How to use the Point Spread Function (PSF) analysis in Daybook (Webinar)

Is there a hint giver to improve the PFS?

• No there is no hint giver. PSF is a general indicator so it could be hard to automate relevant hints based on results other than very general ones.

How can the SBR be higher than the dynamic range of the recording?

• The SBR is a ration between the higher value of the PSF and the lowest value. If the image is made so that the lowest value is 0 the SBR is expressed as “infinity”.

Is it possible to make from a 4 color bead an analysis for the axial chromatic shift since that is also not possible with the Argolight slide yet.

• We are working on axial measurements both with PSF and our hardware solution. But it is not available yet.

The SNR values for x and y are different, why should that be? What is the reproducibility of these values?

• There is no consensus on the way to calculate the SNR for images with beads. We choose to base the SNR on the RMSE of the fitting, as the X and Y line profile are slightly different, the fitting is slightly different, hence the SNR is slightly different.

can multiple beads be analyzed in one dataset simultaneously?

• Not yet.

Can multiple beads of the same configuration and the same date be saved in the manager? Or is there a way to calculate the average results of many beads?

• No there is no averaging of result yet. You can save several beads result in the database, but you must give them different time (limitation is down to second, so just add a few second between each).

Is there a way to do some batch PSF analysis, either by automated beads detection on a single image stack with multiple beads or on several Z stack files cropped to single beads?

• Not yet.

multi-colour analysis?

• Not yet.

The software does not make use of the metadata when saving? So it does not prevent from storing PSFs for example from different systems or objectives in the same database?

• No it does not, but it is by design, we want to leave it to the user to save result in the profile they want.

What determines the fitting function you use for analysis?

• You can select the fitting function in the setting panel of Daybook. The type of fitting is based on the standard/protocol you want to follow. Single gaussian is the most common. But you want to compare results based on the same fitting

Is there a list of all image formats usable on the software?

• We use BioFormat, so the list is on their website. We added a couple of additional format such as Yokogawa’s.

if I understand right the system calculates the PSF out of a stack of x-y images? Is it possible to integrate also x-z stack ?

• No

is the data saved in the data manager just saved localy or is it possible to use it in a network?

• Locally

Your 2nd analhysis shows a lot of spherical aberration, in the axial profile. Doesn’t the software discriminate data that are consistent (or not) with Gaussian fitting ?

• No, the software does not discriminate the fitting based on the data. But we display the R² of the fitting that indicates if the fitting is adequate.

Does it make sense to analyse PSF in different areas of the field of view?

• It does if you want to see if you have image quality fluctuation in the field of view. If you want to compare result through time, you want to compare result from the same image area(s) to get relevant information.

How does one automate analyses of numerous beads in the field

• One does not, yet.

Could the daybook analyze the xy PSF without Z data?

• No it cannot.

Still about the SBR, what is the offset value?

• The offset value is the lowest value of the fit.

The “offset” value can include an instrumental offset, which should be removed. Can this be specified?

• If cannot be specified, but you can do a background substraction on the image using an empty image, hence efficiently substracting the instrumental offset.

Can you generate theoretical psfs based on the objective, camera, and sample properties?

• No, but there are several software that do (SVI, PSF creator etc.)

How do the result of your analysis fit with open Source tools like MetroloJ and others?

• Our results are consistent with other solutions.

what is the practical usuable recommendation regarding the sampling rate in xyz for aquire a 100nm bead

• We do not give recommendation yet.

Do you plan to incorporate some of the ImageJ solutions, e.g. metrolo, etc, so that the analysis can be made with the plugin and data saved in your software?

• It is not on the roadmap, but we are not against it.

can you describe shortly you business model? Is the software free and it provides full functionality or there is also a paid version of the soft.

• All the one-shot analysis is free as well as exporting in CSV, PSF…
• The monitoring part is 40€ per month per system.

More information : www.argolight.com/daybook3

Is there a plan to also look at the OTF of PSF to evaluate the performance of the microscope?

No there is no official feature in development.

Would there be a way to also evaluate the type of aberrations?

No there is not such type of analysis yet.

I wonder if you can measure vibration using the software?

Yes you can, but with another device: either the HM or LM Argolight slides.

How does the analysis of beads relate to the line resolving power test on the Argolight slides? Are the tests complimentary?

These tests are complimentary: the PSF measures the spread of a single point like object, whereas the Resolving power measures the ability to distinguish two objects based on a contrast value. This

What is the lowest SNR and SBR you recommend for reliable and accurate fittings ?

Lowest values should be SNR: 10 to 1,  SBR: 3 to 1

Do you plan on adding subdiffraction features to argolight slides?

The lines on the SIM slides are below 110nm, so they can be considered sub diffraction. However, we cannot yet product point like structures sufficiently small as to be used for PSF.

Will daybook flag acquisition settings that would negatively affect the PSF?

No it doesn’t.

The Confocal ISO specifies really small pixels (10 times less than resolution) what is your comment on this?

We think it is a strict approach (especially in Z). However, until there is a modification of the ISO Standard, you must follow these values to be ISO compatible.