Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Second Round of ACE Fixes #152

Open
wants to merge 27 commits into
base: dev
Choose a base branch
from
Open

Second Round of ACE Fixes #152

wants to merge 27 commits into from

Conversation

anthonyprinaldi
Copy link
Member

  • This round of fixes addresses some bugs present in ACE
  • It also adds a new step and features to the workflow
  • We now have the ability to do instance segmentation (label each individual neuron), count neurons, and get neuron info based on brain regions from an atlas
  • This is added to the ACE workflow through the validate clusters module. Validate clusters will count the neurons per cluster (from ACE stats) and compare it across all subjects.

anthonyprinaldi and others added 27 commits May 21, 2024 18:55
Adds ability to number blobs in binary segmentation files. This allows for things such as counting unique blobs, filtering based on area, or computing stats.
Leave the datatype the same as the original to avoid the issue.
Changes made to processing in parallel and saving information in json file to speed up downstream processing
This is needed for using heatmap in a workflow. Otherwise, when the heatmap command is passed a None arg for `side` (if the workflow has a None arg for `side`), heatmap will fail with an error. Logic to handle this None arg exists in the heatmap script.
The arg for the treated directory was incorrect. It should match the arg for the control dir.
Variable `inclar` was not being used. It gets set manually and does not need to be an input to the function.
Update allows user to skip 0 slices if they'd like. Code would fail on 0 slices before.
`regionprops_table` not available in the version of `skimage` in `miracl`.
Arg is expected in int format (between 0 and 100) and instance seg expects it in float format (between 0.0 and 1.0).
Ace stats arg names were previously changed but were not propagated to the ace workflow.
Model outputs were not being handled correctly. Need to select foreground channel and apply softmax, otherwise threshold has no meaning (when we just compare foreground and background logits). We also need to discretize the values in a separate step than binarizing.
This commit handles all debugging and refactoring of the validate clusters script. Since the script was not in a workign condition before, this contains all debugging changes necessary to get the script working.
@anthonyprinaldi anthonyprinaldi added bug enhancement fix Includes fixes for issues labels Sep 3, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug enhancement fix Includes fixes for issues
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants