EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Correction of orientation dependence in magnetization transfer measures in the context of tractometry: Challenges, pitfalls and solutions

Philippe Karan, Manon Edde, Guillaume Gilbert, Muhamed Barakovic, Stefano Magon, Maxime Descoteaux

Magnetic Resonance Imaging · 2025

DOI: 10.1016/j.mri.2025.110522 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Magnetic Resonance Imaging, 2025. Verfasst von Philippe Karan und 5 weiteren Autor:innen. Beteiligte Einrichtung: Sherbrooke Connectivity Imaging Laboratory (SCIL), Université de Sherbrooke, Sherbrooke, Quebec, Canada. Electronic address: philippe.karan@usherbroo…. Hauptthemen laut MEDLINE: White Matter, Brain, Diffusion Tensor Imaging, Image Enhancement. Die Arbeit nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Nature et technologies, Université de Sherbrooke.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this work we investigate the feasibility of a correction method for removing the orientation dependence of magnetization transfer (MT) measures in the context of tractometry. Following previous work on the track-based characterization of such orientation dependence using diffusion MRI, a correction method was developed. It uses polynomial fits to extrapolate the single-fiber characterizations and allows the MT measures across all white matter tracks to be shifted towards a chosen reference value, effectively removing the bias of fiber orientation with respect to the main magnetic field. Three different references were tested on a dataset of one hundred acquisitions and the performance was accessed by evaluating the removal of the orientation dependence and the reduction of variance between acquisitions, while also exploring the effects on tractometry results. Throughout these experiments, various challenges and pitfalls of an empirical correction method were laid out, like the absence of ground truth or the lack of knowledge about the complex behavior of the phenomenon in crossing-fiber voxels. Nonetheless, a solution was presented, paving the way towards a fully validated correction method for MT measures.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

White MatterBrainDiffusion Tensor ImagingImage EnhancementImage Interpretation, Computer-AssistedArtifactsHumansAlgorithmsReproducibility of ResultsSensitivity and Specificity

Stichwörter der Autor:innen

Diffusion MRIMagnetization transferOrientation dependenceTracksTractometryWhite matter

Autor:innen und Institutionen

  • Sherbrooke Connectivity Imaging Laboratory (SCIL), Université de Sherbrooke, Sherbrooke, Quebec, Canada. Electronic address: philippe.karan@usherbrooke.ca.
  • Sherbrooke Connectivity Imaging Laboratory (SCIL), Université de Sherbrooke, Sherbrooke, Quebec, Canada.
  • MR Clinical Science, Philips Healthcare Canada, Mississauga, Ontario, Canada.
  • Pharma Research and Early Development, Neuroscience and Rare Diseases Roche Innovation Center Basel, Basel, Switzerland.

Förderung

  • Nature et technologies
  • Université de Sherbrooke

Zitieren

APA

Karan, P., Edde, M., Gilbert, G., Barakovic, M., Magon, S., & Descoteaux, M. (2025). Correction of orientation dependence in magnetization transfer measures in the context of tractometry: Challenges, pitfalls and solutions. Magnetic Resonance Imaging. https://doi.org/10.1016/j.mri.2025.110522

Vancouver

Karan P, Edde M, Gilbert G, Barakovic M, Magon S, Descoteaux M. Correction of orientation dependence in magnetization transfer measures in the context of tractometry: Challenges, pitfalls and solutions. Magnetic Resonance Imaging. 2025. doi:10.1016/j.mri.2025.110522

BibTeX
@article{Karan2025,
  author = {Karan, Philippe and Edde, Manon and Gilbert, Guillaume and Barakovic, Muhamed and Magon, Stefano and Descoteaux, Maxime},
  title = {Correction of orientation dependence in magnetization transfer measures in the context of tractometry: Challenges, pitfalls and solutions},
  journal = {Magnetic Resonance Imaging},
  year = {2025},
  doi = {10.1016/j.mri.2025.110522}
}
Kostenlos starten →