Chemical forensics is a new field that collects and attributes chemical information of physical evidence to their sources. Ideally, from a chemical forensic analysis, one can identify chemical signatures of physical evidence and use them to classify or trace the source of it. Headspace chemical forensics is a subdiscipline of chemical forensics, which attempts to detect origins or characteristics of physical evidence from headspace chemical signatures. For the current study, headspace chemical features representing phytocannabinoid profiles were selected based on retention time zones of total ion chromatogram (TIC). With this feature selection, variations of TICs collected from different botanical structures in the same group of marijuana variety was minimized before learning process. This analytical platform of headspace chemical analysis combining with machine learning algorithm provided a nearly non-destructive way to classify marijuana varieties without an analyst's interpretation of chromatograms. The model was able to distinguish marijuana varieties with satisfactory performance. The experimental results demonstrated that headspace chemical analysis approach is a promising analytical platform for chemical forensics. The combination of headspace chemical analysis and machine-learning data processing schemes offers great potential to improve the scope of current forensic analysis of marijuana evidence. (Publisher abstract modified)
Downloads
Similar Publications
- The Effect of Ultraviolet Radiation on Dyed Man-Made Textile Fibers Using UV-Vis Microspectrophotometry (MSP): Technical Aspects on Spectral Alterations in Time
- Human Decomposition Ecology at the University of Tennessee Anthropology Research Facility
- Detecting Various Burial Scenarios In A Controlled Setting Using Ground-penetrating Radar And Conductivity