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

doi:https://doi.org/10.1002/mrm.1910390604

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

Department of Biochemistry and Molecular Biology

Research output: Contribution to journal › Article › Research › peer-review

70 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 language	English
Pages (from-to)	869-877
Journal	Magnetic Resonance in Medicine
Volume	39
DOIs	https://doi.org/10.1002/mrm.1910390604
Publication status	Published - 1 Jan 1998

Keywords

Brain
Cancer
Neural network
Principal component analysis

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

https://doi.org/10.1002/mrm.1910390604

Cite this

Maxwell, R. J., Martínez-Pérez, I., Cerdán, S., Cabañas, M. E., Arús, C., Moreno, À., Capdevila, A., Ferrer, E., Bartomeus, F., Aparicio, A., Conesa, G., Roda, J. M., Carceller, F., Pascual, J. M., Howells, S. L., Mazucco, R., & Griffiths, J. R. (1998). Pattern recognition analysis of ¹H NMR spectra from perchloric acid extracts of human brain tumor biopsies. Magnetic Resonance in Medicine, 39, 869-877. https://doi.org/10.1002/mrm.1910390604

@article{db847a455ca84502b56987ebe6dd6e2f,

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

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.",

keywords = "Brain, Cancer, Neural network, Principal component analysis",

author = "Maxwell, {Ross J.} and Irene Mart{\'i}nez-P{\'e}rez and Sebasti{\'a}n Cerd{\'a}n and Caba{\~n}as, {Miquel E.} and Carles Ar{\'u}s and {\`A}ngel Moreno and Antoni Capdevila and Enrique Ferrer and Frederic Bartomeus and Alberto Aparicio and Gerard Conesa and Roda, {Jos{\'e} Mar{\'i}a} and Fernando Carceller and Pascual, {Jos{\'e} Mar{\'i}a} and Howells, {Si{\^a}n L.} and Roy Mazucco and Griffiths, {John R.}",

year = "1998",

month = jan,

day = "1",

doi = "https://doi.org/10.1002/mrm.1910390604",

language = "English",

volume = "39",

pages = "869--877",

journal = "Magnetic Resonance in Medicine",

issn = "0740-3194",

}

Maxwell, RJ, Martínez-Pérez, I, Cerdán, S, Cabañas, ME, Arús, C, Moreno, À, Capdevila, A, Ferrer, E, Bartomeus, F, Aparicio, A, Conesa, G, Roda, JM, Carceller, F, Pascual, JM, Howells, SL, Mazucco, R & Griffiths, JR 1998, 'Pattern recognition analysis of ¹H NMR spectra from perchloric acid extracts of human brain tumor biopsies', Magnetic Resonance in Medicine, vol. 39, pp. 869-877. https://doi.org/10.1002/mrm.1910390604

TY - JOUR

T1 - Pattern recognition analysis of 1H NMR spectra from perchloric acid extracts of human brain tumor biopsies

AU - Maxwell, Ross J.

AU - Martínez-Pérez, Irene

AU - Cerdán, Sebastián

AU - Cabañas, Miquel E.

AU - Arús, Carles

AU - Moreno, Àngel

AU - Capdevila, Antoni

AU - Ferrer, Enrique

AU - Bartomeus, Frederic

AU - Aparicio, Alberto

AU - Conesa, Gerard

AU - Roda, José María

AU - Carceller, Fernando

AU - Pascual, José María

AU - Howells, Siân L.

AU - Mazucco, Roy

AU - Griffiths, John R.

PY - 1998/1/1

Y1 - 1998/1/1

N2 - 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.

AB - 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.

KW - Brain

KW - Cancer

KW - Neural network

KW - Principal component analysis

U2 - https://doi.org/10.1002/mrm.1910390604

DO - https://doi.org/10.1002/mrm.1910390604

M3 - Article

SN - 0740-3194

VL - 39

SP - 869

EP - 877

JO - Magnetic Resonance in Medicine

JF - Magnetic Resonance in Medicine

ER -