Tackling Backlogs using Artificial Intelligence
“[…] thousands of rape kits awaiting testing in 2022.” (USAFacts, July 3, 2023)
“I think the labs are doing a spectacular job, but they’re drinking from a fire hose.” (Scott Berkowitz, RAINN)
The forensic examination of biological specimens obtained in cases of purported sexual violence, in the United States frequently referred to as "sexual assault evidence collection kits" or colloquially as "rape kits", constitutes a complex and temporally demanding procedure. This intricate process often results in procedural bottlenecks that impede the resolution of alleged sexual assault cases, thereby contributing to an escalating accumulation of unprocessed evidence in forensic laboratories. The multifaceted nature of these analyses, coupled with resource constraints and the meticulous protocols required for maintaining evidence integrity, has led to a significant backlog in many jurisdictions, potentially delaying justice and affecting the efficacy of criminal investigations in this sensitive domain.
The backlog of sexual assault evidence kits in the United States exemplifies a significant challenge in forensic analysis, where crucial evidence frequently remains unexamined. As of 2022, the landscape of untested sexual assault kits varied significantly across states. Maryland reported the highest number of backlogged kits in law enforcement agencies (3,599), while North Carolina had the most in crime labs (9,045)1. This data illustrates the ongoing nature of the backlog problem, despite efforts to reduce it.
To address the growing backlog of sexual assault kits, the U.S. Department of Justice has emphasized the need for innovative solutions, particularly automation in processing these kits. Automated microscopy has emerged as a crucial tool to handle the increasing sample volume efficiently.
MetaSystems, with its extensive experience in automated slide imaging, has been at the forefront of this technological advancement. For nearly four decades, the company has been developing and refining automated imaging solutions, collaborating with forensic laboratories globally to adapt these technologies to their specific requirements.
The company's customized workflows are designed to enhance the efficiency of crime labs' daily operations. By implementing automated sperm detection systems, forensic laboratories can significantly reduce the time and effort required for manual microscopic examination of sexual assault evidence.
Sperm Detection
With our assistance, forensic laboratories have employed Deep Neural Networks (DNNs) to identify sperm cells in cases of alleged sexual assault. Imaging installations operated by the Metafer Platform Software support unattended high-solution image capture of up to 800 slides with secure data handling, full traceability, and reliable documentation.