MRSI spectral quantification using linear tangent space alignment (LTSA)-based manifold learning

Chao Ma, Georges El FakhriProceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition 2023 · DOI

Abstract

Spectral quantification is a critical step in quantitative MRS/MRSI. The recently proposed subspace-based spectral quantification method represents the spectral distribution of each model as a subspace model and enables effective use of spatiospectral priors to improve parametric estimation. However, modeling spectral distribution of each metabolite as a separate subspace leads to a large number of unknowns, which renders the resultant parametric estimation problem challenging when SNR is low. To address this issue, we propose a new linear tangent space alignment-based method for MRSI quantification, leveraging the intrinsic low-dimensional structure information of the underlying MRSI signals for improved parametric estimation.
BibTeX
@inproceedings{ma2023mrsi,
  author = {Chao Ma and Georges El Fakhri},
  title = {{MRSI spectral quantification using linear tangent space alignment (LTSA)-based manifold learning}},
  booktitle = {Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition},
  year = {2023},
  doi = {10.58530/2022/0243},
}

Loading related content...

Figure