Machine Learning Tool Detecting NBOMe Drugs of Abuse based on Geometrical Descriptors

Autor: Steluta Gosav, Adelina Ion, Teodora Gosav, Mirela Praisler
Rok vydání: 2020
Předmět:
Zdroj: 2020 International Conference on e-Health and Bioengineering (EHB).
DOI: 10.1109/ehb50910.2020.9279876
Popis: Drug trafficking is one of the most serious health threats public, worldwide. Thus, both the trend of globalization and the galloping development of technology, which have led to the diversification of trade links and the elimination of border controls, are exploited by criminal organizations in order to feed and control the illicit market of drugs. During the last years, a new type of toxic phenethylamine derivatives, referred to as NBOMe, has been detected on the black market. This study compares a series of artificial neural networks (ANNs) that screen for these new psychoactive drugs based on their geometrical descriptors. The best performing of these machine learning tools has been identified based on a series of figures of merit assessing their efficiency in assessing the class identity of an unknown chemical.
Databáze: OpenAIRE