SlicerCART

SlicerCART : Configurable Annotation and Review Tool.

Description

3D Slicer extension
Adapted from code of Dr. Laurent Létourneau-Guillon and his team in ICH_SEGMENTER_V2, SlicerCART, and Brain_Extraction. This is an effort to create a unified code for a configurable 3D Slicer extension.
Inspired from Neuropoly's workflow.
This tool is made to improve manual segmentation workflows across different teams.

Keywords: medical imaging, manual segmentation, manual correction, workflow, ground-truth segmentation, quality control

Abbreviations:

MRI - Magnetic Resonance Imaging
CT - Computed Tomography
BIDS - Brain Imaging Data Structure
GUI - Graphical User Interface
QC - Quality Control

This module has been adapted to perform several tasks. Among other things, it allows the user to:

Customize configuration preferences (see below for details)
Customize keyboard shortcuts
Identify the name, degree and revision step related to the human annotator
Select folder of interest where volumes are saved (and possibly impose BIDS)
Select the output folder where processing and work is preferred to be saved
Select a ground-truth folder where references studies can be used for iterative self-assessment
Display in the GUI a case list of all the studies of interests for the segmentation task (*from a site directory or a customized list)
Select from the GUI case list any volume of interest to display
Navigate through case list from next and previous buttons
Load automatically the first remaining case for segmentation in a customized list
Create automatically all required segments that may be used according to the project configuration each time a volume is displayed
Toggle interpolation of the volume loaded
Execute multiple automated functions when saving segmentation masks for a given volume. Indeed, the automated functions:
- Save segmentation masks in the selected output folder with volume file hierarchy
- Track the different versions (save the following version if previous version(s) already exist(s)) *N.B. limitation to 99 versions for a single volume
- Save a .csv file with segmentation statistics (e.g. subject, annotator's name and degree, revision step, date and time, total duration, duration of each label annotation)
- Save a .csv file with classification statistics (e.g. subject, annotator's name and degree, revision step, date and time, checkboxes / dropdown / free text fields)
- Go to the next remaining case and make it ready to segment without any further action
Load a pre-existing segmentation for further modification (will be saved as a new version of the segmentation)

Requirements

MacOS Sonoma (recommended)
A working version of 3D Slicer.
- N.B. The version used to develop this module is the version 5.2.2 since more recent versions were not able to support some extensions that previously worked (e.g. JupySlicer)

This module has been developed on:

MacOS Sonoma version 14.1.1
3D Slicer version 5.2.2

Although it may work on other versions and/or operating system, please note that it has not been tested.

Installation steps

Install 3D Slicer.
Clone this repository.
Modify label_config.yml to describe your annotations. There can be as many or as few as you would like. The colors are configurable using RGB integer values between 0 and 255. The default HU thresholds for each label are also configurable. These can also be modified directly in the extension. Note that additional tools will appear in the user interface if one of the labels is either intracerebral hemorrhage (ICH) or perihematomal edema (PHE).
Open 3D Slicer.
Activate the checkbox Enable developer mode in Edit -> Application Settings -> Developer -> Enable developer mode.
Add the path of the folder containing the SlicerCART.py file in Edit -> Application Settings -> Modules -> Additional module paths.
- There might be errors. These would be seen in the Python Console.
The module can be found under Examples -> SlicerCART.
(Optional) Set the SlicerCART module to launch at 3DSlicer startup. To do so, go to Edit -> Application Settings -> Modules -> Default startup module

Trouble shooting

Qt might need to be installed. The first five steps of the following procedure might be useful for this: procedure.

Other extensions that could be useful

SlicerJupyter to be able to use Jupyter Notebooks connected to 3D Slicer.

Documentation

TODO (after sufficient development has been made)

SlicerCART Demo Video (June 24th 2024)
Example of workflow that could be implemented (documentation here) with videos examples (12min30: summary; 40min: detailed).

Video tutorials