The question is about the coordinates at the SPECT processing tool.

The window shows the x, y, z coordinates. In results from several patients they were all positive, however there are significant clusters in both sides. The figures shows the results from one case, upon seting the crosshairs for the first and the third, the blobs are in opposite sides. How I am not getting negative values in any of the items?

It really seems to be a bug, because the Transform Viewer shows the negative values.

Also, saving the coordinates in a text file (with the saving statistics button) gives me the all-positive results as well.

What I am missing?

Thanks,

luis

P.S. It seems I cant upload two images, here are links to them in my dropbox:

1. [url=https://dl.dropboxusercontent.com/u/508558/Captura%20de%20tela%20de%202016-11-16%2009%3A56%3A08.png]https://dl.dropboxusercontent.com/u/508558/Captura%20de%20tela%20de%202016-11-16%2009%3A56%3A08.png[/url]

2. [url=https://dl.dropboxusercontent.com/u/508558/Captura%20de%20tela%20de%202016-11-16%2009%3A55%3A37.png]https://dl.dropboxusercontent.com/u/508558/Captura%20de%20tela%20de%202016-11-16%2009%3A55%3A37.png[/url]

> Not sure. Your Dropbox links give an error "Can't find the page you're
> looking for."
>

The coordinates depend on the file format and perhaps the viewer you are
using. I do not pay attention to the coordinates as long as I can see the
results and know they are in the correct orientation and are properly
coregistered to the reference image I am using. In that case you could just
use the coordinates of the viewer. I use MRIcron and MeVisLab to display
results as overlays along with other images I have processed such as PET.

Also I am puzzled why you would want to do a group analysis of SPECT
results. They vary so dramatically for different patients.

Thanks Larry,

The links are working. I will just paste them, as the forum helper messes with the link:

1. https://dl.dropboxusercontent.com/u/5085...
2. https://dl.dropboxusercontent.com/u/5085...

You have to copy and paste in the browser.

> In that case you could just use the coordinates of the viewer.

It is not that easy... the coordenates in the SPECT Processing Tool seems not to be Talairach. It seems it gets converted into something else (figure). I am tring to find out what coordinates are these, in the viewer, however, it gets converted to MNI space coordinates (there is also another label: C_AP TC). I am trying to find info in the documentation.

> Also I am puzzled why you would want to do a group analysis of SPECT results. They vary so dramatically for different patients.

Indeed, they vary a lot and the data will probably end up being negative, as per the preliminary findings. At this point, I am really curious about previous publication of icta-interictal SPECTS. I couldn't find two papers that used the same method to acquire subtractions. This is the reason why I am still insisting in ISAS, ISAS is open source, open to audit, and has been around for years...

>Your pictures show the same coordinates in the SPECT processing tool as in the viewer after all. One is xyz and the other ijk. Ignore the Mask value.

Keep in mind that the SPECT processing tool is only valid for the MNI space, not the patient MRI. The template T1 has to be used for creating the overlay.

BTW, check out the article from Mayo, comparing SISCOM, ISAS and their new method STATISCOM (search on this term). Both of later perform about twice as well as SISCOM, but only ISAS is available to us.

You may also be interested in my PET methods. You can find it by searching YouTube for Multimodal Imaging. About to publish an article showing it is much much better than SPECT, ISAS, or conventional PET.