生物医学核磁共振中的模式识别方法

摘要/Abstract

摘要：

模式识别(PR)是把具体事物进行正确归类的科学，它能解决许多对复杂体系的认识问题. 生物医学核磁共振波谱(NMR)的理解和分析便是其中一种. 在受到病理或者其他刺激后，生物体内的代谢物水平会发生变化，这种变化可以通过液体高分辨核磁共振的手段来观察. 模式识别把这种认识进一步深化，不仅可以将正常状态与病理状态区分开，还能找到是哪些生化指纹导致两种状态的差异，为生理、病理和药理等研究，以及临床诊断提供依据. 模式识别与生物核磁共振波谱的结合，已经发展成为代谢组学研究的关键技术，甚至被称为基于核磁共振的代谢组学. 主要讨论适用于生物医学核磁共振中的模式识别方法及其最新进展.

关键词: 生物医学核磁共振, 液体高分辨核磁共振, 模式识别, 多元统计, 代谢组学

Abstract:

Pattern recognition (PR) is the technology of making a decision on a concrete object which category it belongs to. PR can be used to solve many problems in understanding complex systems. One of its applications is the comprehension and analysis of biomedical magnetic resonance spectroscopy data. Previous studies using high-resolution liquid NMR spectroscopy have shown that the levels of metabolites in biological samples change with the time after pathological or other perturbation. PR can be used not only to identify the differences of the pathological from the normal, but also to find which bio-fingerprints result in such differences, thus supplying valuable information for diagnosis. In this review, various statistical PR methods used in biomedical magnetic resonance spectroscopy are discussed, and the latest progresses in this field are introduced.

Key words: biomedical magnetic resonance, high-resolution nuclear magnetic resonance (HRNMR), pattern recognition (PR), multivariate statistics, metabonomics

中图分类号:

R445.2

邱玉洁, 夏圣安, 叶朝辉, 刘买利. 生物医学核磁共振中的模式识别方法[J]. 波谱学杂志, 2005, 22(1): 99-111.

QIU Yu-Ji, Jia-Ku-An, Xie-Chao-Hui, LIU Mai-Li. Pattern Recognition Methods in Biomedical Magnetic Resonance[J]. Chinese Journal of Magnetic Resonance, 2005, 22(1): 99-111.

参考文献

[1] Nicholson J K, Wilson I D. High resolution proton magnetic resonance spectroscopy of biological fluids[J]. Prog Nucl Magn Reson Spectrosc, 1989, 21: 449-501.

[2] Kowalski B R, Bender C F. Pattern Recognition. A powerful Approach to Interpreting Chemical Data[J]. J Am Chem Soc, 1972, 94: 5 632-5 639.

[3] Kowalski B R, Bender C F. Pattern Recognition.Ⅱ.Linear and Nonlinear Methods for Displaying Chemical Data[J]. J Am Chem Soc, 1973, 95: 686-692. 

[4] Gartland K P R, Beddell C R, Lindon J C, et al. A pattern recognition approach to the comparison of PMR and clinical chemical data for classification of nephrotoxicity [J]. J Pharm Biomed Anal, 1990, 8: 963-968.

[5] Gartland K P R, Beddell C R, Lindon J C, et al. Application of pattern recognition methods to the analysis and classification of toxicological data derived from proton nuclear magnetic resonance spectroscopy of urine[J]. Mol Pharmacol, 1991, 39: 629-642.

[6] Nicholson J K, Lindon J C, Holmes E. Metabonomics: understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data[J]. Xenobiotica, 1999, 29 (11): 1 181-1 189.

[7] Brindle J T, Antti H, Holmes E, et al. Rapid and noninvasive diagnosis of the presence and severity of coronary heart disease using ¹HNMR-based metabonomics[J]. Nat Med, 2002, 8(12): 1 439-1 444.

[8] Ebbels T, Keun H, Beckonert O, et al. Toxicity classification from metabonomic data using a density superposition approach: 'CLOUDS' [J]. Anal Chim Acta, 2003, 490:109-122.

[9] Nicholson J K, Connelly J, Lindon J C, et al. Metabonomics: a platform for studying drug toxicity and gene function[J]. Nat Rev Drug Disc, 2002, 1: 153-161.

[10] Griffin J L. Metabonomics: NMR spectroscopy and pattern recognition analysis of body fluids and tissues for characterization of xenobiotic toxicity and disease diagnosis[J]. Curr Opin Chem Biol, 2003, 7 (5): 648-654.

