mailto:[hidden email]>
>
> *Subject:* confocal laser scanning mode mean vs sum
>
> Dear confocal listers,
>
> I've got a few questions about our Zeiss LSM510 system. And I am
> wondering if you could help me out with this.
>
> Under the scan control panel in the software, there are two methods to
> do the scanning : mean and sum.
>
> I knew averaging improves the image by increasing the signal to noise
> ratio.
>
> Then what is the advantage of adding up the pixel value of all scan? Is
> this designed for weak signal image acquisition only?
>
> By adding the value of each scan, you get amplification of your "real
> signal" not the noise.
>
> Am I right about the "sum method"?
>
> Another question is about the Line or Fram mode for averaging. People
> claim that fram averaging helps reduce photobleaching.
>
> If I use one direction scan for a fixed sample and my system is a laser
> point scanning confocal microscope.
>
> Would these two method apply the same photobleaching to my sample? In
> another word, would these two method have the same acquire time?
>
> Any feedback would be appreciated.
>
> Best,
>
> Yi
>

Keith Morris

Re: confocal laser scanning mode mean vs sum

In reply to this post by Goodhouse, Joseph G.

Following on from Joe’s comments:

Hi Yi,

From the “Zeiss 510 LSM manual”:

SUM: “ ’Averaging’ can be performed in different ways, depending on whether the Mean or Sum method has been activated. If you are using the Mean method, the image information is generated by adding up all scans pixel by pixel and then calculating the mean value. In the Sum method, the pixel values of all scans are only added up, without a mean value being calculated.”

Thus ’sum’ simply doubles the object brightness after two scans then increases that by a third on the next, and so on [and the image will eventually become ‘overexposed’, completely losing detail in those bright regions, if overdone]. Although the object brightness will be increasing by a set proportion with each scan [as the object hopefully is always in the same place], random noise will be in a different place during each scan so the noise should not increase in brightness to the same extent. The noise will however appear in more parts of the image with each scan, which might be just as objectionable [although it is possible this noise can be reduced a tad with the ‘amplifier offset’ slider]. ‘Sum’ mode is generally used relatively rarely compared to ‘averaging’ [mean] noise reduction mode.

MEAN [AVERAGING, this is exactly the same as KALMAN using Bio-Rad Lasersharp]: “To create the image information using the Line average mode, each line (depending on the setting) is scanned 2, 4, 8 or 16 times during Scan Average, and then the average value per pixel is calculated. This minimizes noise interference during the scanning procedure.”

There should be no difference in bleaching rate using averaging [mean] noise reduction, whether you scan by line or frame mode. Frame mode averaging has the advantage that you can see each frame being scanned and the image averaged with every scan. Again as the object is always in the same place the averaging will not affect the object brightness [assuming minimal bleaching], however random noise will be in a different place with each scan and so that will typically halve in brightness with each scan. Thus after eight frame averages the random noise should be reduced in brightness by around eight times. In frame mode you can see this happening with each scan [the noise slowly reduces every scan]. If you can’t even see the scan line moving down the screen or any image improvement after say two averaging frame scans, then it is pointless to average scan the ‘image’ any more than that. Thus you can use ‘frame’ averaging to help you decide how many averaged frames are required for your sample - balancing this against the increased bleaching rate, doubling per extra ‘mean’ frame [image quality and bleach rate also depends on what you have set the pixel dwell time (scan speed) to]. Image quality and bleaching largely depends on how long it takes you to scan the sample [pixel dwell time], although with modern bleach resistant fluorochromes, anti-fade-mountants [and no large z-stack] there is also the ‘how long you can be bothered to wait for a subtle improvement in image quality’ factor.

Once you have determined, say that two times frame averaging is all that’s required for your sample you then switch to line averaging mode [set naturally to say 2x mean ‘averaging if that’s all that’s needed]. When you average in ‘line’ scanning mode the image appears as just one single frame already averaged [line by line]. This ‘line averaging’ scanning mode has the advantage of being a bit faster than mean frame scanning mode [as the laser galvo mirror doesn’t have to relocate all the way from bottom right to top left of the frame so much], and if you’re scanning say a live cell you probably won’t get a low noise but blurred effect if the cell moves slightly during scanning [or if you knock the microscope, when just the bottom and top of the image will be of slightly different regions]. The only downside of the line averaging mode is that you can’t tell if 2 or 16 averaging frames make any difference [as the averaged image appears in one go] – so you need Frame averaging to determine that.

