TY - JOUR
T1 - Feature selection and classification of high-resolution NMR spectra in the complex wavelet transform domain
AU - Kim, Seoung Bum
AU - Wang, Zhou
AU - Oraintara, Soontorn
AU - Temiyasathit, Chivalai
AU - Wongsawat, Yodchanan
N1 - Funding Information:
We thank the referee for the constructive comments and suggestions, which greatly improved the quality of the paper. We are grateful to Dean P. Jones and Thomas R. Zeigler in the Emory University Medical School for their useful comments. We also thank the nursing and laboratory staff of the Emory General Clinical Research Center for their valuable helps in collecting samples. This work was supported in part by NSF Grant ECS-0528964.
PY - 2008/2/15
Y1 - 2008/2/15
N2 - Successful identification of the important metabolite features in high-resolution nuclear magnetic resonance (NMR) spectra is a crucial task for the discovery of biomarkers that have the potential for early diagnosis of disease and subsequent monitoring of its progression. Although a number of traditional features extraction/selection methods are available, most of them have been conducted in the original frequency domain and disregarded the fact that an NMR spectrum comprises a number of local bumps and peaks with different scales. In the present study a complex wavelet transform that can handle multiscale information efficiently and has an energy shift-insensitive property is proposed as a method to improve feature extraction and classification in NMR spectra. Furthermore, a multiple testing procedure based on a false discovery rate (FDR) was used to identify important metabolite features in the complex wavelet domain. Experimental results with real NMR spectra showed that classification models constructed with the complex wavelet coefficients selected by the FDR-based procedure yield lower rates of misclassification than models constructed with original features and conventional wavelet coefficients.
AB - Successful identification of the important metabolite features in high-resolution nuclear magnetic resonance (NMR) spectra is a crucial task for the discovery of biomarkers that have the potential for early diagnosis of disease and subsequent monitoring of its progression. Although a number of traditional features extraction/selection methods are available, most of them have been conducted in the original frequency domain and disregarded the fact that an NMR spectrum comprises a number of local bumps and peaks with different scales. In the present study a complex wavelet transform that can handle multiscale information efficiently and has an energy shift-insensitive property is proposed as a method to improve feature extraction and classification in NMR spectra. Furthermore, a multiple testing procedure based on a false discovery rate (FDR) was used to identify important metabolite features in the complex wavelet domain. Experimental results with real NMR spectra showed that classification models constructed with the complex wavelet coefficients selected by the FDR-based procedure yield lower rates of misclassification than models constructed with original features and conventional wavelet coefficients.
KW - Classification tree
KW - Complex wavelet transforms
KW - False discovery rates
KW - Gabor coefficients
KW - High-resolution NMR spectra
KW - Metabolomics
UR - http://www.scopus.com/inward/record.url?scp=38349178369&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=38349178369&partnerID=8YFLogxK
U2 - 10.1016/j.chemolab.2007.09.005
DO - 10.1016/j.chemolab.2007.09.005
M3 - Article
AN - SCOPUS:38349178369
SN - 0169-7439
VL - 90
SP - 161
EP - 168
JO - Chemometrics and Intelligent Laboratory Systems
JF - Chemometrics and Intelligent Laboratory Systems
IS - 2
ER -