Pattern recognition analysis of 1H NMR spectra from perchloric acid extracts of human brain tumor biopsies

Ross J. Maxwell, Irene Martínez-Pérez, Sebastián Cerdán, Miquel E. Cabañas, Carles Arús, Àngel Moreno, Antoni Capdevila, Enrique Ferrer, Frederic Bartomeus, Alberto Aparicio, Gerard Conesa, José María Roda, Fernando Carceller, José María Pascual, Siân L. Howells, Roy Mazucco, John R. Griffiths

Research output: Contribution to journalArticleResearchpeer-review

66 Citations (Scopus)

Abstract

Pattern recognition techniques (factor analysis and neural networks) were used to investigate and classify human brain tumors based on the 1H NMR spectra of chemically extracted biopsies (n = 118). After removing information from lactate (because of variable ischemia times), unsupervised learning suggested that the spectra separated naturally into two groups: meningiomas and other tumors. Principal component analysis reduced the dimensionality of the data. A back-propagation neural network using the first 30 principal components gave 85% correct classification of meningiomas and nonmeningiomas. Simplification by vector rotation gave vectors that could be assigned to various metabolites, making it possible to use or to reject their information for neural network classification. Using scores calculated from the four rotated vectors due to creatine and glutamine gave the best classification into meningiomas and nonmeningiomas (89% correct). Classification of gliomas (n = 47) gave 62% correct within one grade. Only inositol showed a significant correlation with glioma grade.
Original languageEnglish
Pages (from-to)869-877
JournalMagnetic Resonance in Medicine
Volume39
DOIs
Publication statusPublished - 1 Jan 1998

Keywords

  • Brain
  • Cancer
  • Neural network
  • Principal component analysis

Fingerprint Dive into the research topics of 'Pattern recognition analysis of <sup>1</sup>H NMR spectra from perchloric acid extracts of human brain tumor biopsies'. Together they form a unique fingerprint.

Cite this