Trash Talk: How UNO Students Are Teaching Machines to Spot Dangerous Waste Drums

What began as a classroom project in IT Project Management led University of Nebraska at Omaha (UNO) students into the world of machine learning and nuclear safety.

➡️ What’s new:

• A student team at UNO turned an undergraduate IT Project Management class assignment into an interdisciplinary research partnership with a federal agency. The project originally began with students tasked by the U.S. Air Force Office of Special Investigations (OSI) to design outreach strategies for OSI’s cybersecurity services.
• That classroom effort soon sparked a collaboration with Los Alamos National Laboratory (LANL) focused on nuclear safety. Over the summer, Magie Hall, Ph.D., associate professor of Information Systems and Quantitative Analysis in UNO's College of Information Science & Technology (CIST), secured a grant from LANL and developed a prototype machine learning tool to detect rust and dents in hazardous waste storage drums.
• The prototype, currently at Technology Readiness Level 4 (TRL-4), demonstrates the concept in a lab setting and lays out the groundwork for further refinement.

💡 Why it matters:

• Detecting corrosion or dents in nuclear waste drums is essential to national safety. Left unchecked, damage in these drums can lead to dangerous leaks or dangerous contamination events. By developing automated detection tools using machine learning, UNO students are helping to improve early warning systems for hazardous material storage.
• Beyond its environmental implications, this project reflects the impact of experiential learning. Students gained hands-on research experience, built software with national security applications, and collaborated with federal scientists. The project underscores UNO’s commitment to preparing students for high-impact careers in data science, cybersecurity, and engineering.

🔍 Zoom in:

• The team’s prototype uses computer vision to scan images of 55-gallon drums and detect visible signs of rust or impact damage, which are key indicators of structural risk. Reaching TRL-4 validates the concept and opens the door to further refinement and real-world deployment.
• Students involved in the project came from a variety of disciplines, including computer science, software engineering, IT innovation, and cybersecurity. The work culminated in a hands-on summer research experience, including a site visit to LANL in New Mexico where students toured a nuclear research facility and collaborated with LANL engineers and scientists.
• Their efforts were supported and guided by UNO faculty, and the partnership continues to grow, with the potential for additional innovation in environmental safety technologies.

🎤 What they’re saying:

• Magie Hall, Ph.D., CIST Associate Professor of Information Systems and Quantitative Analysis: “This project is the perfect example of what happens when students are given the freedom to explore real-world problems with real-world partners. They rose to the challenge and exceeded expectations.”
• Mohammed Njie, student in AI and Computer Science and Scott Scholar: "Going to New Mexico was such a fun experience, and seeing the inside of a nuclear facility was a once in a lifetime experience. This position gave us direction on what we want to do in the work force, and how real teams operate with deadlines and a user in mind."

🧾 Grant information:

• Funding source: Los Alamos National Laboratory
• Award number: C5702