- Feb 2024
-
docs.google.com docs.google.com
-
Future work
Actually ask people the abridged questionnaire
-
The CBCL can be reduced to 20 questions without impacting prediction by >5%.
I think this is too strong
-
-
github.com github.com
-
using the most impactful questions
but these questions were asked and answered in the context of the full battery, right?
-
- Jan 2024
-
www2.childmind.org www2.childmind.org
-
The following have been updated in the C-PAC container: Ubuntu (18.04 LTS Bionic Beaver → 22.04 LTS Jammy Jellyfish) AFNI (23.0.07 'Commodus' → 23.1.10 'Publius Helvius Pertinax') FSL (6.0.4:ddd0a010c → 6.0.6.5) A large number of C-PAC's dependencies has been upgraded to more recent versions. Please refer to the release page on our GitHub for a full list of these upgrades.
I'm not sure this being two bullet points makes sense. Maybe instead something like
• A large number of C-PAC's dependencies has been upgraded to more recent versions, including
◦ Ubuntu (18.04 'Bionic Beaver' → 22.04 'Jammy Jellyfish')
◦ Python (3.7.13 → 3.10.6; holding back from 3.11 for sdcflows, torch, and torchvision)
◦ AFNI (21.1.00 'Domitian' → 23.3.09 'Septimius Severus')
◦ ANTs (2.3.3 'Leptomyrmex' → 2.4.3 'Emplastus')
◦ FSL (5.0.9 → 6.0.6.5)
◦ Nipype (1.5.1 → 1.8.6)
Please refer to the "Upgraded dependencies" section of the changelog for 1.8.6 or the release notes in the user guide for a full list of these upgrades.
(This suggestion includes my other two suggestions for this section)
-
Improvements
Maybe add
torch, a large dependency that is used byunetand nothing else inC-PAC, has been removed from the preinstalled dependencies in the C-PAC image and is now installed in the container at runtime ifunetis enabled.
-
Fixed a bug where sparsity thresholds were not being scaled for network centrality.
Note that this bug is present in v1.8.0 ‒ v1.8.5?
-
the release page on our GitHub
I think either the "Upgraded dependencies" section of the changelog for 1.8.6 or the release notes in the user guide would be preferable documentation to the release notes on GH
-
23.1.10 'Publius Helvius Pertinax')
~~23.1.10 'Publius Helvius Pertinax'~~ 23.3.09 'Septimius Severus'
-
Freesurfer
Do we want to capitalize "FreeSurfer" like how they do?
-
fmriprep
Do we want to capitalize "fMRIPrep" like how they do?
-
- Nov 2017
-
childmind.org childmind.org
-
Parent-Child Interaction Therapy (PCIT) PCIT works with parents and children together, teaching them skills to interact in a positive, productive way. It is effective for kids between the ages of 2 and 7, and usually requires 14 to 17 weekly sessions. In PCIT, parents receive live coaching (via a bug in the ear) from a therapist who watches from behind a one-way mirror as they and their child perform a series of tasks, and parents practice specific responses to both desired and undesired behavior. PCIT is the most practice-intensive, Dr. Rouse says, as parents demonstrate mastery of each skill before going on to the next one. “It starts out with positive interactions, then waits till parents reach mastery of these skills before moving on to discipline strategies to improve oppositional behavior.” Dr. Rouse says he might propose PCIT if he feels that the parents need a lot of one-on-one attention in terms of how they’re interacting with their child, and “especially if it feels like there have been a lot of coercive negative interactions.” Sometimes it’s very important for parents, he adds, “to learn how to be with their kid in a positive way.” He adds that he doesn’t always suggest PCIT even if the child falls within PCIT’s age range. “It’s not just age, it’s looking at the situation. If there needs to be a really strong dose of positive interactions as the first thing and the child is under 7, I’ll lean more toward PCIT.”
-