Calibration constructing models

In order to construct a calibration model, the values of the parameters to be determined must be obtained by using a reference method. The optimum choice of reference method will be that providing the highest possible accuracy and precision. The quality of the results obtained with a multivariate calibration model can never exceed that of the method used to obtain the reference values, so the choice should be carefully made as the quality of the model will affect every subsequent prediction. The averaging of random errors inherent in regression methods can help construct models with a higher precision than the reference method. 

In contras to unsupervised methods, supervised pattern-recognition methods (Section 4.3) use class membership information in the calculations. The goal of these methods is to construct models using analytical measurements to predict class membership of future samples. Class location and sometimes shape are used in the calibration step to construct the models. In prediction, these moddsare applied to the analytical measurements of unknowu samples to predict dsss membership. 

Multivariate calibration tools are used to construct models for predicting some characteristic of future samples. Chapter 5 begins with a discussion of the reasons for choosing multivariate over univariate calibration methods. The most widely used multivariate calibration tools are then presented in two categories classical and inverse methods. 

The classical models of spiral galaxies were constructed using rotation velocities. In contrast, the models of elliptical galaxies were found from luminosity profiles and calibrated using central velocity dispersions or motions of companion galaxies. An overview of classical methods to construct models of galaxies is given by Perek (1962). 

Concentrations are useful for assessing trends, but do not adequately describe the fate of a contaminant in the environment. The rates of movement of contaminants from one compartment to another are necessary to assess fate, and to construct models that can predict fate under different environmental conditions. Models that describe contaminant transport and fate can range from simple equilibrium box models to highly complex dynamic models. For modelling PCB fate, accuracy and precision are limited by our ability to describe the processes involved, and the availability of actual field data for calibrating and validating the models. 

This requirement also makes good sense. A calibration is nothing more than a mathematical model that relates the behavior of the measureable data to the behavior of that which we wish to predict. We construct a calibration by finding the best representation of the fit between the measured data and the predicted parameters. It is not surprising that the performance of a calibration can deteriorate rapidly if we use the calibration to extrapolate predictions for... 

In case of fast gradient (below 15 min), S could be considered constant for all the investigated molecules and wiU only have a small influence on the retention time of the compounds. Thus, the gradient retention times, of a calibration set of compounds are linearly related to the ( )o values [39]. Moreover, Valko et al. also demonstrated that the faster the gradient was, the better the correlation between t, and < )o [40]. Once the regression model was established for the calibration standards, Eq. 8 allowed the conversion of gradient retention times to CHI values for any compound in the same gradient system. Results are then suitable for interlaboratory comparison and database construction. The CH I scale (between 0 and 100) can be used as an independent measure of lipophilicity or also easily converted to a log P scale. 

For PyMS to be used for (1) routine identification of microorganisms and (2) in combination with ANNs for quantitative microbiological applications, new spectra must be comparable with those previously collected and held in a data base.127 Recent work within our laboratory has demonstrated that this problem may be overcome by the use of ANNs to correct for instrumental drift. By calibrating with standards common to both data sets, ANN models created using previously collected data gave accurate estimates of determi-nand concentrations, or bacterial identities, from newly acquired spectra.127 In this approach calibration samples were included in each of the two runs, and ANNs were set up in which the inputs were the 150 new calibration masses while the outputs were the 150 old calibration masses. These associative nets could then by used to transform data acquired on that one day to data acquired at an earlier data. For the first time PyMS was used to acquire spectra that were comparable with those previously collected and held in a database. In a further study this neural network transformation procedure was extended to allow comparison between spectra, previously collected on one machine, with spectra later collected on a different machine 129 thus calibration transfer by ANNs was affected. Wilkes and colleagues130 have also used this strategy to compensate for differences in culture conditions to construct robust microbial mass spectral databases. 

It must be made clear that WSL has only used olfactometers constructed on its own premises and calibrated by WSL staff one such instrument is the WSL Transportable model. (It is emphasised that this olfactometer is not the same as the commercially-available Portable model.) The WSL Transportable olfactometer and its operation have been described in detail previously (1) and only its main features need be reviewed here. From time to time, certain applications have required a special re-design of the olfactometer, e.g. to measure very weak odours, but the basic principles and the mode of operation have remained unchanged for some years. The main features of the standard WSL Transportable olfactometer are as follows ... 

