EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

WAND: Wavelet Analysis‐Based Neural Decomposition of MRS Signals for Artifact Removal

Julian P Merkofer iD, Dennis M J van de Sande iD, Sina Amirrajab, Kyung Min Nam iD, Ruud J G van Sloun, Alex A Bhogal iD

NMR in Biomedicine · 2025

DOI: 10.1002/nbm.70038 ↗

Kurz gefasst

Fachartikel in der Zeitschrift NMR in Biomedicine, 2025. Verfasst von Julian P Merkofer und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical Engineering, Eindhoven University of Technology, Eindhoven, the Netherlands. Hauptthemen laut MEDLINE: Artifacts, Wavelet Analysis, Neural Networks, Computer. Die Arbeit nennt 74 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: NWO VIDI (VI.Vidi.223.085), Spectralligence (EUREKA IA Call, ITEA4 project 20209).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate quantification of metabolites in magnetic resonance spectroscopy (MRS) is challenged by low signal‐to‐noise ratio (SNR), overlapping metabolites, and various artifacts. Particularly, unknown and unparameterized baseline effects obscure the quantification of low‐concentration metabolites, limiting MRS reliability. This paper introduces wavelet analysis‐based neural decomposition (WAND), a novel data‐driven method designed to decompose MRS signals into their constituent components: metabolite‐specific signals, baseline, and artifacts. WAND takes advantage of the enhanced separability of these components within the wavelet domain. The method employs a neural network, specifically a U‐Net architecture, trained to predict masks for wavelet coefficients obtained through the continuous wavelet transform. These masks effectively isolate desired signal components in the wavelet domain, which are then inverse‐transformed to obtain separated signals. Notably, an artifact mask is created by inverting the sum of all known signal masks, enabling WAND to capture and remove even unpredictable artifacts. The effectiveness of WAND in achieving accurate decomposition is demonstrated through numerical evaluations using simulated spectra. Furthermore, WAND's artifact removal capabilities significantly enhance the quantification accuracy of linear combination model fitting. The method's robustness is further validated using data from the 2016 MRS Fitting Challenge and in vivo experiments.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ArtifactsWavelet AnalysisNeural Networks, ComputerMagnetic Resonance SpectroscopySignal-To-Noise RatioHumansAlgorithmsReproducibility of Results

Stichwörter der Autor:innen

artifact removaldeep learningmagnetic resonance spectroscopywavelet analysis

Autor:innen und Institutionen

  • Department of Electrical Engineering, Eindhoven University of Technology, Eindhoven, the Netherlands.
  • Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, the Netherlands.
  • High Field Research Group, Center for Image Sciences, University Medical Center Utrecht, Utrecht, the Netherlands.

ORCID-Profile: Julian P Merkofer, Dennis M J van de Sande, Kyung Min Nam, Alex A Bhogal

Förderung

  • NWO VIDI (VI.Vidi.223.085)
  • Spectralligence (EUREKA IA Call, ITEA4 project 20209)

Zitieren

APA

Merkofer, J. P., van de Sande, D. M. J., Amirrajab, S., Min Nam, K., van Sloun, R. J. G., & Bhogal, A. A. (2025). WAND: Wavelet Analysis‐Based Neural Decomposition of MRS Signals for Artifact Removal. NMR in Biomedicine. https://doi.org/10.1002/nbm.70038

Vancouver

Merkofer JP, van de Sande DMJ, Amirrajab S, Min Nam K, van Sloun RJG, Bhogal AA. WAND: Wavelet Analysis‐Based Neural Decomposition of MRS Signals for Artifact Removal. NMR in Biomedicine. 2025. doi:10.1002/nbm.70038

BibTeX
@article{Merkofer2025,
  author = {Merkofer, Julian P and van de Sande, Dennis M J and Amirrajab, Sina and Min Nam, Kyung and van Sloun, Ruud J G and Bhogal, Alex A},
  title = {WAND: Wavelet Analysis‐Based Neural Decomposition of MRS Signals for Artifact Removal},
  journal = {NMR in Biomedicine},
  year = {2025},
  doi = {10.1002/nbm.70038}
}
Kostenlos starten →