Featured Post from University of Delaware

AI-powered model predicts post-concussion injury risk in college athletes

Apr 16, 2025

3 min

CaptionResizeWrap TextRemove


Athletes who suffer a concussion have a serious risk of reinjury after returning to play, but identifying which athletes are most vulnerable has always been a bit of a mystery, until now.


Using artificial intelligence (AI), University of Delaware researchers have developed a novel machine learning model that predicts an athlete’s risk of lower-extremity musculoskeletal (MKS) injury after concussion with 95% accuracy. A recent study published in Sports Medicine details the development of the AI model, which builds on previously published research showing that the risk of post-concussion injury doubles, regardless of the sport. The most common post-concussive injuries include sprains, strains, or even broken bones or torn ACLs.


“This is due to brain changes we see post-concussion,” said Thomas Buckley, professor of kinesiology and applied physiology at the College of Health Sciences.


These brain changes affect athletes’ balance, cognition, and reaction times and can be difficult to detect in standard clinical testing.


“Even a minuscule difference in balance, reaction time, or cognitive processing of what’s happening around you can make the difference between getting hurt and not,” Buckley said.


How AI is changing injury risk assessment


Recognizing the need for enhanced injury reduction risk tools, Buckley collaborated with colleagues in UD’s College of Engineering, Austin Brockmeier, assistant professor of electrical and computer engineering, and César Claros, a fourth-year doctoral student; Wei Qian, associate professor of statistics in the College of Agriculture and Natural Resources; and former KAAP postdoctoral fellow Melissa Anderson, who’s now an assistant professor at Ohio University. To assess injury risk, Brockmeier and Claros developed a comprehensive AI model that analyzes more than 100 variables, including sports and medical histories, concussion type, and pre- and post-concussion cognitive data.


“Every athlete is unique, especially across various sports,” said Brockmeier. “Tracking an athlete’s performance over time, rather than relying on absolute values, helps identify disturbances, deviations, or deficits that, when compared to their baseline, may signal an increased risk of injury.”


While some sports, such as football, carry higher injury risk, the model revealed that individual factors are just as important as the sport played.


“We tested a version of the model that doesn’t have access to the athlete’s sport, and it still accurately predicted injury risk,” Brockmeier said. “This highlights how unique characteristics—not just the inherent risks of a sport—play a critical role in determining the likelihood of future injury,” said Brockmeier.


The research, which tracked athletes over two years, also found that the risk of MSK injury post-concussion extends well into the athlete’s return to play.


“Common sense would suggest that injuries would occur early in an athlete’s return to play, but that’s simply not true,” said Buckley. “Our research shows that the risk of future injury increases over time as athletes compensate and adapt to small deficits they may not even be aware of.”


The next step for Buckey’s Concussion Research Lab is to further collaborate with UD Athletics’ strength and conditioning staff to design real-time interventions that could reduce injury risk.


Beyond sports: AI’s potential in aging research


The implications of the UD-developed machine-learning model extend far beyond sports. Brockmeier believes the algorithm could be used to predict fall risk in patients with Parkinson’s disease.


Claros is also exploring how the injury risk reduction model can be applied to aging research with the Delaware Center for Cognitive Aging.


“We want to use brain measurements to investigate whether baseline lifestyle measurements such as weight, BMI, and smoking history are predictive of future mild cognitive impairment or Alzheimer’s disease,” said Claros.


To arrange an interview with Buckley, email UD's media relations team at MediaRelations@udel.edu

Powered by

You might also like...

Check out some other posts from University of Delaware

2 min

Discovery pinpoints potential Achilles’ Heel in HIV, opening new frontier in drug development

Scientists have long known that detecting HIV early is crucial in slowing and treating the virus. During the acute stage of infection, a single human cell can produce as many as 10,000 new HIV particles. A discovery led by the University of Delaware's Juan Perilla offers hope: A new drug target early in the virus's life cycle that could save millions of lives. In the surprising discovery, published Feb. 18 in Nature, Perilla and collaborators in the U.S. and the United Kingdom, have revealed a previously unknown role for the viral protein integrase. Scientists already knew that integrase helps HIV insert itself into human DNA. But this new study provides the first direct evidence that integrase plays a critical structural role earlier on in HIV’s life cycle — when the virus matures into an infectious force. Using high-resolution cryo-electron microscopy (cryo-EM), the research team – which also included UD doctoral student Juan S. Rey – found that integrase proteins form gluey filaments that line the inside of the capsid. Each segment of the filament slots neatly into the capsid’s hexagon-shaped tiles, while gripping tightly to HIV’s RNA genome. This zipper-like arrangement organizes and packs the virus, preparing it to hijack a cell and start making copies of itself. “Integrase plays a structural role inside the HIV capsid — nobody expected that,” Perilla said. “This protein forms filaments that anchor the RNA to the capsid. Without these filaments, the virus is non-infective.” Seeing inside HIV is no small feat. The capsid is only about 120 nanometers wide —roughly 1/800th the thickness of a human hair. It is incredibly small, fragile, densely packed and constantly changing, Perilla said. To reveal its hidden architecture, the researchers relied on deep collaboration and a combination of sophisticated microscopy, molecular modeling and experimentation. Read more about the science behind the study here. “The thing with HIV is that people are chronically living with it,” Perilla said. “Treatments are effective, but patients always need new therapeutics. We want to help develop the next generation of inhibitors and hope to have a significant contribution.” To reach Perilla directly and arrange an interview, visit his profile page. Interested reporters can also email MediaRelations@udel.edu.