CONTINOUS AVERAGING: And from the manual, there’s also a continuously ‘averaging’ mode [until you tell it to stop]:

“If the Frame average mode is used to create the image information, the complete frame is scanned 2, 4, 8 or 16 times, depending on the setting. The average value is recalculated after each frame scan. The Frame average mode also permits continuous averaging. For this, select the Continuous option in the Number selection box.

If you have selected the Continuous option, the Finish button for ending continuous averaging is displayed instead of the Cont. button. Use the Single button in this case to start continuous scanning. When you click on the Finish button, the scan currently in progress will be completed before the process is stopped.”

Regards

Keith

---------------------------------------------------------------------------
Dr Keith J. Morris,
Molecular Cytogenetics and Microscopy Core,
Laboratory 00/069 and 00/070,
The Wellcome Trust Centre for Human Genetics,
Roosevelt Drive,
Oxford OX3 7BN,
United Kingdom.

Telephone: +44 (0)1865 287568
Email: [hidden email]

Web-pages: http://www.well.ox.ac.uk/molecular-cytogenetics-and-microscopy

From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Goodhouse, Joseph G.
Sent: 03 March 2010 17:23
To: [hidden email]
Subject: Re: confocal laser scanning mode mean vs sum

Yi,

Joe Goodhouse
Confocal Core Lab Manager
Dept. of Molecular Biology
Princeton University

Washington Road

Princeton, NJ. 08544-1014
609-258-5432

Visit us at http://www.molbio1.princeton.edu/facility/confocal/

Dear confocal listers,

I've got a few questions about our Zeiss LSM510 system. And I am wondering if you could help me out with this.

Under the scan control panel in the software, there are two methods to do the scanning : mean and sum.

I knew averaging improves the image by increasing the signal to noise ratio.

Then what is the advantage of adding up the pixel value of all scan? Is this designed for weak signal image acquisition only?

By adding the value of each scan, you get amplification of your "real signal" not the noise.

Am I right about the "sum method"?

Another question is about the Line or Fram mode for averaging. People claim that fram averaging helps reduce photobleaching.

If I use one direction scan for a fixed sample and my system is a laser point scanning confocal microscope.

Would these two method apply the same photobleaching to my sample? In another word, would these two method have the same acquire time?

Any feedback would be appreciated.

Best,

Stanislav Vitha

Re: confocal laser scanning mode mean vs sum

In reply to this post by SUBSCRIBE CONFOCALMICROSCOPY Sara Smith

Hi Carl,
regarding the slow scan vs. summing (or averaging) several fast scans, the
triplet state relaxation (T-rex) comes to mind:
"...ensuring that transient molecular dark states with lifetimes >1 micros,
such as the triplet state relax between two molecular absorption events".

Donnert, G., C. Eggeling, and S.W. Hell, Major signal increase in
fluorescence microscopy through dark-state relaxation. Nature methods, 2007.
4(1): p. 81-6.

Stan Vitha
Microscopy and Imaging Center
Texas A&M University

On Thu, 4 Mar 2010 11:47:20 -0700, Carl Boswell <[hidden email]>
wrote:

