Extinction coefficients in GetExtinctions.m

[Johann]

Hello everyone,

I had a question about the extinction coefficients that are in the GetExtinctions.m file being used by hmrR_Intensity2OD.m. This function is very useful because it gives the ability to use different sets of extinction coefficients to get the optical densities from intensities. I was looking at the dataset used by default by GetExtinctions.m which is data from Gratzer and Kollias compiled by Scott Prahl, but it appears that the dataset in this file is different from the data compiled by Scott Prahl available online. Before 650 nm, the two extinction sets are the same, but from 650 nm they differ. I was wondering what was the dataset used from 650 nm?

GetExtinctions.m: https://github.com/BUNPC/Homer3/blob/889dbebcf349ddf38bdd00958767ae4e1e6d184a/FuncRegistry/UserFunctions/GetExtinctions.m#L356

[...]

640     442       4345.2;

642     423.6    4216.8;

644     405.2    4088.4;

646     390.4    3965.08;

648     379.2    3857.6;

650.0  506.0    3743.0;   <=

652.0  488.0    3677.0;

654.0  474.0    3612.0;

656.0  464.0    3548.0;

658.0  454.3    3491.3;

660.0  445.0    3442.0;

[...]

Online Scott Prahl's data: https://omlc.org/spectra/hemoglobin/summary.html [...] 640    442       4345.2 642    423.6    4216.8 644    405.2    4088.4 646    390.4    3965.08 648    379.2    3857.6 650    368       3750.12   <= 652    356.8    3642.64 654    345.6    3535.16 656    335.2    3427.68 658    325.6    3320.2 660    319.6    3226.56 [...]

I found this post in the mailing list talking about it but couldn't find any answer: https://mail.nmr.mgh.harvard.edu/pipermail/homer-users/2017-September/001052.html

Sorry if this was answered already...

Thanks for your help,

Johann

[Boas, David]

Fascinating discovery.

I compiled the numbers in this function over 10 years ago… maybe more than 15 years ago… from the sources referenced in the file. I can’t imagine what happened.

 

Can you make a plot of the differences and share it here?

I wonder how big they are.

We should probably update this file.

 

David

--
You received this message because you are subscribed to the Google Groups "Homer-fNIRS and AtlasViewer Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to homer-fnirs...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/homer-fnirs/f9656234-a32a-4d6a-8527-6f9e2cdf3784%40googlegroups.com.

[Johann]

Hello David,

Thank you very much for your reply.

Here is the dataset from Scott Prahl's website:

And here Homer's dataset:

So it looks like there are discontinuities in the Homer dataset at 650 and 900 nm (just noticed that after 900nm the data is back to normal), whereas Scott Prahl's data look smoother.

Interestingly enough, here is the dataset used in NIRSLab (from NIRx):

I believe it uses the same dataset as Homer. We cannot see the data before 650 nm but there is the discontinuity at 900 nm.

Is there a link between the development of Homer and the development of NIRSLab (like one relying on the other)?

Kind regards,

Johann

---

To unsubscribe from this group and stop receiving emails from it, send an email to homer...@googlegroups.com.

[Boas, David]

Mystery solved!

 

Looking at getExtinctions.m I see we created the file in 1999!

I also see that on April 27, 2001 we noted that “Maria Angela Updated the extinction coefficients for Hb and HbO”

Well, it turns out that Maria Angela always uses the extinction coefficients published in

Wray, S., Cope, M., Delpy, D., Wyatt, J., Reynolds, E. (1988). Characterization of the near infrared absorption spectra of cytochrome aa3 and haemoglobin for the non-invasive monitoring of cerebral oxygenation Biochimica et Biophysica Acta (BBA) - Bioenergetics 933(1), 184-192. https://dx.doi.org/10.1016/0005-2728(88)90069-2

 

If you look at table II in that paper, you find exactly the extinction coefficients from 650nm and above to 900nm that we have in getExtinctions.m. While Wray goes up to 1042nm, we apparently only copied from 650 to 900nm into the getExtinctions.m

 

So, this means that for everyone using this function since 2001 for this 650 to 900nm wavelength range, we have by default been using the Wray extinction coefficients.

 

I will change the function to explicitly cite that Wray is the default and I will update the function to have the proper “Prahl” option available for those who want it. That new function will appear in the Homer3 repository on GitHub. I will also update it for Homer2 on nitrc.

 

I am glad we finally got to the bottom of that!

 

David

To unsubscribe from this group and stop receiving emails from it, send an email to homer-fnirs...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/homer-fnirs/098b5c84-a1f2-4552-a2af-4726aa0ebd0b%40googlegroups.com.

[Johann]

Well done David, thanks a lot for finding out.

Wray data matches what's in Homer indeed. Please note however that some of the points seem interpolated from Wray et al. 1988. Indeed while we have the wavelengths for even numbers in Homer (650, 652, 654, ...), those are a bit different in the Wray et al. paper (650, 653, 657, 660, ...). But not that much a problem I believe as the Homer function is interpolating anyway if the wavelengths do not match exactly the data in the GetExtinctions.m file.

Thank you very much for your help David.

Best wishes,

Johann

---

Le jeudi 7 mai 2020 13:49:21 UTC+1, Boas, David a écrit :

Mystery solved!

 

Looking at getExtinctions.m I see we created the file in 1999!

I also see that on April 27, 2001 we noted that “Maria Angela Updated the extinction coefficients for Hb and HbO”

Well, it turns out that Maria Angela always uses the extinction coefficients published in

Wray, S., Cope, M., Delpy, D., Wyatt, J., Reynolds, E. (1988). Characterization of the near infrared absorption spectra of cytochrome aa3 and haemoglobin for the non-invasive monitoring of cerebral oxygenation Biochimica et Biophysica Acta (BBA) - Bioenergetics 933(1), 184-192. https://dx.doi.org/10.1016/0005-2728(88)90069-2

 

If you look at table II in that paper, you find exactly the extinction coefficients from 650nm and above to 900nm that we have in getExtinctions.m. While Wray goes up to 1042nm, we apparently only copied from 650 to 900nm into the getExtinctions.m

 

So, this means that for everyone using this function since 2001 for this 650 to 900nm wavelength range, we have by default been using the Wray extinction coefficients.

 

I will change the function to explicitly cite that Wray is the default and I will update the function to have the proper “Prahl” option available for those who want it. That new function will appear in the Homer3 repository on GitHub. I will also update it for Homer2 on nitrc.

 

I am glad we finally got to the bottom of that!

 

David

 

 

From: <homer...@googlegroups.com> on behalf of Johann <johann....@gmail.com>

To view this discussion on the web visit https://groups.google.com/d/msgid/homer-fnirs/098b5c84-a1f2-4552-a2af-4726aa0ebd0b%40googlegroups.com.

[Boas, David]

The GetExtinctions.m file has been uploaded in the svn version of Homer2 at nitrc.org and for Homer3 at GitHub

And here Homer's dataset:

So it looks like there are discontinuities in the Homer dataset at 650 and 900 nm (just noticed that after 900nm the data is back to normal), whereas Scott Prahl's data look smoother.

 

Interestingly enough, here is the dataset used in NIRSLab (from NIRx):

To unsubscribe from this group and stop receiving emails from it, send an email to homer-fnirs...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/homer-fnirs/a83be2cd-9c42-4cb9-aa1a-b2ee44a829e2%40googlegroups.com.