Pattern recognition algorithms responses

In the field of chemical sensors, the revolution in software and inexpensive hardware means that not only nonlinear chemical responses can be tolerated, but incomplete selectivity to a variety of chemical species can also be handled. Arrays of imperfectly selective sensors can be used in conjunction with pattern recognition algorithms to sort out classes of chemical compounds and thek concentrations when the latter are mixed together. 

The multivariate tools typically used for the NIR-CI analysis of pharmaceutical products fall into two main categories pattern recognition techniques and factor-based chemometric analysis methods. Pattern recognition algorithms such as spectral correlation or Euclidian distance calculations basically determine the similarity of a sample spectrum to a reference spectrum. These tools are especially useful for images where the individual pixels yield relatively unmixed spectra. These techniques can be used to quickly define spatial distributions of known materials based on external reference spectra. Alternatively, they can be used with internal references, to locate and classify regions with similar spectral response. 

In general, carbon black composites demonstrate fast response times, good reversibility, reproducibility, and stability. However, they lack the ability to react selectively to different gaseous analytes, making them better suited for a sensor array application where pattern recognition algorithms can be used to identify analytes. 

Surface Acoustic Wave (SAW) sensors detect changes in the properties of acoustic waves as they travel at ultrasonic frequencies in piezoelectric materials. The basic transduction mechanism involves interaction of these waves with surface-attached matter. Multiple sensor arrays with multiple coatings and pattern recognition algorithms provide the means to identify agent classes and reject interferant responses that could cause false alarms. Acoustic wave sensors are used in mobile detectors to detect nerve and blister agents. 

Additional specificity is derived through the use of a matrix of SAW devices where each of the elements, a separate SAW oscillator, is coated with a different polymer. For example, one of the SAW systems currently under evaluation contain six SAW oscillators and use six different polymers two have affinities for nerve agents, two for blister agents, one for water vapor and one for non-polar compounds such as hydrocarbons. By determination of the magnitudes of frequency shifts for each of the crystals when the device is exposed to a certain compound, a library of patterns of responses can be generated and, by use of appropriate pattern recognition algorithms, specificity of response can be realized [5]. 

Another significant environmental factor for vapor-phase applications is humidity. The ubiquitous nature of water vapor requires development of means to exclude or correct for interferences from water [92a,b]. Careful selection of coating materials, for example, can minimize the effect of water vapor on the sensor response. Alternatively, a coating with appropriate sensitivity to water can be used in the development of correction algorithms [93]. Other instrumental or system approaches, such as preconcentrators or sensor arrays with pattern recognition [94a-c], will be discussed in Section 5.5 and in Chapter 6. 

FIGURE 3.1 An ALOPEX system. The stimulus is presented on the cathode ray tube (CRT). The observer or any pattern recognition device (PRD) faces the CRT the subject s response is sent to the ALOPEX interface unit where it is recorded and integrated and the fi nal response is sent to the computer. The computer calculates the values of the new pattern to be presented on the CRT according to the ALOPEX algorithm and the process continues until the desired pattern appears on the CRT. At this point the response is considered to be optimal and the process stops. 

---