• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 29 subscribers
  • Views 1284 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Use of 3rd party NIR prediction models

Mate Fuchsz
Prodigy 25 points

Hi Community,

is it possible to use existing prediction models with the NIRscannano equipment and software? It would be possible to get for my research prediction models from other researchers, but they made these models with other NIR systems (e.g. Polytec). The measurement range was the same. The last option would be to buy a prediction model from Auris.

Do you have any idea, if it can work, or i have to build my own model? So i can do the measurement with the NIRscannano, than i can do the chemical analyses in the lab and join the results, but it would be really a lot of work, thats what i want to avoid.

Thanks for your help.


Cheers, Mate

  • 0
    Prodigy 10 points

    Hi Mate

    I am also interested in applying chemometrics to the TI modules scan data. I am currently looking at the use of RStudio with the chemometrics and pls packages.

    What is your experience so far. Please contact me at PeterRobertson (at) hispeed.ch

  • 0
    TI__Guru 54585 points

    Hello Mate,

    Welcome to the DLP Advanced Light Control section of the TI-E2E Community.

    DMD based systems are relatively new, so whether it can be used with existing prediction models is not certain.

    In theory, it should be possible, but some correlation to existing samples may be needed to adjust unknown sample readings to the models.

    I hope there are other community members that may be able to add to the conversation on this.

    Fizix

  • 0
    Prodigy 25 points
    Hi Peter,

    i have contacted some companies with prediction models, for FOSS, Bruker, Carl Zeiss und tec5 labratory and handheld devices. Most of them have never heard about the TI NIRScan units, so far they had no idea if their prediction model can work with the NIRscan. One technician said, it would be possible to use the NIRscan with their model, the problem is, that they should use the NIRscan for a while, to do several test measurements to know where is a chnage in the existing model necessary.

    I would like to use the NIRscan in the agriculture, assuming that the prediction model licence for one unit costs ca. 5-6000 EUR plus the technician hours for the customization, meetings, etc. for one model the price could be more than 10000 euro. This is actually such a high price, that i can build my own prediction model and dataset with probes analysed in a normal labratory with wet chemical methods, would have ca. 500 different probes - which is generally more than enough for an own prediction model and i can use it on as much NIRscan units as i want.

    It would be nice if you could inform us about your experiences with RStudio.

    Cheers, Máté