| 1. |
- Andersson, M, et al.
(författare)
-
NIR spectroscopy on moving solids using a scanning grating spectrometer - impact on multivariate process analysis
- 2005
-
Ingår i: Chemometrics and Intelligent Laboratory Systems. - : Elsevier BV. - 0169-7439. ; 75:1, s. 1-11
-
Tidskriftsartikel (refereegranskat)abstract
- The effect of sample movement on spectral response during fiber probe diffuse reflectance near-infrared spectrometry (NIR) sampling was characterized. This is of central importance in Process Analytical Chemistry (PAC) and Process Analytical Technology (PAT). The incitement to this study was the observation of spectral artifacts during measurements of powder samples in process streams when using a mechanically scanning spectrometer. Artifacts appeared as momentary changes in the spectral response during acquisition of a scan. These transitions emanate from continuous replacement of the sample subfraction seen by the probe and are typical for turbid media where sample properties may vary locally with respect to scattering and/or absorption. The impact on qualitative and quantitative analysis using chemometric methods such as principal component analysis (PCA) and partial least squares (PLS) regression was evaluated through experimental and theoretical simulations. It was generally found that spectra with the smallest residuals after projection onto the models came from non-moving samples or samples moving only slowly. It is shown that the magnitude of the spectral residuals is directly connected to the effective sample size, which relates both to sample speed as well as to the sample area presented to the probe. Implications for in-line/on-line process analysis of solids are discussed. (C) 2004 Elsevier B.V. All rights reserved.
|
|
| 2. |
- Björk, Anders, et al.
(författare)
-
Modeling of pulp quality parameters from distribution curves extracted from process acoustic measurements on a thermo mechanical pulp (TMP) process
- 2007
-
Ingår i: Chemometrics and Intelligent Laboratory Systems. - : Elsevier BV. - 0169-7439 .- 1873-3239. ; 85:1, s. 63-69
-
Tidskriftsartikel (refereegranskat)abstract
- In this paper the feasibility of modeling strength and optical pulp properties from length distribution curves extracted from acoustic data using continuous wavelet transform-fiber length extraction, CWT-FLE (A Björk and L-G Danielsson, 'Extraction of Distribution Curves from Process Acoustic Measurements on a TMP-Process', Pulp and Paper Canada 105 No. 11 (2004), T260-T264) by use of Partial Least Squares (PLS) have been tested. The curves used have earlier been validated against length distribution curves obtained by analyzing pulp samples with a commercial analyzer (FiberMaster). The curves were extracted from acoustic data without any "calibration" against fiber length analyses. The acoustic measurements were performed using an accelerometer affixed to the refiner blow-line during a full-scale trial with a Sunds Defibrator double disc refiner at SCA Ortviken, Sweden. Pulp samples were collected concurrently with the acoustic measurements and extensive physical testing has been made on these samples. For each trial point three pulp samples were collected. PLS1 and PLS2 models were successfully made linking the distribution curves obtained using CWT-FLE to pulp tensile strength properties as well as optical properties. The resulting Root Mean Square Error of Prediction (RMSEP) for all parameters is comparable to what can be obtained by pooling the standard deviations of reference measurements from the different trial points. The results obtained are compared to FiberMaster data modeled in the same fashion, yielding lower prediction errors than the CWT-FLE data. However, this can be partly due to the five-year storage of pulp samples between pulp sampling/acoustic measurement and FiberMaster analyses/sheet testing. The acoustic method is fast and produces results without dead time and could constitute a new tool for improving process control and optimizing the fiber characteristics in a specific process and for a specific purpose. The technique could be implemented in a PC-environment at a fairly low cost.
|
|
| 3. |
- Chauchard, Fablen, et al.
