Skip to content

v1.0.0 ClinicalASL Tool release

Choose a tag to compare

View all tags
@JSIERO JSIERO released this 24 Jun 08:31
· 84 commits to main since this release
v1.0.0
ca240d2

Release Date: 2025-06-11

Version: v1.0.0

Official ClinicalASL Tool for processing of quantitative multi-delay ASL MRI data designed for clinical use with PACS export and visual reporting on CBF, AAT and CVR

Features

  • DICOM to NIfTI conversion
  • Look-Locker correction
  • Motion correction
  • Quantitative perfusion parameter estimation:
    • Cerebral Blood Flow (CBF)
    • Arterial Arrival Time (AAT)
    • Arterial Transit Artefacts (ATA)
    • Cerebrovascular Reactivity (CVR)
  • Export of results as NIfTI, PNG, and DICOM (PACS-ready) formats.
  • Handles Philips multiframe enhanched MRI DICOMS

General improvements:

  • Switched from fixed preACZ / postACZ for acetazolamide naming → generalized BIDS-compliant ASL context:
    • baseline
    • stimulus
    • with customizable context_study_tags (e.g. preACZ, postACZ, preHypercapnia, Hypercapnia, etc.)

Advanced multi-delay ASL fitting using Quantified Imaging - SSVB:

  • Integration of the Spatially and Spatially-Varying Bayesian (SSVB) model-based fitting framework
  • Enables robust voxel-wise estimation of CBF, AAT, and CVR, especially in low-SNR clinical settings
  • Developed and maintained by Quantified Imaging
  • For methodological details, see:
    Thomas F. Kirk, Georgia G. Kenyon, Martin S. Craig, Michael A. Chappell
    Stochastic variational inference improves quantification of multiple timepoint arterial spin labelling perfusion MRI,
    Frontiers in Neuroscience, 2025.
    https://doi.org/10.3389/fnins.2025.1536752

Flexible configuration:

  • Introduced config.json:
    • All pipeline parameters are now configurable
    • Different configs per acquisition type or study possible
    • Hardcoded fallback defaults remain available

Improved logging:

  • Logs written both to console and to clinicalasl.log in output directory

Cleaner main_pipeline:

  • Unified loop over contexts → more compact and extensible
  • Reduced redundant code blocks

Improved saving:

  • PNG and DICOM outputs now use the configurable context_study_tags → matches original acquisition
  • Smarter and more consistent file naming

Utilities:

  • Added utility for loading config.json safely
  • Added version field to config.json

Known limitations

  • No external dependency manager yet → tested with manual install + requirements.txt
  • SimpleITK / Elastix / ANTs registration alternatives under evaluation

Next steps

  • Finalize and document Elastix fallback / alternative to ANTs
  • Add automatic test suite to validate pipeline outputs
  • Add support for more ASL variants (variable-TR, time-encoded, hybrid, velocity-selective ASL, etc.)
  • Start creating BIDS-derivatives-compliant output structure

How to upgrade

  • Pull latest from GitHub
  • Use new run_pipeline.py --config my_config.json option
  • Update your configs per acquisition type (see config/config_default.json as example)

Thank you for testing, contributing, and providing feedback!

Feel free to open issues / suggestions here: https://github.com/JSIERO/ClinicalASL/issues

date: 11-06-2025
Jeroen C.W. Siero, associate professor - translational neuroimaging, Utrecht Medical Center Utrecht, moyamoya team
email: j.c.w.siero@umcutrecht.nl

Full Changelog: https://github.com/JSIERO/ClinicalASL/commits/R1.0.0

Assets 2
Loading