[11] Chen Wen-xue(陈文学), Deng Feng(邓风），Yue Yong(岳勇）. Application of NMR techniques to biological tissues(核磁共振技术在生物组织中的应用)[J]. Chinese J Magn Reson(波谱学杂志）, 2004, 21(1):127-139.

[12] Lindon J C, Holmes E, Nicholson J K. Pattern recognition methods and applications in biomedical magnetic resonance[J]. Prog Nucl Magn Reson Spectrosc, 2001, 39: 1-40.

[13] Geladi P, Kowalski B R. Partial least-squares regression: a tutorial[J]. Anal Chim Acta, 1986, 185: 1-17.

[14] Bollard M E, Holmes E, Lindon J C, et al. Investigations into Biochemical Changes Due to Diurnal Variation and Estrus Cycle in Female Rats Using High-Resolution \{\}\+1H NMR Spectroscopy of Urine and Pattern Recognition[J]. Anal Biochem, 2001, 295: 194-202.

[15] Holmes E, Nicholls A W, Lindon J C, et al. Chemometric Models for Toxicity Classification Based on NMR Spectra of Biofluids[J]. Chem Res Toxicol, 2000, 13: 471-478.

[16] Han Jiawei, Kamber M. Data Mining[M]. Beijing: Higher Education Press, 2001. 340.

[17] Siedlecki W, Siedlecka K, Sklansky J. An overview of mapping techniques for exploratory pattern analysis[J]. Pattern Recogn, 1988, 21 (5): 411-429.

[18] Bian Zhao-qi（边肇祺），Zhang Xue-gong（张学工）. Pattern Recognition（2nd ed）（模式识别第二版）[M]. Beijing(北京): Tsinghua university press(清华大学出版社）, 2000. 8.212.244.

[19] Waters N J, Holmes E, Williams A, et al. NMR and Pattern Recognition Studies on the Time-Related Metabolic Effects of α-Naphthylisothiocyanate on Liver, Urine, and Plasma in the Rat: An Integrative Metabonomic Approach[J]. Chem Res Toxicol, 2001, 14: 1 401-1 412.

[20] Lenz E M, Bright J, Wilson I D, et al. A ¹H NMR-based metabonomic study of urine and plasma samples obtained from healthy human subjects[J]. J Pharm Biomed Anal, 2003, 33 (5): 1 103-1 115. 

[21] Wang Yulan, Bollard M E, Keun H, et al. Spectral editing and pattern recognition methods applied to high-resolution magic-angle spinning ₁H nuclear magnetic resonance spectroscopy of liver tissues[J]. Anal Biochem, 2003, 223 (1): 26-32.

[22] Ren Ruo-en（任若恩）, Wang Hui-wen（王惠文）.Mutlivariate Statistical Data Analysis（多元统计数据分析）[M]. Beijing(北京）: National defence industry press（国防工业出版社），1999. 104.

[23] Sammon J W. A Nonlinear Mapping for Data Structure Analysis[J]. IEEE T COMPUT, 1969, 18(5): 401-409.

[24] Chen Nian-yi（陈念贻），Qin Pei（钦佩），Chen Rui-liang（陈瑞亮），et al. Pattern Recognition Methods' Applications in Chemical Engineering（模式识别方法在化学化工中的应用）[M]. Beijing(北京）: Science press（科学出版社）, 2000. 24.54.56.60.63.

[25] Gartland K P, Sanins S M, Nicholson J K, et al. Pattern recognition analysis of high resolution 1H NMR spectra of urine: A nonlinear mapping approach to the classification of toxicological data[J]. NMR Biomed, 1990, 3 (4): 166-172. 

[26] Bechonert O, Bollard M E, Ebbels T M D, et al. NMR-based metabonomic toxicity classification: hierarchical cluster analysis and k-nearest-neighbour approaches[J]. Anal Chim Acta, 2003, 490: 3-15.

[27] Sheng Zhou（盛骤），Xie Shi-qian（谢式千），Pan Cheng-yi（潘承毅）.Probability Theory and Mathematical Statistics (3rd ed)（概率论与数理统计）[M]. Beijing(北京）: Higher Education Press（高等教育出版社）, 2001. 312.

[28] Wang Hui-wen（王惠文）.Partial Least-squares Regression Method and Applications（偏最小二乘方法及其应用）[M]. Beijing(北京）: National defence industry press（国防工业出版社），1999. 57.

