EUVIMEDEUVIMED entdecken
VergleichsstudieOpen Access · CC BY 4.0

WALINET: A water and lipid identification convolutional neural network for nuisance signal removal in 1H MR spectroscopic imaging

Paul J Weiser iD, Georg Langs, Stanislav Motyka iD, Wolfgang Bogner, Sébastien Courvoisier iD, Malte Hoffmann, Antoine Klauser iD, Ovidiu C Andronesi iD

Magnetic Resonance in Medicine · 2025

DOI: 10.1002/mrm.30402 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift Magnetic Resonance in Medicine, 2025. Verfasst von Paul J Weiser und 7 weiteren Autor:innen. Beteiligte Einrichtung: Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Boston, Massachusetts, USA. Hauptthemen laut MEDLINE: Brain, Convolutional Neural Networks, Image Processing, Computer-Assisted, Lipids. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCI NIH HHS, NICHD NIH HHS, The Christian Doppler Laboratory for MR Imaging Biomarkers (BIOMAK).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

Proton magnetic resonance spectroscopic imaging (‐MRSI) provides noninvasive spectral‐spatial mapping of metabolism. However, long‐standing problems in whole‐brain ‐MRSI are spectral overlap of metabolite peaks with large lipid signal from scalp, and overwhelming water signal that distorts spectra. Fast and effective methods are needed for high‐resolution ‐MRSI to accurately remove lipid and water signals while preserving the metabolite signal. The potential of supervised neural networks for this task remains unexplored, despite their success for other MRSI processing.

Methods

We introduce a deep learning method based on a modified Y‐NET network for water and lipid removal in whole‐brain ‐MRSI. The WALINET (WAter and LIpid neural NETwork) was compared with conventional methods such as the state‐of‐the‐art lipid L2 regularization and Hankel–Lanczos singular value decomposition (HLSVD) water suppression. Methods were evaluated on simulated models and in vivo whole‐brain MRSI using NMRSE, SNR, CRLB, and FWHM metrics.

Results

WALINET is significantly faster and needs 8s for high‐resolution whole‐brain MRSI, compared with 42min for conventional HLSVD+L2. WALINET suppresses lipid and water in the brain by 25–45 and 34–53‐fold, respectively. WALINET has better performance than HLSVD+L2, providing: (1) more lipid removal with 41% lower NRMSE; (2) better metabolite signal preservation with 71% lower NRMSE in simulated data; 155% higher SNR and 50% lower CRLB in in vivo data. Metabolic maps obtained by WALINET in healthy subjects and patients show better gray‐/white‐matter contrast with more visible structural details.

Conclusions

WALINET has superior performance for nuisance signal removal and metabolite quantification on whole‐brain ‐MRSI compared with conventional state‐of‐the‐art techniques. This represents a new application of deep learning for MRSI processing, with potential for automated high‐throughput workflow.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

BrainConvolutional Neural NetworksImage Processing, Computer-AssistedLipidsNeural Networks, ComputerProton Magnetic Resonance SpectroscopyWaterAdultFemaleHumansMaleAlgorithmsDeep LearningMagnetic Resonance ImagingMagnetic Resonance Spectroscopy

Stichwörter der Autor:innen

brainmetabolite quantificationmr spectroscopic imagingultrahigh‐field mrwater and lipid removal

Substanzen

LipidsWater

Autor:innen und Institutionen

  • Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Boston, Massachusetts, USA.
  • Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA.
  • Computational Imaging Research Lab-Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria.
  • High Field MR Center-Department of Biomedical Imaging and Image-Guided Therapy, Medical University of Vienna, Vienna, Austria.
  • Center for Biomedical Imaging (CIBM), Geneva, Switzerland.
  • Department of Radiology and Medical Informatics, University of Geneva, Geneva, Switzerland.

ORCID-Profile: Paul J Weiser, Stanislav Motyka, Sébastien Courvoisier, Antoine Klauser, Ovidiu C Andronesi

Förderung

  • NCI NIH HHS
  • NICHD NIH HHS
  • The Christian Doppler Laboratory for MR Imaging Biomarkers (BIOMAK)
  • NIBIB NIH HHS
  • Austrian Science Fund

Zitieren

APA

Weiser, P. J., Langs, G., Motyka, S., Bogner, W., Courvoisier, S., Hoffmann, M., Klauser, A., & Andronesi, O. C. (2025). WALINET: A water and lipid identification convolutional neural network for nuisance signal removal in 1H MR spectroscopic imaging. Magnetic Resonance in Medicine. https://doi.org/10.1002/mrm.30402

Vancouver

Weiser PJ, Langs G, Motyka S, Bogner W, Courvoisier S, Hoffmann M, et al. WALINET: A water and lipid identification convolutional neural network for nuisance signal removal in 1H MR spectroscopic imaging. Magnetic Resonance in Medicine. 2025. doi:10.1002/mrm.30402

BibTeX
@article{Weiser2025,
  author = {Weiser, Paul J and Langs, Georg and Motyka, Stanislav and Bogner, Wolfgang and Courvoisier, Sébastien and Hoffmann, Malte and Klauser, Antoine and Andronesi, Ovidiu C},
  title = {WALINET: A water and lipid identification convolutional neural network for nuisance signal removal in 1H MR spectroscopic imaging},
  journal = {Magnetic Resonance in Medicine},
  year = {2025},
  doi = {10.1002/mrm.30402}
}
Kostenlos starten →