Authors
Abstract
Along with the development of laboratory methods for diagnosing addiction, concealment ways, either physically or chemically, for creating false results have been in progress. In this research based on the Laser Induced Breakdown Spectroscopy technique (LIBS) and analyzing hair of addicted and normal people, we are proposing a new method to overcome problems in conventional methods and reduce possibility of cheating in the process of diagnosing addiction. For this purpose, at first we have sampled hair of 17 normal and addicted people and recorded 5 spectrums for each sample, overall 170 spectrums. After analyzing the recorded LIBS spectra and detecting the atomic and ionic lines as well as molecular bands, relative intensities of emission lines for Aluminum to Calcium (Al/Ca) and Aluminum to Sodium (Al/Na) were selected as the input variables for the Support Vector Machine model (SVM).The Radial Basis, Polynomial Kernel functions and a linear function were chosen for classifying the data in SVM model. The results of this research showed that by the combination of LIBS technique and SVM one can distinguish addicted person with precision of 100%. Because of several advantages of LIBS such as high speed analysis and being portable, this method can be used individually or together with available methods as an automatic method for diagnosing addiction through hair analysis.
Keywords
2. S George and R Braithwaite, Journal of analytical toxicology 20 (1996) 195.
3. M Bahreini and S H Tavassoli, Journal of Lasers in Medical Sciences 3 (2012) 127.
4. M Gazmeh, et al., Applied optics 54 (2015) 123.
5. S Musazzi and U Perini, “Laser-induced breakdown spectroscopy”, Springer Series in Optical Sciences, 182. ISBN 978-3-642-45084-6. Springer-Verlag Berlin Heidelberg, (2014) 1.
6. NIST, “Atomic spectra database,” http://www.nist.gov/pml/data/asd.cfm.
7. Ż Trzebiatowska, et al., Laser Physics and Applications, Mar 15 International Society for Optics and Photonics (2013) 87700B.
8. J Cisewski, et al,. Journal of Chemometrics 26 (2012) 143.
9. N C Dingari, et al,. Analytical chemistry 84 (2012) 2686.
10. N Cristianini and J Shawe-Taylor, “An introduction to support vector machines and other kernel-based learning methods”, Cambridge university press, (2000).
11. V Vapnik, “The nature of statistical learning theory”, Springer Science & Business Media, (2013).
12. V N Vapnik and V Vapnik, “Statistical learning theory” Wiley New York, (1998).