National Forensic Science Week

The National Forensic Science Week celebrated its 12th anniversary last week. The event is organized by a committee from various forensic associations, under the Consortium of Forensic Science Organizations (CFSO). Their mission is to raise awareness and encourage participation through outreach to educational institutions, government bodies, forensic agencies, media, and organizations.

As the National Forensic Science Week came to an end, we would like to take the opportunity to highlight the advantages of our Metafer Customization Package Sperm Detection.

Automation Is Key to Reducing Backlogs

The labor-intensive search for sperm cells in forensic samples often leads to bottlenecks in solving cases of suspected sexual assault, contributing to a growing backlog in crime labs. A glaring example of this is the backlog of rape samples in the U.S. According to a 2019 report by the U.S. Government Accountability Office, the number of backlogged DNA analyses rose from 91,000 to 169,000 between 2011 and 2017. To combat this, the U.S. Department of Justice has called for innovation, including the use of automation in the processing of sexual assault kits.

MetaSystems has more than 35 years of experience in automated slide imaging and has adapted this technology to the needs of forensics. Our Metafer Customization Package Sperm Detection can optimize slide scanning workflows to detect sperm on slides to speed up rape kit investigations.

Metafer Customization Package Sperm Detection

At the intersection of precision and automation, MetaSystems' Metafer software controls microscope and accessory hardware to capture digital images. The Customization Package Sperm Detection is adapted to the visual detection of sperm in forensic samples by integrating intelligent software functions and advanced microscopy hardware.

After scanning, all detected objects are presented in a gallery and scored based on the predicted likelihood of being sperm, aiding forensic teams in sorting during the review. Expert assessments are made seamlessly on-screen, with results securely stored in the Neon case and image management tool.

Deep Learning Advances Brightfield Techniques for Sperm Detection

In the field of forensic analysis, staining methods like Christmas Tree and Baecchi stain have long been indispensable tools. Now, leveraging the latest advancements in artificial intelligence, we have integrated Deep Learning techniques into our Metafer Customization Package Sperm Detection to further enhance confirmatory sperm detection.

At MetaSystems, we employ Deep Neural Networks (DNNs) to tackle complex image processing tasks, like identifying potential spermatozoa in forensic specimens. DNNs learn from example images that are labeled by human experts. In our internal validation study, our fully trained DNN correctly classified 99% of images showing spermatozoa stained with Christmas Tree staining.

Versatility in Automated Microscopy Imaging

we can configure a microscope operated by our Metafer software with the two illumination modes brightfield and fluorescence to support three different staining techniques, i.e., Christmas Tree staining (Nuclear Fast Red/Picroindigocarmine), Baecchi stain (acid fuchsin and methylene blue), and commercial fluorescent staining kits.

If capacity requires 24/7 microscopy operation with the ability to prioritize urgent slides, an automated slide feeder can be added to the scanning system. This allows for unattended slide scanning overnight or on weekends. With a scalable capacity from 8 up to 800 slides being automatically scanned, we support our customers in high-throughput applications.

Ongoing Collaborations

One indication that a condom was used in a sexual assault is to detect starch particles using light microscopy. Most latex condoms are coated with a thin layer of corn starch to prevent the latex sticking to itself, which makes the condom easier to unroll. When polarized light is used, the starch granules show the typical “Maltese cross” birefringence pattern when examined under the microscope. Our colleagues are currently working on the incorporation of artificial intelligence to support the detection of starch granules in an automated microscope solution, which could be a promising addition to the DNN-supported sperm detection in forensic samples that we can already customize.

Want to know more about our products and services?