Detection and Classification of Organophosphate Nerve Agent Simulants Using Support Vector Machines with Multiarray Sensors

Abstract

The need for rapid and accurate detection systems is expanding and the utilization of cross-reactive sensor arrays to detect chemical warfare agents in conjunction with novel computational techniques may prove to be a potential solution to this challenge. We have investigated the detection, prediction, and classification of various organophosphate (OP) nerve agent simulants using sensor arrays with a novel learning scheme known as support vector machines (SVMs). The OPs tested include parathion, malathion, dichlorvos, trichlorfon, paraoxon, and diazinon. A new data reduction software program was written in MATLAB V. 6.1 to extract steady-state and kinetic data from the sensor arrays. The program also creates training sets by mixing and randomly sorting any combination of data categories into both positive and negative cases. The resulting signals were fed into SVM software for "pairwise" and "one" vs all classification. Experimental results for this new paradigm show a significant increase in classification accuracy when compared to artificial neural networks (ANNs). Three kernels, the S2000, the polynomial, and the Gaussian radial basis function (RBF), were tested and compared to the ANN. The following measures of performance were considered in the pairwise classification: receiver operating curve (ROC) A z indices, specificities, and positive predictive values (PPVs). The ROC A z values, specifities, and PPVs increases ranged from 5% to 25%, 108% to 204%, and 13% to 54%, respectively, in all OP pairs studied when compared to the ANN baseline. Dichlorvos, trichlorfon, and paraoxon were perfectly predicted. Positive prediction for malathion was 95%.

Multicomponent analysis of alcohol vapors using integrated gas chromatography with sensor arrays

“…Although the performance of the GC-TCD was better than the GC-PSAD in terms of peak resolution and reproducibility, there is a lot of promise on the improvement of the PSAD because its inherent flexibility in changing the nature and ratios of monomers and dopants to tune its selectivity on one hand or to widen the scope of responding analytes on the other. Also, because of the many and variable sensor signals that are possible from this kind of system, there is the possibility of combining novel machine learning programs that can aid in the rapid and cost effective identification of analytes as we have just recently demonstrated [20]. In that case it would have been impossible to identify with an electronic nose system alone or too expensive with a GC-MS system.…”

Data fusion and hyperspectral imaging in tandem with least squares-support vector machine for prediction of sensory quality index scores of fish fillet

“…LS-SVM is an evolutionary version of the standard support vector machines (SVM) and has been widely developed for the optimal control of nonlinear systems and spectral calibration (Suykens et al, 2002;Bao, Liu, Kong, Sun, He & Qiu, 2014). This method utilizes nonlinear map function, projects input features to a high dimensional space, and adopts the Lagrange multiplier to compute the partial differentiation of each feature for converting the optimization problem into resolving the linear algebraic equation (Sadik et al, 2004). The radial basis function (RBF) as a popular kernel function is normally used in LS-SVM and the grid-search technique is usually applied to determine the optimal parameters obtained from the RBF kernel: the regularization parameter g and the width of RBF parameter s 2 (Suykens, Vandewalle, & De Moor, 2001).…”

Feature Selection Using Support Vector Machines and Independent Component Analysis for Wound Infection Detection by Electronic Nose

“…SVM is not only good at classification (13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23) but also widely used in feature selection. (9,(24)(25)(26)(27)(28)(29)(30)(31)(32)(33) In this paper, a new method such as the embedded method for feature selection, which is specifically designed to work with SVM and ICA, is introduced.…”