Using Artificial Intelligence in the Forensic Science for the Analysis of Microparticles: A Systematic Review

Abstract

Technological advances in different industries have a tremendous impact on various aspects of human activities, including criminal activities. More complex and well-organized crimes often leave no room for the traditional analytical techniques, and require analysis of the tiniest pieces of the evidence – microobjects. Detection and study of these pieces of information obviously are time-consuming and demands more sophisticated equipment. This systematic review estimated the current status of the application of artificial intelligence systems in analysis of a specific type of the forensic evidence such as microparticles. Analysis of 27 articles extracted according to the PRISMA guidelines confirms the rationale behind using the AI for forensic investigation mainly to achieved automation of the most laborious aspects of the evidence investigation: image processing, matching a piece of the evidence to the created database and identification of the evidence. The AI technologies assist in identification of the victim or suspect personality thorough AI-assisted analysis of the DNA and RNA from the blood, saliva, urine; time of death and place of death via AI-assisted investigation of soil and fabric traces, and specific microbiome; tracking abusive substances; identification the cause of fires. The AI mainly serves as assistant to the convenient forensic methods, such as microscopy and spectroscopy, to process a big amount of data generated by the traditional techniques or to enhance the outcomes of the traditional techniques, such as image processing. The types of the AL the most widely used in the forensic science are machine learning algorithm.