December 15, 2020 tedana newsletter

1 view
Skip to first unread message

tedana-newsletter

unread,
Jan 18, 2024, 11:11:40 AMJan 18
to tedana-newsletter

Ported message from our previous tinyletter.com system:

We hope that everyone is doing well. It has been some time since our last newsletter. We have continued to expand and improve tedana. In particular, we have added a feature, started at the hackathon, where you can use html-based dynamic reports to view the outputs of tedana. The community has also formalized a new governance structure.

 

OHBM

 Multi-echo fMRI has continued to have a solid presence at OHBM. You can view a curated list of all multi-echo MRI presentations at OHBM. That link includes the tedana poster with a recorded demonstration of the new dynamic reports and you can also interact with those reports. Additionally, several tedana contributors organized a symposium “Two is better than one (and many are better): Multi-echo fMRI methods and applications” There was also an Open Science Room discussion on multi-echo fMRI. Those are both great places to gain a better understanding of the current state of multi-echo fMRI as well as plans relating to tedana.

 

Breaking changes to tedana

- The --manacc argument now accepts a space-separated list of integers instead of a comma-separated string.

New Features

- Dynamic reports! Check them out.

- Many improvements in documentation

- Added information for Philips scanner multi-echo acquisition.

 

Fixes

- T2* maps now have floor maps to avoid divide-by-zero errors

- Masks are now correctly included in the workflow

- Fixed math discrepancy between paper and implementation for PAID

- Fixed a minor PCA bug

 

Upcoming changes

- We are currently reviewing code to modularize a large portion of the code #591 [link]. This change should minimally affect how users run tedana, but will be the biggest under-the-hood change since people started to use tedana, and is critical to helping the tedana continue to improve. While several of us are examining the code, it would be very useful if more users tried the current and new versions on their data to identify what, if any output differences happen. If you are interested in trying this, please contact us soon.

 

Governance

 

We’ve transitioned from having one person who would have the final say on all decisions to a system that strives for consensus while providing a path for when consensus cannot be achieved. We built off other open source governance systems to find one that contains sufficient structure for situations when structure is needed, but keeps the administrative burden low given the fact that tedana is a fairly small project. In addition to having a steering committee and maintainers to help guide decisions, we have focused leaders who can provide support to contributors - new and existing - in may areas of the project. The new governance structure is here. Feedback is very welcome.

Reply all
Reply to author
Forward
0 new messages