[29] Wold S. Pattern recognition by means of disjoint principal components models[J]. Pattern Recogn, 1976, 8:127-139.

[30] Tobias R D. An Introduction to Partial Least Squares Regression[EB/OL]. TS-509,SAS Institute Inc, Cary N C. ftp.sas.com/techsup/download/technote/ts509.pdf,1997-04.

[31] Wold S, Antti H, Lindgren F, et al. Orthogonal signal correction of near-infrared spectra[J]. Chemom Intell Lab Syst, 1998, 44:175185.

[32] Hauksson J B, Edlund U, Trygg J. NMR processing techniques based on multivariate data analysis and orthogonal signal correction. 13C CP/MAS NMR spectroscopic characterization of softwood kraft pulp[J]. Magn Reson Chem, 2001, 39: 267-275.

[33] eckwith-Hall B M, Brindle J T, Barton R H, et al. Application of orthogonal signal correction to minimise the effects of physical and biological variation in high resolution ¹H NMR spectra of biofluids[J]. Analyst, 2002, 127: 1 283-1 288.

[34] Gavaghan C L, Wilson I D, Nicholson J K. Physiological variation in metabolic phenotyping and functional genomic studies: use of orthogonal signal correction and PLS-DA[J]. FEBS Lett, 2002, 530: 191-196.

[35] Keun H C, Ebbles T M D, Antti H, et al. Improved analysis of multivariate data by variable stability scaling: application to NMR-based metabolic profiling[J]. Anal Chim Acta, 2003, 490: 265-276.

[36] Charlton A J, Farrington W H H, Brereton P. Application of ¹H NMR and Multivariate Statistics for Screening Complex Mixtures: Quality Control and Authenticity of Instant Coffee[J]. J Agric Food Chem, 2002, 50:3 098-3 103.

[37] Amendolia S R, Doppiu A, Ganadu M L, et al. Classification and Quantitation of \{\}\+1H spectra of Alditols Binary Mixtures Using Artificial Neural Networks[J]. Anal Chem, 1998, 70: 1 249-1 254.

[38] Anker L S, Jurs P C. Prediction of carbon-13 nuclear magnetic resonance chemical shifts by artificial neural networks[J]. Anal Chem, 1992, 64 (10): 1 157-1 164.

[39] Hare B J, Prestegard J H. Application of neural networks to automated assignment of NMR spectra of proteins[J]. J Biomol NMR, 1994, 4: 35-46.

[40] Ala-Korpela M, Hiltunen Y, Bell J D. Quantification of biomedical NMR data using artificial neural network analysis: lipoprotein lipid profiles from ¹H NMR data of human plasma[J]. NMR Biomed. 1995, 8(6):235-244.

[41] Duda R O, Hart P E, Stork D G. Pattern Classification（2nd ed）[M]. Beijing: China Machine Press, 2004.288.

[42] Specht D F. Probabilistic Neural Networks[J]. Neural Networks, 1990, 3: 109-118.

[43] Romero R D, Touretzky D S, Thibadeau R H. Optical Chinese character recognition using probabilistic neural networks[J]. Pattern Recogn, 1997, 30 (8): 1 279-1 292.

[44] Kim S H, Chun S H. Graded forecasting using an array of bipolar predictions: application of probabilistic neural networks to a stock market index[J]. Int J Forecast, 1998, 14: 323-337.

[45] Hammond M H, Riedel C J, Rose-Pehrsson S L, et al. Training set optimization methods for a probabilistic neural network[J]. Chemom Intell Lab Syst, 2004, 71: 73-78.

[46] Holmes E, Nicholson J K, Tranter G. Metabonomic Characterization of Genetic Variations in Toxicological and Metabolic Responses Using Probabilistic Neural Networks[J]. Chem Res Toxicol, 2001, 14: 182-191.

[47] Parzen E. On estimation of a probability density function and mode[J]. Annu Math Stat, 1962, 33: 1 065-1 076.

[48] Bishop C. Neural networks for pattern recogniton[M]. Oxford: University Press, 1995.P164.

[49] Yan Xian-zhong(颜贤忠）, Zhao Jian-yu(赵剑宇）, Peng Shuang-qing(彭双清），et al. Metabonomics in post-genomic era(代谢组学在后基因组时代的作用)[J]. Chinese J Magn Reson(波谱学杂志）, 2004, 21(2): 263-271.