P-II
Raman spectroscopy and machine learning for medical diagnostics and forensic purposes
Igor Lednev
Department of Chemistry, Department of Biological Sciences, University at Albany, State University of New York (SUNY), USA
Raman spectroscopy combined with advanced statistics is uniquely suitable for characterizing microheterogeneous samples. Understanding the structure and (bio)chemical composition of samples at the microscopic level is important for many practical applications including material science, pharmaceutical industry, etc. We have recently demonstrated a great potential of Raman hyperspectroscopy for disease diagnostics and forensic purposes. In this presentation, we will discuss the development of a new, noninvasive method for Alzheimer's disease (AD) diagnostics based on Raman spectroscopy of blood. Near infrared (NIR) Raman hyperspectroscopy coupled with advanced multivariate statistics was utilized for differentiating patients diagnosed with Alzheimer's disease, other types of dementia and healthy control subjects with more than 95% sensitivity and specificity. When fully developed, this fast, inexpensive noninvasive method could be used for screening at risk patient populations for AD development and progression.
Raman spectroscopy has already found numerous applications in forensic chemistry providing confirmatory identification of analytes. The technique is non-destructive, rapid and requires little or no sample preparation. Furthermore, portable Raman instruments are readily available allowing for crime scene accessibility. We have recently demonstrated that Raman microspectroscopy can be used for the identification of biological stains at a crime scene indicating the type of body fluid. In addition, peripheral and menstrual blood as well as human and animal blood can be differentiated. The time since deposition of bloodstain can be estimated up to two years. Most recently, we demonstrated the proof-of-concept for phenotype profiling based on Raman spectroscopy of dry traces of body fluids including the determination of sex, race and age group of the donor.