Independent component analysis for automated decomposition of in vivo magnetic resonance spectra

• Published in: Magnetic resonance in medicine Vol. 50; no. 4; pp. 697 - 703
• Main Authors: Ladroue, Christophe, Howe, Franklyn A., Griffiths, John R., Tate, A. Rosemary
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc., A Wiley Company 01.10.2003; Williams & Wilkins
• Subjects: automatic decomposition; Biological and medical sciences; Brain Neoplasms - pathology; Humans; independent component analysis; independent component analysis (ICA); Magnetic Resonance Spectroscopy - methods; Medical sciences; MRS; Principal Component Analysis; principal component analysis (PCA); Radiotherapy. Instrumental treatment. Physiotherapy. Reeducation. Rehabilitation, orthophony, crenotherapy. Diet therapy and various other treatments (general aspects); Signal Processing, Computer-Assisted; Technology. Biomaterials. Equipments. Material. Instrumentation; model; independent component analysis (ICA); principal component analysis (PCA); automatic decomposition; Biomedical engineering; MRS
• ISSN: 0740-3194; 1522-2594
• DOI: 10.1002/mrm.10595

Fully automated methods for analyzing MR spectra would be of great benefit for clinical diagnosis, in particular for the extraction of relevant information from large databases for subsequent pattern recognition analysis. Independent component analysis (ICA) provides a means of decomposing signals into their constituent components. This work investigates the use of ICA for automatically extracting features from in vivo MR spectra. After its limits are assessed on artificial data, the method is applied to a set of brain tumor spectra. ICA automatically, and in an unsupervised fashion, decomposes the signals into interpretable components. Moreover, the spectral decomposition achieved by the ICA leads to the separation of some tissue types, which confirms the biochemical relevance of the components. Magn Reson Med 50:697–703, 2003. © 2003 Wiley‐Liss, Inc.