Rapport (Rapport De Recherche) Année : 2015

A Wavelet-based Method for the Metabolite Detection and Quantification in Proton Magnetic Resonance Spectroscopy

Résumé

Purpose: In healthy tissue, metabolites are present in steady-state concentrations typical for that specific tissue. Metabolite concentrations may shift due to stress, functional disturbances, tumors or metabolic diseases. These changes are detectable with MRS, and provide valuable information for both diagnosis and therapeutic surveillance. Various methods have been developed to quantify the metabolite concentrations using methods ranging from the simple integration of spectral peak to complex algorithms.Theory and Methods: The purpose of this work is to develop an analysis system for in vivo NMR spectroscopy for brain metabolites quantification based on continuous wavelet transforms without prior knowledge. Tests are done on both simulated and real in vivo MRS signals using spectroscopic data acquired in 16 healthy subjects (range 20-50) using 3T MR system.Results: Results show that even with the presence of low SNRs and baseline, the proposed method is able to derive the parameters such as the frequencies, the amplitudes and the damping factors of metabolites directly from the raw data and without any beforehand preprocessing or prior knowledge.Conclusion: The CWT analysis has shown to be accurate, robust and in agreement with the time domain fitting method AMARES for the quantification of short echo time in vivo MRS data.

Fichier principal
Vignette du fichier
article.pdf (1.24 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence
Loading...

Dates et versions

hal-02042878 , version 1 (20-02-2019)

Licence

Identifiants

  • HAL Id : hal-02042878 , version 1

Citer

Jessica Hanna, Sandy Rihana, David Helbert, Rita Zrour, Philippe Carré, et al.. A Wavelet-based Method for the Metabolite Detection and Quantification in Proton Magnetic Resonance Spectroscopy. [Research Report] Université de Poitiers (France). 2015. ⟨hal-02042878⟩
380 Consultations
296 Téléchargements

Partager

  • More