EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Model‐informed unsupervised deep learning approaches to frequency and phase correction of MRS signals

Amirmohammad Shamaei iD, Jana Starcukova iD, Iveta Pavlova iD, Zenon Starcuk iD

Magnetic Resonance in Medicine · 2023-03 · 89 (3) · 1221-1236

DOI: 10.1002/mrm.29498 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Magnetic Resonance in Medicine, online erschienen am 11.11.2022. Verfasst von Amirmohammad Shamaei und 3 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Scientific Instruments of the Czech Academy of Sciences, Brno, Czech Republic. Hauptthemen laut MEDLINE: Deep Learning. Die Arbeit wurde bisher 26-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Akademie Věd České Republiky.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

A supervised deep learning (DL) approach for frequency and phase correction (FPC) of MRS data recently showed encouraging results, but obtaining transients with labels for supervised learning is challenging. This work investigates the feasibility and efficiency of unsupervised deep learning–based FPC.

Methods

Two novel deep learning–based FPC methods (deep learning–based Cr referencing and deep learning–based spectral registration), which use a priori physics domain knowledge, are presented. The proposed networks were trained, validated, and evaluated using simulated, phantom, and publicly accessible in vivo MEGA‐edited MRS data. The performance of our proposed FPC methods was compared with other generally used FPC methods, in terms of precision and time efficiency. A new measure was proposed in this study to evaluate the FPC method performance. The ability of each of our methods to carry out FPC at varying SNR levels was evaluated. A Monte Carlo study was carried out to investigate the performance of our proposed methods.

Results

The validation using low‐SNR manipulated simulated data demonstrated that the proposed methods could perform FPC comparably with other methods. The evaluation showed that the deep learning–based spectral registration over a limited frequency range method achieved the highest performance in phantom data. The applicability of the proposed method for FPC of GABA‐edited in vivo MRS data was demonstrated. Our proposed networks have the potential to reduce computation time significantly.

Conclusions

The proposed physics‐informed deep neural networks trained in an unsupervised manner with complex data can offer efficient FPC of large MRS data in a shorter time.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningNeural Networks, ComputerPhantoms, ImagingMonte Carlo MethodImage Processing, Computer-Assisted

Stichwörter der Autor:innen

MR spectroscopydeep learningedited MRSfrequency correctionphase correction

Autor:innen und Institutionen

  • Institute of Scientific Instruments of the Czech Academy of Sciences, Brno, Czech Republic.
  • Department of Biomedical Engineering, Brno University of Technology, Brno, Czech Republic.

ORCID-Profile: Amirmohammad Shamaei, Jana Starcukova, Iveta Pavlova, Zenon Starcuk

Förderung

  • Akademie Věd České Republiky

Zitieren

APA

Shamaei, A., Starcukova, J., Pavlova, I., & Starcuk, Z. (2023). Model‐informed unsupervised deep learning approaches to frequency and phase correction of MRS signals. Magnetic Resonance in Medicine, 89(3), 1221-1236. https://doi.org/10.1002/mrm.29498

Vancouver

Shamaei A, Starcukova J, Pavlova I, Starcuk Z. Model‐informed unsupervised deep learning approaches to frequency and phase correction of MRS signals. Magnetic Resonance in Medicine. 2023;89(3):1221-1236. doi:10.1002/mrm.29498

BibTeX
@article{Shamaei2023,
  author = {Shamaei, Amirmohammad and Starcukova, Jana and Pavlova, Iveta and Starcuk, Zenon},
  title = {Model‐informed unsupervised deep learning approaches to frequency and phase correction of MRS signals},
  journal = {Magnetic Resonance in Medicine},
  year = {2023},
  volume = {89},
  number = {3},
  pages = {1221--1236},
  doi = {10.1002/mrm.29498}
}
Kostenlos starten →