Artificial intelligence (AI) in New Psychoactive Substances (NPS) analysis: state-of-art and future perspectives.

Abstract

Analytical toxicology is a discipline of forensic toxicology which applies analytical techniques for the determination of drugs of abuse in biological and non-biological matrices. To this concern, artificial intelligence (AI), particularly machine learning (ML), is innovating analytical toxicology by improving data processing and facilitating the identification of New Psychoactive Substances (NPS). The aim of this review was to explore the current application of AI in this field and to highlight the future perspectives. A literature search was performed in several scientific databases to review articles reporting the implementation of AI models for analytical toxicological purposes. The most frequent applications of these technologies were for compound identification, molecular structure prediction and retention time prediction. AI proved to be a valuable tool for analytical toxicologists for the capability to process large amount of data which are typically obtained by untargeted approaches.