(författare)
-
Localization of embedded inclusions using detection of fluorescence: Feasibility study based on simulation data, LS-SVM modeling and EPO pre-processing
- 2008
-
Ingår i: Chemometrics and Intelligent Laboratory Systems. - : Elsevier BV. - 0169-7439. ; 91:1, s. 34-42
-
Tidskriftsartikel (refereegranskat)abstract
- Fluorescence spectroscopy is a useful technique for tissue diagnostics and is also a promising tool in the characterization of embedded structures in tissue. The emitted fluorescence from an embedded inclusion, marked with a fluorescent compound, is affected by several factors as the light propagates through the medium to the tissue boundary, where the fluorescence light is detected. Tissue absorption, scattering and autofluorescence, as well as the size and depth of the inclusion, affect the detected fluorescence light. The aim of this study is to investigate if the size and location of a fluorescent inclusion could be determined using models based a combination of External Parameter Orthogonalisation (EPO) and Least Squares Support Vector Machine (LS-SVM). This can be very useful for data pre-processing before a full fluorescence tomography reconstruction. The data set consisted of simulated multispectral fluorescence, where depth and radius of a spherical fluorescent inclusion were varied as well as the fluorescence contrast and optical properties of the surrounding tissue. The results showed that the non-linear models based on LS-SVM can simultaneously predict both radius and depth. It was observed that EPO acts as a useful pre-processing tool on spectra for this nonlinear model and that it was necessary to perform EPO to be able to predict the depth with the LS-SVM model. (C) 2007 Elsevier B.V. All rights reserved.
|
|
| 4. |
- Danielsson, Rolf, et al.
(författare)
-
Rapid multivariate analysis of LC/GC/CE data (single or multiple channel detection) without prior peak alignment
- 2006
-
Ingår i: Chemometrics and Intelligent Laboratory Systems. - : Elsevier BV. - 0169-7439 .- 1873-3239. ; 84:1-2, s. 33-39
-
Tidskriftsartikel (refereegranskat)abstract
- One- or two-dimensional data obtained with LC/GC/CE and single or multiple channel detection (MS, UV/VIS) are often used as 'fingerprints' in order to characterize complex samples. The relation between samples is then explored by multivariate data analysis (PCA, hierarchical clustering), but inevitable more or less random variation in separation conditions obstructs the analysis. Several methods for peak alignment have been developed, with more or less increase of time and efforts for computations. In this work another approach is presented, based on a correlation measure less sensitive for variations in retention/migration time. The merits of the method as a fast initial data exploration tool are demonstrated for a case study of urine profiling with CE/MS.
|
|
| 5. |
- Eliasson, Charlotte, 1973, et al.
(författare)
-
Multivariate methodology for surface enhanced Raman chemical imaging of lymphocytes
- 2006
-
Ingår i: Chemometrics and Intelligent Laboratory Systems. - : Elsevier BV. - 0169-7439 .- 1873-3239. ; 81:1, s. 13-20
-
Tidskriftsartikel (refereegranskat)abstract
- Surface enhanced Raman spectroscopy (SERS) was used to study the uptake of rhodamine 6G in human lymphocytes. In total four Raman images of lymphocytes were used. The aim was to find a multivariate methodology capable of separating spectra with chemical information from those that mainly contained the surface enhanced background, in order to create chemical images. The standard PCA procedure was compared with PCA of standard normal variate (SNV) corrected spectra, spectra baseline corrected in the wavelet domain, and variable trimming before PCA, to isolate unique spectra. It was not straightforward to perform a standard PCA for overview, since the small background variation in many variables dominated over the Raman band variation that only occur in few variables. It was shown that wavelet filtering could remove background variations and that variable trimming followed by PCA modelling left the unique Raman spectra as outliers, which facilitated interpretation of the Raman score images.
|
|
| 6. |
- Forshed, Jenny, et al.
(författare)
-
Enhanced multivariate analysis by correlation scaling and fusion of LC/MS and 1H-NMR data
- 2007
-
Ingår i: Chemometrics and Intelligent Laboratory Systems. - : Elsevier B.V. - 0169-7439 .- 1873-3239. ; 85:2, s. 179-185
-
Tidskriftsartikel (refereegranskat)abstract
- A method to enhance the multivariate data interpretation of, for instance, metabolic profiles is presented. This was done by correlation scaling of 1H NMR data by the time pattern of drug metabolite peaks identified by LC/MS, followed by parallel factor analysis (PARAFAC). The variables responsible for the discrimination between the dosed and control rats in this model were then eliminated in both data sets. Next, an additional PARAFAC analysis was performed with both LC/MS and 1H NMR data, fused by outer product analysis (OPA), to obtain sufficient class separation. The loadings from this second PARAFAC analysis showed new peaks discriminating between the classes. The time trajectories of these peaks did not agree with the drug metabolites and were detected as possible candidates for markers. These data analyses were also compared with the PARAFAC analysis of raw data, which showed very much the same loading peaks as for the correlation-scaled data, although the intensities differed. Elimination of the variables correlated with the drug metabolites was therefore necessary to be able to select the peaks which were not drug metabolites and which discriminated between the classes.1
|
|
| 7. |
|
|
| 8. |
- Forshed, Jenny, et al.
