Expert system for monitoring the tributyltin content in inland water samples

Abstract In this study we discuss an attempt to build an expert system that can support decision making by analytical chemists regarding the presence of tributyltin (TBT) in inland Polish water samples in detail. It is possible to conclude with at least a 0.93 probability that a sample is free of TBT using the expert system that was constructed (if a sample is analyzed in accordance with the European norm PN-EN ISO 17353:2006). This idea, which is based on the efficient use of the information that is stored in a chromatographic database, can easily be extended to monitor other priority substances in water samples. Our on-going research, which is focused on octylphenols in water samples, has provided very encouraging results and additionally supports this hypothesis. The proposed framework can also be attractive to other testing laboratories that have a similar scope of expertise and follow the same analytical protocols. Moreover, as a natural consequence of our research further efforts should lead to the development of a ready-to-use product that would offer testing laboratories validated chromatographic libraries along with the expert system(s) with the possibility of upgrading them with respect to an increasing pool of analyzed samples. Such a solution when implemented in a testing laboratory environment may have a wide economic impact on its further functioning and increase throughput efficiency, especially in a case in which monitoring priority substances in water is a major concern.

[1]  R. Tibshirani,et al.  Sparsity and smoothness via the fused lasso , 2005 .

[2]  Armando C. Duarte,et al.  Application of Chemometrics in Separation Science , 2006 .

[3]  Hein Putter,et al.  The bootstrap: a tutorial , 2000 .

[4]  M Daszykowski,et al.  Target selection for alignment of chromatographic signals obtained using monochannel detectors. , 2007, Journal of chromatography. A.

[5]  P. Eilers.  A perfect smoother. , 2003, Analytical chemistry.

[6]  Ryan Gosselin,et al.  A Bootstrap-VIP approach for selecting wavelength intervals in spectral imaging applications , 2010 .

[7]  B. Walczak,et al.  About kernel latent variable approaches and SVM , 2005 .

[8]  Lutgarde M. C. Buydens,et al.  Interpretation of variable importance in Partial Least Squares with Significance Multivariate Correlation (sMC) , 2014 .

[9]  M. Daszykowski,et al.  Metabolomics provide new insights on lung cancer staging and discrimination from chronic obstructive pulmonary disease. , 2014, Journal of pharmaceutical and biomedical analysis.

[10]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[11]  G. M. Escandar,et al.  Determination of tributyltin at parts-per-trillion levels in natural waters by second-order multivariate calibration and fluorescence spectroscopy , 2013 .

[12]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[13]  Tarja Rajalahti,et al.  Discriminating variable test and selectivity ratio plot: quantitative tools for interpretation and variable (biomarker) selection in complex spectral or chromatographic profiles. , 2009, Analytical chemistry.

[14]  E. K. Kemsley,et al.  Discriminant analysis of high-dimensional data: a comparison of principal components analysis and partial least squares data reduction methods , 1996 .

[15]  José Manuel Amigo,et al.  ChroMATHography: solving chromatographic issues with mathematical models and intuitive graphics. , 2010, Chemical reviews.

[16]  Ivana Stanimirova,et al.  Detection of discoloration in diesel fuel based on gas chromatographic fingerprints , 2014, Analytical and Bioanalytical Chemistry.

[17]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[18]  Yizeng Liang,et al.  Preprocessing of analytical profiles in the presence of homoscedastic or heteroscedastic noise , 1994 .

[19]  A. Ng.  Feature selection, L1 vs. L2 regularization, and rotational invariance , 2004, Twenty-first international conference on Machine learning - ICML '04.

[20]  M. P. Gautier,et al.  Determination of organotin compounds by headspace solid-phase microextraction–gas chromatography–pulsed flame-photometric detection (HS-SPME–GC–PFPD) , 2005, Analytical and bioanalytical chemistry.

[21]  Chih-Jen Lin,et al.  LIBLINEAR: A Library for Large Linear Classification , 2008, J. Mach. Learn. Res..

[22]  M. Barker,et al.  Partial least squares for discrimination , 2003 .

[23]  D. Massart,et al.  The Radial Basis Functions — Partial Least Squares approach as a flexible non-linear regression technique , 1996 .

[24]  H. Zou,et al.  Regularization and variable selection via the elastic net , 2005 .

[25]  Beata Walczak,et al.  USE AND ABUSE OF CHEMOMETRICS IN CHROMATOGRAPHY , 2006 .

[26]  Ricardo Erthal Santelli,et al.  Occurrence and chemical speciation analysis of organotin compounds in the environment: a review. , 2010, Talanta.

[27]  Peter M. Williams,et al.  Bayesian Regularization and Pruning Using a Laplace Prior , 1995, Neural Computation.

[28]  J. Carstensen,et al.  Aligning of single and multiple wavelength chromatographic profiles for chemometric data analysis using correlation optimised warping , 1998 .

[29]  Yoav Freund,et al.  A Short Introduction to Boosting , 1999 .