3 min

Rethinking AI in the classroom: A literacy-first approach to generative technology

As schools nationwide navigate the rapid rise of generative artificial intelligence, educators are searching for guidance that goes beyond fear, hype or quick fixes. Rachel Karchmer-Klein, associate professor of literacy education at the University of Delaware, is helping lead that conversation. Her latest book, Putting AI to Work in Disciplinary Literacy: Shifting Mindsets and Guiding Classroom Instruction, offers research-based strategies for integrating AI into secondary classrooms without sacrificing critical thinking or deep learning. Here is how she is approaching the complex topic.  Q: Your new book focuses on AI in disciplinary literacy. What is the central message? Karchmer-Klein: Rather than positioning AI as a shortcut or replacement for student thinking, the book emphasizes a literacy-first approach that helps students critically evaluate, interrogate, and apply AI-generated information. This is important because schools and universities are grappling with rapid AI adoption, often without clear guidance grounded in learning theory, literacy research, or classroom practice. Q: What inspired this research? Karchmer-Klein: The book grew directly out of my work with preservice teachers, practicing educators, and school leaders who were asking practical but complex questions about AI: How do we use it responsibly? How do we prevent over-reliance? How do we teach students to question what AI produces? I also saw a gap between public conversations about AI which often focused on fear or efficiency and what teachers actually need: research-informed strategies that support deep learning. My long-standing research in digital literacies provided a natural foundation for addressing these questions. Q: What are some of the key findings from your work? Karchmer-Klein: AI is most effective when it is embedded within strong instructional design and disciplinary literacy practices, not treated as a stand-alone tool. The research and classroom examples illustrate that AI can support student learning when it is used to prompt reasoning, reveal misconceptions, provide feedback for revision, and encourage multiple perspectives. Another important development is the emphasis on teaching students to evaluate AI outputs critically by recognizing bias, inaccuracies, and limitations, rather than assuming correctness. Q: How could this work impact schools, teacher education programs and the broader public? Karchmer-Klein: For educators, this work provides concrete, evidence-based literacy strategies coupled with AI in ways that strengthen, not dilute, student thinking. For teacher education programs and school districts, it offers a research-based framework for professional development and policy conversations around AI use. More broadly, the work speaks to a public concern about how emerging technologies are shaping learning, helping to reframe AI as something that requires human judgment, ethical consideration, and strong literacy skills to use well. ABOUT RACHEL KARCHMER-KLEIN Rachel Karchmer-Klein is an associate professor in the School of Education at the University of Delaware where she teaches courses in literacy and educational technology at the undergraduate, graduate, and doctoral levels. She is a former elementary classroom teacher and reading specialist. Her research investigates relationships among literacy skills, digital tools, and teacher preparation, with particular emphasis on technology-infused instructional design. To speak with Karchmer-Klein further about AI in literacy education, critical evaluation of AI-generated content and teacher preparation in the era of generative AI, reach out to MediaRelations@udel.edu.

2 min

How AI can improve poor leadership writing and boost productivity

Poor written communication from leaders can create the kind of confusion it intended to avoid. University of Delaware career expert Jill Gugino Panté suggests using AI to sharpen emails, clarify expectations and reduce unnecessary calls. Getting through to employees with strong messaging can boost productivity by saving time and reducing unwanted meetings, she says. Panté, director of UD's Lerner Career Services Center, says that good leadership writing should be direct and outcome-driven, with no fluff, and offered the following advice for improvement. ✅ Don’t bury the lead. Start with what decision needs to be made, what action is required, and the deadline. If your writing doesn’t reduce ambiguity, it’s going to add to it. Vague communication can create interpretation gaps which, in turn, can create more meetings. When ownership isn’t defined, decisions aren’t documented, or outcomes aren’t clear, teams default to “Let’s hop on a call.” Meetings then become the fallback for unclear thinking. ✅ Generative AI can be a powerful clarity tool if it’s used intentionally. When used well, it can sharpen your ask and structure communication for action. The key is prompting it to refine your message, not just polish it. Leaders can use prompts like: • “Rewrite this message so the action, owner, deadline, and success metrics are explicitly stated" • “What assumptions or ambiguities exist in this message?” ✅ Good writing can replace unnecessary meetings. If communication is not direct, outcome-driven, and structured for action, it will cost you time somewhere else. Here are some practical actions that leaders can make in their writing: • Start with the Ask - Be explicit about what decision or action is needed. Don’t make people search for it. • Define Outcomes - Clarify deliverables, timelines, budgets and state what success looks like. • Clarify Ownership - Identify who is responsible for the request. • Document Decisions - Write down what has been decided and reiterate next steps, owners, and deadlines. To connect with Panté directly and arrange an interview, visit her profile and click on the "contact" button. Interested media can also send an email to MediaRelations@udel.edu.

View all posts
Powered by