Peak Detection and Integration

[Rannulph]

I am working on a project involving Laser Induced Flourescence. I have been able to output the desired graph and am now facing the next obstacle - peak detection and integration of peak to determine area under curve. I assume this to be a previously accomplished task and wanted to see if anyone had suggestions, example programs, or preferred methods. Thank you very much. Any help would be extremely appreciated.

[BEWIN]

In the time domain palette, you can find Peak Detector.vi. Once the peaks found,you could extract them out of the spectrum assuming a known FWHM of the peaks. With Integral x(t).vi you could do the integration of the peaks

[Rannulph]

Thank you very much for your reply. The concern I have is that I need to be precise in the decision of where to start and stop integration - a possible downfall of the FWHM approach. I have been attempting to use the threshold function to determine the start and stop of integration with little success - quite a few bugs in program. Does anyone know of a simple approach or possibly have a tested program? Thanks again for your help.

[BEWIN]

I don't have examples. Probably there you could also find automatic peak detection and integration vi's for chromatography (GC, HPLC)

[jfxcoll]

I am working on the GC problem myself. I have just started l I am working with the peak picking VI now. one thing is clear you need to load a typical signal from the GC to a file for development purposes. I would think that NI had crossed this bridge a long time ago.
The PP vi actually finds the positive an negative 2nd derivatives the peaks being the negative 2nd D's My plan was to integrate from positive 2nds and hope the threshold function works well.
Here is the integrator that I am trying to creaet a VI for
. http://www.chem.agilent.com/Library/usermanuals/Public/G1213-90120.pdf
Of course intigtaing from every minimum is going to be intersting. I guess it workes by size of the area under the curve but the intigtation funtions (it is my impression) of much of the chromatogrphy is much more compicated.
Finding min to the right and left of a peak is one way. Sounds simple enough the there is the slope (Derivitive ) of the cureve at FWHM
It seems incredable the the VI for the 3396 has not been produced. For any chart one would wish to run the volume under the curve is at least half the point... at least in our labs. Retention times are nice. Then the peak volumes need to be added and %'s worked out.
The grand finale is reporting the peaks by retention time and volumes along with the chart and retention times.
The code exixts already but someone at Hewlett-Packard. i mean Agiient I mean.... keysifght is sittting on it.

[jfxcoll]

I might try intigrating from valley to valley

[natasftw]

11-12-2004 03:33 AM

I'm pretty sure they're no longer working on this.

[jfxcoll]

I am sure they have implemented it in every digital O scope they ship.

[natasftw]

How many O scopes do you think BEWIN has shipped?