EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Noise suppression of proton magnetic resonance spectroscopy improves paediatric brain tumour classification

Teddy Zhao iD, James T Grist iD, Dorothee P Auer iD, Shivaram Avula iD, Simon Bailey iD, Nigel P Davies iD, Richard G Grundy iD, Omar Khan iD, Lesley MacPherson iD, Paul S Morgan iD, Barry Pizer, Heather E L Rose iD und 5 weitere

NMR in Biomedicine · 2024

DOI: 10.1002/nbm.5129 ↗

Kurz gefasst

Fachartikel in der Zeitschrift NMR in Biomedicine, 2024. Verfasst von Teddy Zhao und 16 weiteren Autor:innen. Beteiligte Einrichtung: Cancer and Genomic Sciences, University of Birmingham, Birmingham, UK. Hauptthemen laut MEDLINE: Brain Neoplasms, Proton Magnetic Resonance Spectroscopy, Signal-To-Noise Ratio. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Action Medical Research and the Brain Tumour Charity, Cancer Research UK, NIHR Research Professorship.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Proton magnetic resonance spectroscopy ( 1 H‐MRS) is increasingly used for clinical brain tumour diagnosis, but suffers from limited spectral quality. This retrospective and comparative study aims at improving paediatric brain tumour classification by performing noise suppression on clinical 1 H‐MRS. Eighty‐three/forty‐two children with either an ependymoma (ages 4.6 5.3/9.3 5.4), a medulloblastoma (ages 6.9 3.5/6.5 4.4), or a pilocytic astrocytoma (8.0 3.6/6.3 5.0), recruited from four centres across England, were scanned with 1.5T/3T short‐echo‐time point‐resolved spectroscopy. The acquired raw 1 H‐MRS was quantified by using Totally Automatic Robust Quantitation in NMR (TARQUIN), assessed by experienced spectroscopists, and processed with adaptive wavelet noise suppression (AWNS). Metabolite concentrations were extracted as features, selected based on multiclass receiver operating characteristics, and finally used for identifying brain tumour types with supervised machine learning. The minority class was oversampled through the synthetic minority oversampling technique for comparison purposes. Post‐noise‐suppression 1 H‐MRS showed significantly elevated signal‐to‐noise ratios ( P < .05, Wilcoxon signed‐rank test), stable full width at half‐maximum ( P > .05, Wilcoxon signed‐rank test), and significantly higher classification accuracy ( P < .05, Wilcoxon signed‐rank test). Specifically, the cross‐validated overall and balanced classification accuracies can be improved from 81% to 88% overall and 76% to 86% balanced for the 1.5T cohort, whilst for the 3T cohort they can be improved from 62% to 76% overall and 46% to 56%, by applying Naïve Bayes on the oversampled 1 H‐MRS. The study shows that fitting‐based signal‐to‐noise ratios of clinical 1 H‐MRS can be significantly improved by using AWNS with insignificantly altered line width, and the post‐noise‐suppression 1 H‐MRS may have better diagnostic performance for paediatric brain tumours.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Brain NeoplasmsProton Magnetic Resonance SpectroscopySignal-To-Noise RatioHumansChildFemaleMaleChild, PreschoolAdolescentRetrospective StudiesInfant

Stichwörter der Autor:innen

machine learningmetabolite concentrationnoise suppressionpaediatric brain tumourproton magnetic resonance spectroscopywavelet

Autor:innen und Institutionen

  • Cancer and Genomic Sciences, University of Birmingham, Birmingham, UK.
  • Oncology, Birmingham Children's Hospital, Birmingham, UK.
  • Clinical Neuroscience, University of Nottingham, Nottingham, UK.
  • Radiology, Alder Hey Children's NHS Foundation Trust, Liverpool, UK.
  • Paediatric Oncology, Great North Children's Hospital, Newcastle upon Tyne, UK.
  • Imaging and Medical Physics, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK.

ORCID-Profile: Teddy Zhao, James T Grist, Dorothee P Auer, Shivaram Avula, Simon Bailey, Nigel P Davies, Richard G Grundy, Omar Khan, Lesley MacPherson, Paul S Morgan, Heather E L Rose, Yu Sun, Martin Wilson, Theodoros N Arvanitis, Andrew C Peet

Förderung

  • Action Medical Research and the Brain Tumour Charity
  • Cancer Research UK
  • NIHR Research Professorship
  • Cancer Research UK and EPSRC Cancer Imaging Programme at the Children's Cancer and Leukaemia Group (CCLG) in association with the MRC and Department of Health…
  • NIHR Nottingham Biomedical Research Centre
  • Children's Research Fund
  • Help Harry Help Others
  • Cancer Research UK and NIHR Experimental Cancer Medicine Centre Paediatric Network

Zitieren

APA

Zhao, T., Grist, J. T., Auer, D. P., Avula, S., Bailey, S., Davies, N. P., Grundy, R. G., Khan, O., MacPherson, L., Morgan, P. S., Pizer, B., Rose, H. E. L., Sun, Y., Wilson, M., Worthington, L., Arvanitis, T. N., & Peet, A. C. (2024). Noise suppression of proton magnetic resonance spectroscopy improves paediatric brain tumour classification. NMR in Biomedicine. https://doi.org/10.1002/nbm.5129

Vancouver

Zhao T, Grist JT, Auer DP, Avula S, Bailey S, Davies NP, et al. Noise suppression of proton magnetic resonance spectroscopy improves paediatric brain tumour classification. NMR in Biomedicine. 2024. doi:10.1002/nbm.5129

BibTeX
@article{Zhao2024,
  author = {Zhao, Teddy and Grist, James T and Auer, Dorothee P and Avula, Shivaram and Bailey, Simon and Davies, Nigel P and Grundy, Richard G and Khan, Omar and MacPherson, Lesley and Morgan, Paul S and Pizer, Barry and Rose, Heather E L and Sun, Yu and Wilson, Martin and Worthington, Lara and Arvanitis, Theodoros N and Peet, Andrew C},
  title = {Noise suppression of proton magnetic resonance spectroscopy improves paediatric brain tumour classification},
  journal = {NMR in Biomedicine},
  year = {2024},
  doi = {10.1002/nbm.5129}
}
Kostenlos starten →