Skip to content

mervinTEE/ExploreASL-Tools

Repository files navigation

ExploreASL-Tools

Here are some scripts that helps in manage some pre- and post-processing from ExploreASL. Feel free to use any of the scripts to help in your workflow!

check_processing_status.sh

ExploreASL will include a "999_ready.status" found in the lock folder. This script will search for this status and report if any of the subject is without this status.

Current limitations

  • If subject did not begin processing in the specific module, it will fail to detect it. So, do counter check if your the subj list matches your rawdata.

Example usage: ./check_processing_status.sh /path/to/module_locked_folder

find_missing_rawdata.sh

This script will help report the subj with the missing files you are looking for. For example, if you are looking for subjects with missing "*run-1*". This will loop through all your folders and look for the specific string indicated. You can repeat this for either the 'anat' or 'perf' folder by supplying the argument to the script. Example usage: ./find_missing_rawdata.sh /path/to/rawdata anat "*filename_pattern*"

moving_dicoms_to_subj_root.sh

ExploreASL import relied on sourcestructure.json to detect DICOMs for conversion. As the folder hierarchy is fixed based on what was defined in sourcestructure.json, this can fail to detect subjects with different folder depths. This script will help you move your defined folders to the subj root folder. Example: ./moving_dicoms_to_subj_root.sh /path/to/sourcedatafolder scans

QC.R and QC_TwoPanel.R

This runs in R Shiny. To facilitate QC process you can provide the path to the jpegs generated by ExploreASL and conduct your QC on the Shiny's interface. This script is made to allow QC to be conducted using just your keyboards keys (1- CBF, 2- Vascular, 3- Artifact, 4- Unknown). You can also further revisit your previous image to change your gradings by using the left and right keys. Once you are ready to export, you can click export to csv and it will generate a csv file in the same folder as your images. Windows can be resized according to the size you prefer.

Recommended for two panel function, first group all your Coronal and Transverse images into separate folders. Then use these two paths as the directory paths to load the images. The csv will always be saved to the first directory loaded.

Steps to Get Started

  1. Load first directory
  2. Load second directory (optional)
  3. Check if both images matches (optional, if you did not load second panel)
  4. Load csv file (optional, if you do not have previously graded csv saved)
  5. Start grading, either using mouse or keys.

New Features Added

  • Load of previous grading from csv file with past gradings (limitations: needs to ensure that it is the right csv file)
  • Added prompts when you are trying to load the path to avoid accidental refresh.
  • Added field to name your csv export file and prompts to warn about overwriting existing files.
  • Added additional version (QC_TwoPanel.R) which allows you to scroll through two views concurrently. Pre load your images into two folders and provide the paths accordingly. Ensure that each folder has the same subjects' images. CSV file path will always be taking reference from first directory Loaded
  • Added counter to inform the number of images left to grade
  • Added comment field to provide additional information about the image when needed.

Alt Text

Current limitations

  • Yet to implement a dropdown for you to select the specific image without scrolling through all subjs, but it doesn't take too long to scroll using the keys at this moment.
  • Two-panel will not know if images mismatch between sessions or subjects, so make sure to check your images before your screening.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published