Advancing Forensic Science through Artificial Intelligence: A Systematic Review

Abstract

Abstract

AI is the ability of a system to think based on experience. It simulates human reasoning, which is currently transforming forensic science and the criminal justice system. This systematic review was conducted to estimate the increasing trend of incorporating AI in the improvement of forensic sciences. The findings of the present study indicated that contemporary forensic science faces numerous challenges, including overwhelming volumes of data, trace evidence embedded in chaotic and complex environments, traditional laboratory structures and limited specialist knowledge. Together, these factors can hinder investigations and increase the risk of miscarriages of justice. Moreover, machine learning and deep learning are two types of AI that can help address these challenges. In a variety of forensic fields, neural networks and case-based reasoning are used to provide errorless, objective, and reproducible results. AI has recently been integrated into almost every major field of forensic sciences, using various approaches such as data analysis, pattern recognition, image processing, computer vision, data mining, statistical analysis and probabilistic methods, computational and mathematical methods, and graphical modeling. Furthermore, AI assists forensic experts and investigators by articulating logical evidence in Deoxyribonucleic acid (DNA) analysis, fingerprint analysis, footprint analysis, drug analysis, ballistics analysis, voice analysis, facial recognition to identify individuals in images and videos, three-dimensional (3D) reconstructions of crime scenes, autopsy, and cause of death determination. As a result, despite its hurdles, the incorporation of AI into forensic analysis constitutes a significant leap in the field of crime investigation, opening up new avenues for efficiency, accuracy, and scalability in solving complicated cases and providing justice.