Separation of Metabolite and Macromolecule Signals for 1 H-Mrsi Using Learned Nonlinear Models

Yahang Li, Zepeng Wang, Fan Lam

Research output: Chapter in Book/Report/Conference proceedingConference contribution


This paper presents a novel method to reconstruct and separate metabolite and macromolecule (MM) signals in 1 H magnetic resonance spectroscopic imaging (MRS I) data using learned nonlinear models. Specifically, deep autoencoder (DAE) networks were constructed and trained to learn the nonlinear low-dimensional manifolds, where the metabolite and MM signals reside individually. A regularized reconstruction formulation is proposed to integrate the learned models with signal encoding model to reconstruct and separate the metabolite and MM components. An efficient algorithm was developed to solve the associated optimization problem. The performance of the proposed method has been evaluated using simulation and experimental 1 H-MRSI data. Efficient low-dimensional signal representation of the learned models and improved metabolite/MM separation over the standard parametric fitting based approach have been demonstrated.

Original languageEnglish (US)
Title of host publicationISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging
PublisherIEEE Computer Society
Number of pages4
ISBN (Electronic)9781538693308
StatePublished - Apr 2020
Event17th IEEE International Symposium on Biomedical Imaging, ISBI 2020 - Iowa City, United States
Duration: Apr 3 2020Apr 7 2020

Publication series

NameProceedings - International Symposium on Biomedical Imaging
ISSN (Print)1945-7928
ISSN (Electronic)1945-8452


Conference17th IEEE International Symposium on Biomedical Imaging, ISBI 2020
CountryUnited States
CityIowa City


  • ADMM
  • deep autoencoder
  • proton magnetic resonance spectroscopic imaging' deep learning
  • regularization

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Separation of Metabolite and Macromolecule Signals for <sup>1</sup> H-Mrsi Using Learned Nonlinear Models'. Together they form a unique fingerprint.

Cite this