A5059040456- Spectroscopy and Chemometric Analyses
- Advanced Chemical Sensor Technologies
- Analytical Chemistry and Chromatography
- Advanced Statistical Methods and Models
- Water Quality Monitoring and Analysis
- Gene expression and cancer classification
- Metabolomics and Mass Spectrometry Studies
- Fault Detection and Control Systems
- Scientific Measurement and Uncertainty Evaluation
- Analytical Chemistry and Sensors
- Blind Source Separation Techniques
- Advanced Biosensing Techniques and Applications
- Mass Spectrometry Techniques and Applications
- Pesticide Residue Analysis and Safety
- Statistical and numerical algorithms
- Sensory Analysis and Statistical Methods
- NMR spectroscopy and applications
- Molecular Biology Techniques and Applications
- Color Science and Applications
- Structural Health Monitoring Techniques
- Identification and Quantification in Food
- Chemical Thermodynamics and Molecular Structure
- Advanced Proteomics Techniques and Applications
- Bioinformatics and Genomic Networks
- Electrochemical Analysis and Applications
Dalhousie University
2015-2024
Michigan State University
1989-1992
University of British Columbia
1989-1991
The theoretical principles and practical implementation of a new method for multivariate data analysis, maximum likelihood principal component analysis (MLPCA), are described. MLCPA is an analog to (PCA) that incorporates information about measurement errors develop PCA models optimal in sense. foundations MLPCA initially established using regression model extended the framework singular value decomposition (SVD). An efficient reliable algorithm based on alternating Generalization allows its...
Most cells on earth exist in a quiescent state. In yeast, quiescence is induced by carbon starvation, and exit occurs when source becomes available. To understand how survive in, from this state, mRNA abundance was examined using oligonucleotide-based microarrays quantitative reverse transcription-polymerase chain reaction. Cells stationary-phase cultures exhibited coordinated response within 5-10 min of refeeding. Levels >1800 mRNAs increased dramatically (>or=64-fold), smaller group...
The characteristics of baseline drift are discussed from the perspective error covariance. From this standpoint, operation derivative filters as preprocessing tools for multivariate calibration is explored. It shown that convolution filter coefficients with covariance matrices data tend to reduce contributions correlated error, thereby reducing presence noise. This theory corroborated by examination experimental before and after preprocessing. proposed maximum likelihood principal components...
Modeling of gene expression data from time course experiments often involves the use linear models such as those obtained principal component analysis (PCA), independent (ICA), or other methods. Such methods do not generally yield factors with a clear biological interpretation. Moreover, implicit assumptions about measurement errors limit application these to log-transformed data, destroying structure in untransformed data.In this work, method for decomposition by multivariate curve...
Two new approaches to multivariate calibration are described that, for the first time, allow information on measurement uncertainties be included in process a statistically meaningful way. The methods, referred as maximum likelihood principal components regression (MLPCR) and latent root (MLLRR), based principles of parameter estimation. MLPCR MLLRR generalizations (PCR), which has been widely used chemistry, (LRR), virtually ignored this field. Both methods decomposition data matrix by...
In forensic examination of questioned documents a type casework often encountered are the frauds that occur by mean addition and adulteration parts text or numbers on document. The goal this work is to evaluate performance hyperspectral images (HSI) in near (NIR) middle (MIR) regions, combined with unsupervised pattern recognition techniques Principal Component Analysis (PCA) Projection Pursuit (PP) for rapid, reliable non-destructive identification document falsifications means alterations...
Medidas analíticas modernas são comumente apresentadas na forma de vetores (e.g., espectros) ou estruturas dados ordem maior como matrizes, que normalmente sujeitas a estratégias análise multivariada para extração informações.Um aspecto dessas medidas é pouco compreendida natureza subentendida dos erros, e eles afetam obtenção da informação química.Este Account descreve alguns métodos podem ser utilizados caracterizar erros multivariadas esta pode usada melhorar os resultados das análises...
The application of trilinear decomposition (TLD) to the analysis fluorescence excitation-emission matrices mixtures polycyclic aromatic hydrocarbons (PAHs) is described. variables constituting third-order tensor are excitation wavelength, emission and concentration a quencher (nitromethane). addition PAH selectively reduces intensity mixture components according Stern-Volmer equation. TLD allows three-way matrix be decomposed give unique solutions for spectrum, quenching profiles each...
ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTChemical acoustic emission analysis in the frequency domainPeter D. Wentzell and Adrian P. WadeCite this: Anal. Chem. 1989, 61, 23, 2638–2642Publication Date (Print):December 1, 1989Publication History Published online1 May 2002Published inissue 1 December 1989https://pubs.acs.org/doi/10.1021/ac00198a010https://doi.org/10.1021/ac00198a010research-articleACS PublicationsRequest reuse permissionsArticle Views184Altmetric-Citations29LEARN ABOUT THESE...
The transmittance spectra of a long-period fiber grating element immersed in different mixtures water and dimethyl sulfoxide were recorded. obtained data compared with theoretical model based on linearly polarized modes treated coupled-mode approach. Excellent agreement between the measurements results was found over wide wavelength range.