>Hi Sarah,
>
>Summation and integration are the same. Each option sums the intensity at
>each pixel from all the scans selected. This will bring out the signal from
>the noise a bit simply because the noise (different from what I call
>background, which is nonspecific staining or autofluorescence) is a random
>event, while the signal is constant. This is used when the signal is weak
>but still obvious compared to "non-signal".
>
>I suppose this is better than slowing the scan speed because I"m speculating
>that increasing pixel dwell time produces more photobleaching than several
>fast scans. In other words, the same total dwell time as a single slow
>scan, but broken up over several rapid scans, is better. Is there any math
>or physics that supports or refutes this logic?
>
>Cheers,
>C
>
>Carl A. Boswell, Ph.D.
>Molecular and Cellular Biology
>University of Arizona
>520-954-7053
>FAX 520-621-3709
>----- Original Message -----
>From: "Sarah Chacko" <[hidden email]>
>To: <[hidden email]>
>Sent: Thursday, March 04, 2010 2:22 AM
>Subject: Re: confocal laser scanning mode mean vs sum
>
>
>Going back to the original question, wouldn't summing just give the same
>result as averaging, only multiplied by a constant? Or is it different?
>
>It's something I was wondering about, on my system the modes are called
>average and integrate, but I think integrate just sums the pixel values.
>
>Sarah
>
>
>________________________________________
>From: Confocal Microscopy List [[hidden email]] On Behalf
>Of Dale Callaham [[hidden email]]
>Sent: 04 March 2010 00:40
>To: [hidden email]
>Subject: Re: confocal laser scanning mode mean vs sum
>
>Biorad had a Kalman filter collection mode that was very effective. I
>forget the formula, and I think it was handled in hardware on those
>older machines, but it did work well. The formula was in the MRC-600
>manual. Does anyone know if there is any equivalent method - maybe
>through a macro? - available on the Zeiss 510?
>
>Dale
>
>Goodhouse, Joseph G. wrote:
>>
>>
>> Yee,
>>
>> You can run a fine filter or a 3 x 3 kernel median filter over the
>> images. You will find this under the Process tab. If you have a low
>> expression GFP and are working with fixed samples I suggest using an
>> anti-GFP antibody which will enhance the signal greatly. I do not
>> suggest real slow scan speeds if you are optically sectioning a lot as
>> you definitely will photo bleach.
>>
>> Joe Goodhouse
>> Confocal Core Lab Manager
>> Dept. of Molecular Biology
>> Princeton University
>>
>> Washington Road
>>
>> Princeton, NJ. 08544-1014
>> 609-258-5432
>>
>> Visit us at http://www.molbio1.princeton.edu/facility/confocal/
>>
>> *From:* Confocal Microscopy List
>> [mailto:[hidden email]] *On Behalf Of *yee sarah
>> *Sent:* Wednesday, March 03, 2010 4:09 PM
>> *To:* [hidden email]
>> *Subject:* Re: confocal laser scanning mode mean vs sum
>>
>> Hi Joe,
>>
>> Thank you for your reply.
>>
>> I am very interested in knowing how to filter the noise sum post
>> collection.
>>
>> Would you please explain more about this?
>>
>> I think it will be very helpful for weak signal acquisition.
>>
>> Best,
>>
>> Yi
>>
>> On Wed, Mar 3, 2010 at 12:23 PM, Goodhouse, Joseph G.
>> <[hidden email] <mailto:[hidden email]>> wrote:
>>
>> Yi,
>>
>> Yes summation is a way to collect weak signals but you will also be
>> summing the noise. Not as much depending on how much of is random. You
>> can filter it post collection to improve the S/N ratio. Another way to
>> collect weak signals is to scan very slow to get good S/N ratio. This
>> often works better. Uni-direction line scanning will not bleach more
>> than frame mode. This you can easily test.
>>
>> Joe Goodhouse
>> Confocal Core Lab Manager
>> Dept. of Molecular Biology
>> Princeton University
>>
>> Washington Road
>>
>> Princeton, NJ. 08544-1014
>> 609-258-5432
>>
>> Visit us at http://www.molbio1.princeton.edu/facility/confocal/
>>
>> *From:* Confocal Microscopy List
>> [mailto:[hidden email]
>> <mailto:[hidden email]>] *On Behalf Of *yee sarah
>> *Sent:* Wednesday, March 03, 2010 11:09 AM
>>
>>
>> *To:* [hidden email]
>> <mailto:[hidden email]>
>>
>> *Subject:* confocal laser scanning mode mean vs sum
>>
>> Dear confocal listers,
>>
>> I've got a few questions about our Zeiss LSM510 system. And I am
>> wondering if you could help me out with this.
>>
>> Under the scan control panel in the software, there are two methods to
>> do the scanning : mean and sum.
>>
>> I knew averaging improves the image by increasing the signal to noise
>> ratio.
>>
>> Then what is the advantage of adding up the pixel value of all scan? Is
>> this designed for weak signal image acquisition only?
>>
>> By adding the value of each scan, you get amplification of your "real
>> signal" not the noise.
>>
>> Am I right about the "sum method"?
>>
>> Another question is about the Line or Fram mode for averaging. People
>> claim that fram averaging helps reduce photobleaching.
>>
>> If I use one direction scan for a fixed sample and my system is a laser
>> point scanning confocal microscope.
>>
>> Would these two method apply the same photobleaching to my sample? In
>> another word, would these two method have the same acquire time?
>>
>> Any feedback would be appreciated.
>>
>> Best,
>>
>> Yi
>>