Figure 1. Plots showing the Calibration Process. A. Response transformation to constant variance Examples showing a. too little, b. appropriate, and c. too much transformation power. B. Amount Transformation in conforming to a (linear) model. C. Construction of p. confidence bands about the regressed line, q. response error bounds and intersection of these to determine r. the estimated amount interval.

Figure 14.3 schematizes the use of spectral information for constructing mnltivariate calibration models. The pretreatment of the sample spectrnm and variable selection processes involve constructing both identification libraries and calibration models for qnantifying APIs however, the samples and spectra shonld snit the specific aim in each case. 

Calibration is the process by which a mathematical model relating the response of the analytical instrument (a spectrophotometer in this case) to specific quantities of the samples is constructed. This can be done... 

The body of samples selected is split into two subsets, namely the calibration set and the validation set. The former is used to construct the calibration model and the latter to assess its predictive capacity. A number of procedures for selecting the samples to be included in each subset have been reported. Most have been applied to situations of uncontrolled variability spanning much wider ranges than those typically encountered in the pharmaceutical field. One especially effective procedure is that involving the selection of as many samples as required to span the desired calibration range and encompassing the whole possible spectral variability (i.e. the contribution of physical properties). The choice can be made based on a plot of PCA scores obtained from all the samples. 

NIR models are validated in order to ensure quality in the analytical results obtained in applying the method developed to samples independent of those used in the calibration process. Although constructing the model involves the use of validation techniques that allow some basic characteristics of the model to be established, a set of samples not employed in the calibration process is required for prediction in order to conhrm the goodness of the model. Such samples can be selected from the initial set, and should possess the same properties as those in the calibration set. The quality of the results is assessed in terms of parameters such as the relative standard error of prediction (RSEP) or the root mean square error of prediction (RMSEP). 

The model can lose its predictive capacity for reasons such as a dramatic change in the nature of the samples. A new calibration model better adjnsted to the new situation must therefore be constructed. In other cases, however, the model can be re-adjnsted (restandardized) in order to ensure a predictive capacity not significantly different from the original one. 

Standardizing the coefficients of the model entails modifying the calibration equation. This procedure is applicable when the original equipment is replaced (situation 1 above). Forina et al. developed a two-step calibration procedure by which a calibration model is constructed for the master (F-X), its spectral response correlated with that of the slave X-X) and, finally, a global model correlating variable Y with both X and X is obtained. The process is optimized in terms of SEP and SEC for both instruments as it allows the number of PLS factors used to be changed. Smith et al. propose a very simple procedure to match two different spectral responses. 

Standardizing the spectral response is mathematically more complex than standardizing the calibration models but provides better results as it allows slight spectral differences - the most common between very similar instruments - to be corrected via simple calculations. More marked differences can be accommodated with more complex and specific algorithms. This approach compares spectra recorded on different instruments, which are used to derive a mathematical equation, allowing their spectral response to be mutually correlated. The equation is then used to correct the new spectra recorded on the slave, which are thus made more similar to those obtained with the master. The simplest methods used in this context are of the univariate type, which correlate each wavelength in two spectra in a direct, simple manner. These methods, however, are only effective with very simple spectral differences. On the other hand, multivariate methods allow the construction of matrices correlating bodies of spectra recorded on different instruments for the above-described purpose. The most frequent choice in this context is piecewise direct standardization... 

Standardizing the predicted values is a simple, useful choice that ensures smooth calibration transfer in situations (a) and (b) above. The procedure involves predicting samples for which spectra have been recorded on the slave using the calibration model constructed for the master. The predicted values, which may be subject to gross errors, are usually highly correlated with the reference values. The ensuing mathematical relation, which is almost always linear, is used to correct the values subsequently obtained with the slave. 

Tablets account for more than 80% of all pharmaceutical formulations therefore, the development and implementation of NIR methods for determining APIs in intact tablets is of a high interest with a view to assuring content uniformity and quality in the end product. Blanco et al developed an innovative strategy to prepare calibration samples for NIR analysis by using laboratory-made samples obtained by mixing the API and excipients in appropriate proportions and compacting the mixture at a pressure similar to that used industrially. This way of matching laboratory and production samples affords more simple and robust NIR methods which require the use of neither HPLC nor UV-vis spectroscopy as reference rather, reference values are obtained by weighing during preparation of the samples. The PLS calibration models thus constructed exhibited a good predictive ability with various production batches.

---