The Open Science Initiative for Perfusion Imaging (OSIPI): Results from the ASL MRI Challenge

Udunna Anazodo, Joana Pinto, Flora McConnell, Cassandra van Praag, Henk Mutsaerts, Aaron Taylor, Jan Petr, Diego Pineda-Ordóñez, Maria‐Eleni Dounavi, Irène Brumer, Wei Siang Marcus Chan, Jack Toner, Jian Hu, Logan Zhang, Laura C. Bell, Joseph G. Woods, Moss Zhao, Paula Croal, André Monteiro PaschoalProceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition 2023 · DOI

Abstract

The OSIPI ASL MRI Challenge is an initiative of the ASL community aiming to characterize the variability of CBF quantification arising from different pipelines. The goal of this challenge is to establish best practice in ASL data processing, understand the sources of variability, make ASL analysis more reproducible, and enable fair comparison between studies. Here, we analyzed 3 submitted entries from 7 teams registered in the challenge. The preliminary results showed pipelines based in different programming languages and analysis tools, leading to important variability in the quantitative CBF maps compared to the ground-truth.
BibTeX
@inproceedings{anazodo2023open,
  author = {Udunna Anazodo and Joana Pinto and Flora McConnell and Cassandra van Praag and Henk Mutsaerts and Aaron Taylor and Jan Petr and Diego Pineda-Ordóñez and Maria‐Eleni Dounavi and Irène Brumer and Wei Siang Marcus Chan and Jack Toner and Jian Hu and Logan Zhang and Laura C. Bell and Joseph G. Woods and Moss Zhao and Paula Croal and André Monteiro Paschoal},
  title = {{The Open Science Initiative for Perfusion Imaging (OSIPI): Results from the ASL MRI Challenge}},
  booktitle = {Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition},
  year = {2023},
  doi = {10.58530/2022/0910},
}

Loading related content...

Figure