Application of Raman Spectroscopy for an Easy-to-Use, on-Field, Rapid, Nondestructive, Confirmatory Identification of Body Fluids
Igor K. Lednev
Abstract
According to the major goals of this project, a novel method was developed for the nondestructive, confirmatory identification of body fluid traces based on Raman spectroscopy combined with advanced statistics. Our prior hypothesis that the Raman spectroscopic signature of each body fluid is unique and can be used for identification purposes was confirmed. A library of Raman signatures for various body fluids, specifically blood, semen, vaginal fluid, saliva and sweat was built and software for the automatic identification of an unknown sample was developed. The method was expanded to include body fluid stains on various substrates and contaminated stains.
Human body fluids were collected from anonymous groups of donors. The donors’ age, race and gender were disclosed to assure the required diversity and sample size. Samples of each body fluid were obtained from donors representing five races, including: Caucasian, African American, Asian, Hispanic and Native American. Dry traces of body fluids were tested using Raman microscope equipped with a computer-controlled stage for rapid mapping. It was found that dry traces of body fluids are intrinsically heterogeneous meaning that Raman spectra measured from different spots varied significantly. We proposed a new approach to identification by utilizing multidimensional spectroscopic signature to account for sample heterogeneity and variations with donor. Our developed software algorithms compare experimental Raman spectra with the library of Raman signatures, providing quantitative measures of similarity. It was demonstrated that an unknown stain of a pure body fluid can be identified with near 100% accuracy provided that good quality Raman spectra are acquired.
We demonstrated that mixed samples can be identified through automatic mapping if two body fluids are not thoroughly mixed. Thoroughly mixed samples can be identified using support vector machine (SVM)-based methods. Specifically, SVM-based method was able to identify a small contribution (2-5%) of thoroughly mixed semen in blood and blood in semen samples. As a specific objective of this project, the problem of substrate interference was also explored. We demonstrated that dry traces of blood and semen can be identified successfully on several substrates of practical importance.
Overall, this study demonstrates great potential of Raman spectroscopy for the nondestructive, confirmatory identification of body fluid traces, including stains on various substrates and contaminated mixtures. Additionally, a program was developed for the automatic identification of body fluids in dry mixtures.