Skip to main content
Department of Information Technology

Magnetic Resonance Spectroscopy

Magnetic Resonance is an important instrument for diagnosis of various diseases and for discovering the molecular structure of chemical compounds. There exist primarily two implementations of MR: Magnetic Resonance Imaging (MRI) which allows one to obtain images of the soft tissues in the body, and Magnetic Resonance Spectroscopy (MRS) which is a spectroscopic technique that can be used to determine the concentrations of different substances in a sample. Both MRI and MRS are useful tools for medical applications and, among other things, they can be used to find cancer tumors. MRS is also used in chemistry to analyze complex molecular structures.

The most common signal processing method in clinical use today for MR spectroscopy is the Fourier transform (FT). While the FT is a robust method which is also able to process lots of data in a short time, it suffers from a lower resolution than many modern spectral estimators. By, e.g., introducing a mathematical model of the MR signal and estimating the parameters of that model, we can obtain much higher resolution than that of the FT. Methods exploiting such a model are often called parametric methods.

Our recent MRS research include the following topics

  • Fast and Accurate Estimation of the T1 parameter in MRI (article in preparation)
  • Separation of Fat and Water in MRI (article in preparation)
  • Frequency-Selective Estimation of MRS Signal Parameters [1], [3]
  • Multichannel Spectral Analysis of MRS Data [4]
  • Estimation of MRS Signal Parameters using Prior Information [2]
  • Area-Selective Spectral Analysis of Two-Dimensional MRS Data [5]

Selected publications

  1. Advanced Spectral Analysis with Applications. Niclas Sandgren. Ph.D. thesis, , Institutionen för informationsteknologi, Uppsala, 2007.
  2. Frequency-Selective Magnetic Resonance Spectroscopy using Prior Information for Data with Low Signal to Noise Ratio. Niclas Sandgren and Peter Stoica. In Proceedings of the 18th international BIOSIGNAL conference, 2006.
  3. Signal Parameter Estimation using Multichannel Magnetic Resonance Spectroscopy Analysis of In-Vivo 1H Data. Niclas Sandgren, Frederick Frigo, and Peter Stoica. In Proceedings of the 18th international BIOSIGNAL conference, 2006.
  4. Area-Selective Signal Parameter Estimation for Two-Dimensional MR Spectroscopy Data. Niclas Sandgren, Peter Stoica, and Frederick Frigo. In Journal of magnetic resonance, volume 183, number 1, pp 50-59, 2006. (DOI).
  5. Spectral analysis of multichannel MRS data. Niclas Sandgren, Peter Stoica, Frigo Frederick J., and Yngve Selén. In Journal of Magnetic Resonance, volume 175, number 1, pp 79-91, 2005. (External link).
  6. Frequency-Selective Analysis of Multichannel Magnetic Resonance Spectroscopy Data. Niclas Sandgren and Peter Stoica. In Proceedings of the 27th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2005.
  7. Subspace-based MRS data quatitation of multiplets using prior knowledge. T. Laudadio, Y. Selén, L. Vanhamme, Peter Stoica, P. Van Hecke, and S. Van Huffel. In 12th meeting of the International Society for Magnetic Resonance in Medicine, 2004.
  8. Frequency-selective SVD-based magnetic resonance spectroscopy with prior knowledge. Niclas Sandgren, Peter Stoica, and Yngve Selén. In Conference Record of the 38th Asilomar Conference on Signals, Systems, and Computers, 2004.
  9. Parametric methods for frequency-selective MR spectroscopy - a review. Niclas Sandgren, Yngve Selén, Peter Stoica, and Jian Li. In Journal of Magnetic Resonance, volume 168, number 2, pp 259-272, 2004.
  10. Subspace-based MRS Data Quantitation of Multiplets using Prior Knowledge. Teresa Laudadio, Yngve Selén, Leentje Vanhamme, Peter Stoica, and Paul Van Hecke. In Journal of Magnetic Resonance, volume 168, number 1, pp 53-65, 2004.

Updated  2009-09-07 10:45:31 by Erik Gudmundson.