Populate tools on niQC Website

48 views
Skip to first unread message

Pradeep Reddy Raamana

unread,
May 12, 2019, 10:27:02 AM5/12/19
to NIQC
Hi Everyone,

We have a live website
big thanks to Annibal Solon.

We chose an easy to manage format (in Markdown) and populated the tools page quickly:

It is by no means exhaustive and need your help to add various other tools and improve it. It is as simple as forking the niQC repo, editing the markdown page at https://github.com/INCF/niQC/tree/master/content/pages/tools and sending us a PR.

Please suggest other ideas for improving the website, which is meant to be the public face for this group. I hope to announce in few weeks (end of May) in time for OHBM, so the community can learn more about our efforts. 

Thanks,
Pradeep

--
Postdoctoral fellow
Neuroinformatics,
3560 Bathurst Street, Toronto, ON
Canada, M6A 2E1.
Phone: +1 416-785-2500 x6408

Ben A. INGLIS

unread,
May 12, 2019, 1:03:23 PM5/12/19
to Pradeep Reddy Raamana, NIQC
Thanks, Pradeep! How about a page for example data sets that folks can use to test their own QC software on? I'm thinking four types of data: phantom, and three from human brains, "wild,"  "spiked" and "good." The phantom data could potentially be sub-divided into "phantom-spiked" and "phantom-good." Here, I mean for good data to be that ostensibly acceptable data that doesn't exhibit any uncontrollable errors. "Wild" data are exactly that; data from real studies. So I suppose these could also be split into "wild - good" and "wild - bad" with the latter having clear demonstrations of one or more artifacts that would lead most people to reject the data from further analysis.

"Spiked" data probably needs renaming, but I can't think of a suitable alternative word right now. (I mean contaminated, or intentionally bad data.) These data have artifacts created intentionally, ideally but not necessarily one at a time. When I am thinking about QC as an engineering project I tend to think of pairs of data, starting with "phantom - good" and "phantom - spiked" pairs, then moving to pairs of "good" and "spiked" data from human brains, ideally collected back-to-back in the same session. This way we are controlling most of the variables and contaminating our data systematically. In my opinion, "wild" data should be the very last category to be fed into a QC pipeline. Until the pipeline is known to handle spiked data properly, analyzing wild data is a reverse-engineering project. The forward problem is far more tractable!


--
You received this message because you are subscribed to the Google Groups "niQC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to niqc+uns...@googlegroups.com.
To post to this group, send email to ni...@googlegroups.com.
Visit this group at https://groups.google.com/group/niqc.
To view this discussion on the web visit https://groups.google.com/d/msgid/niqc/CAKYYBvp_SE5YJ%2BYVuhZh58eoOJQuL%2BBwV_enjfgEP-RrTs9qvg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Pradeep Reddy Raamana

unread,
May 12, 2019, 1:07:12 PM5/12/19
to Ben A. INGLIS, NIQC
Excellent idea, thanks Ben.

Just added a Resources page with links to educational content and papers. Datasets would be an essential addition to that page, or even on its own at the top. 

I will add it soon - any more volunteers to help maintain the website?


For more options, visit https://groups.google.com/d/optout.

Stephen Strother

unread,
May 12, 2019, 6:33:10 PM5/12/19
to ni...@googlegroups.com
I think Ben's idea is a good one and we can contribute wild fBRIN phantom scans with both normal and spiked data of a variety of kinds. We will be releasing ~ 4 years worth of monthly fBIRN scans across 12 scanners before our educational session in Rome as I intend to discuss some of the results.

Cheers, Stephen

For more options, visit https://groups.google.com/d/optout.


-- 
----------------------------------------------------------------------------------------------------------------
To reduce e-mail overload follow the e-mail charter of Chris Anderson: http://emailcharter.org/.

Stephen Strother, PhD
Senior Scientist, Rotman Research Institute, Baycrest
Professor of Medical Biophysics, University of Toronto
E-mail: sstr...@research.baycrest.org
Tel. Office: 416-785-2500 x2956
Fax: 426-785-2862

Pradeep Reddy Raamana

unread,
May 18, 2019, 1:10:40 PM5/18/19
to Stephen Strother, NIQC
In addition to this awesome release, I am giving an overview of the various QC/QA tools available in this course, hence I’d encourage tool developers to send a PR to us to add it to the website, so I’ll try to cover it. It would be even better (easier for me) if you can send a slide or two on your tool, detailing both its purpose (primary focus), limitations (what it doesn’t do) and who should use it. This is a great way to promote your new and existing tools, as many potential users would be seeing it.

Course description: 

Just added a page Datasets page under Resources now - if you have, or come across anything useful for others, please send it our way.

Thanks,
Pradeep


For more options, visit https://groups.google.com/d/optout.

Pradeep Reddy Raamana

unread,
May 21, 2019, 10:30:49 AM5/21/19
to Stephen Strother, NIQC
Hi Everyone,

Many reasons it would be great if you would help me with your slides on your tool: 1) ensure we don't miss covering it, 2) ensure I don't misrepresent it, and 3) great exposure and potential contributors.

Here is a template I am planning to follow - please download and send it to me by June 1st, thanks:

Even if we won't be able to cover it extensively in the course, lots of people will come back to these later on.

Thanks,
Pradeep

Kay Chioma Igwe

unread,
May 28, 2019, 11:02:39 PM5/28/19
to Pradeep Reddy Raamana, Stephen Strother, NIQC
By chance, for the meeting at OHBM, will there be a Zoom or Skype link for those of us that can't make it to the meeting?

- Kay

Pradeep Reddy Raamana

unread,
May 30, 2019, 1:40:34 PM5/30/19
to Kay Chioma Igwe, Stephen Strother, NIQC
Dear Kay,

That's a great idea - I will try to facilitate it, but depending on the setup and facilities there, this might not work too well. I will keep everyone posted.

As I mentioned in some other email, I am hoping to run periodic virtual meetups (monthly or biweekly depending on availability and interest) to discuss the survey results, get domain experts to teach, tool makers to demo their tools etc. Stay tuned.

Thanks,
Pradeep

Kay Chioma Igwe

unread,
May 31, 2019, 9:04:03 PM5/31/19
to Pradeep Reddy Raamana, Stephen Strother, NIQC
Sounds good.

- Kay
Reply all
Reply to author
Forward
0 new messages