Schebique

Re: confocal laser scanning mode mean vs sum

In reply to this post by SUBSCRIBE CONFOCALMICROSCOPY Sara Smith

Dear confocalists,

the most common scanning mode is line averaging. Usually 8x or 16x times line averaging strongly improves details in the image compared to lower line averaging. Depending on sample brightness this can be sufficient in the mean of S/N ratio. If there is stil too much noise in the image, frame averaging can be efectivelly used to diminished the amount of noise and subsequent highlight details (although images seems subjectively more blure :)). However, this is possible only with well fixed samples and stable stage! For scanning living cells line averaging is only option.

In the case of weak signals there are several ways how to obtain better image. To increase laser power obviously leads to higher photobleaching. This can be done only with higly stable fluorochromes like Alexa fluor. The other possibilities is to 1) use smaller image resolution (512x512 instaed of 1024 to 1024) in the case that pixel size is not limiting factor and you don't need small details, 2) open the pinhole - this leads to loss of confocality, but you obtain more light on detector. 3) similar is to compromise laser beam thicknes by beam expander (Leica SP2 - I don't know other systems). 4) Better signal can be detected by slowing down the scanning - mentioned increase of photobleaching is not IMHO so crucial. 5) Instead of averaging images can be summed (or integrated).

I case of low signal with needs to have good resolution I would decrease laser power, slower scan speed, switch to 12 or 16 bit image depth and use image integration until reasonable grey level values in the image appear. If number of sums offered by system is not sufficient, this can be simulated by XY(Z)T mode and images can be summed in postproccesing e.g. in imageJ. Advantage of this approache is low photobleaching and very low noise is integrated with correctly adjusted PMT offset and gain. Disadvantage is that this is very time-consuming :) However, very good images can be obtained also with well stained samples with limited confocality by pinhole sizes about 0.5 Airy.

New Leica SP5 system also offer the possibility of summing line as well as frame averaged images. But I did never try.

Sincerely Ondrej Sebesta

Dne 3.3.2010 17:09, yee sarah napsal(a):

Dear confocal listers,

I've got a few questions about our Zeiss LSM510 system. And I am wondering if you could help me out with this.

Under the scan control panel in the software, there are two methods to do the scanning : mean and sum.

I knew averaging improves the image by increasing the signal to noise ratio.

Then what is the advantage of adding up the pixel value of all scan? Is this designed for weak signal image acquisition only?

By adding the value of each scan, you get amplification of your "real signal" not the noise.

Am I right about the "sum method"?

Another question is about the Line or Fram mode for averaging. People claim that fram averaging helps reduce photobleaching.
If I use one direction scan for a fixed sample and my system is a laser point scanning confocal microscope.

Would these two method apply the same photobleaching to my sample? In another word, would these two method have the same acquire time?

Any feedback would be appreciated.

Best,

Yi
Příchozí zpráva neobsahuje viry.
Zkontrolováno AVG - www.avg.cz 
Verze: 8.5.435 / Virová báze: 271.1.1/2720 - Datum vydání: 03/03/10 07:34:00

  

-- 
------------------------------------------
Mgr. Ondřej Šebesta
Laboratory of Confocal and Fluorescence Microscopy
Faculty of Science, Charles University in Prague
Vinicna 7
128 44 Prague
Czech Republic

Phone: +420 2 2195 1943
e-mail: [hidden email]