(författare)
-
Evaluation of different techniques for fusion of LC/MS and 1HNMR data
- 2007
-
Ingår i: Chemometrics and Intelligent Laboratory Systems. - 0169-7439 .- 1873-3239. ; 85:1, s. 102-109
-
Tidskriftsartikel (refereegranskat)abstract
- In the analyses of highly complex samples (for example, metabolic fingerprinting), the data might not suffice for classification when using only a single analytical technique. Hence, the use of two complementary techniques, e.g., LUMS and H-1-NMR, might be advantageous. Another possible advantage from using two different techniques is the ability to verify the results (for instance, by verifying a time trend of a metabolic pattern). In this work, both LC/MS and H-1-NMR data from analysis of rat urine have been used to obtain metabolic fingerprints. A comparison of three different methods for data fusion of the two data sets was performed and the possibilities and difficulties associated with data fusion were discussed. When comparing concatenated data, full hierarchical modeling, and batch modeling, the first two approaches were found to be the most successful. Different types of block scaling and variable scaling were evaluated and the optimal scaling for each case was found by cross validation. Validations of the final models were performed by means of an external test set.(2)
|
|
| 9. |
- Gabrielsson, Jon, et al.
(författare)
-
OPLS methodology for analysis of pre-processing effects on spectroscopic data
- 2006
-
Ingår i: Chemometrics and Intelligent Laboratory Systems. - : Elsevier BV. - 0169-7439. ; 84:1-2, s. 153-8
-
Tidskriftsartikel (refereegranskat)abstract
- Pre-processing of spectroscopic data is commonly applied to remove unwanted systematic variation. Possible loss of information and ambiguity regarding discarded variation are issues that complicate pre-treatment of data. In this paper, OPLS methodology is applied to evaluate different techniques for pre-processing of spectroscopic data gathered from a batch process. The objective is to present a rational scheme for analysis of pre-processing in order to understand the influence and effect of pre-treatment.O2PLS uses linear regression to divide the systematic variation in X and Y into three parts; one part with joint X–Y covariation, i.e. related to both X and Y, one part of X with Y-orthogonal variation and one part of Y with X-orthogonal variation.All of the investigated pre-treatment methods removed an additive baseline as expected. In the analysis of raw and differentiated data variation associated with the baseline was found in the Y-orthogonal part of X. Orthogonal information was also found in Y, which suggests that this pre-processing procedure not only removed variation. This would have been more difficult to detect without the O2PLS model since both raw and differentiated data must be analysed simultaneously.Development of a knowledge based strategy with OPLS methodology is an important step towards eliminating trial and error approaches to pre-processing.
|
|
| 10. |
- Gut, Luiza, et al.
(författare)
-
Assessment of a two-step partial nitritation/Anammox system with implementation of multivariate data analysis
- 2007
-
Ingår i: Chemometrics and Intelligent Laboratory Systems. - : Elsevier BV. - 0169-7439 .- 1873-3239. ; 86:1, s. 26-34
-
Tidskriftsartikel (refereegranskat)abstract
- Complexity of biological wastewater treatment, in which a variety of physical and (bio)chemical processes concurrently take place, demands appropriate approach in the data analysis. In this study, the data set collected during a 20 month operation of a two-step partial nitritation/Anammox system for nitrogen removal from wastewater (semi-industrial pilot plant scale) are subjected to Principal Component Analysis (PCA) and Partial Least Squares projections to latent structures (PLS) analysis. Interpretation of PCA- and PLS models enable to discern relationships between different factors for the start-up period and stable operation of the pilot plant. Variables like conductivity, pH value, dissolved oxygen concentration and nitrite-to-ammonium ratio (NAR) appear to be the key factors in the process control and monitoring. Extension of the Anammox reactor capacity demands accurate monitoring, principally by scrutinizing nitrite nitrogen concentration in the reactor. These findings suggest that the two methods complement each other in assessing the partial nitritation/Anammox system. This study demonstrated that multivariate data analysis provides the powerful implement in the field of wastewater treatment, especially in investigating novel systems.
|
|
