Pharmaceutical commercials with catchy jingles, smiling actors, and promises of better health rarely mention the time and expense required to get drugs to market. In reality, clinical drug trials can take over a decade and cost billions of dollars.

American University (AU) computer science professor Bei Xiao uses artificial intelligence (AI) to predict the success of future trials, which are critical for assessing a drug’s safety and effectiveness. Xiao’s AI model boasts an 80% accuracy rate, helping to improve efficiency and cut back on wasted resources.

“The whole thing takes a lot of money, especially for [clinical trial] phases,” Xiao said. “If you abandon it early on, you won't have to endure the heartache that will only lead to failure.”

Benefits of Improving Clinical Drug Trials

Real Connections Drive Innovations
 

Xiao’s research is supported by AU’s Translating Research into Action Center, or TRAC, and benefits from direct mentorship through the University of Michigan’s (U-M) Innovation Partnerships.

“AI is transforming many sectors, including healthcare,” said Bryce Pilz, U-M’s Assistant Vice President for Research, Licensing, and Strategic Alliances. “The potential for societal impact is immense, especially in addressing inefficiencies and inequities in healthcare.”

The AU-UMich collaboration—supported by National Science Foundation award #2331399—is helping Xiao engage with the private sector to explore intellectual property considerations, licensing, and patents. Xiao aims to make her AI model open source, amplifying its use and broadening its societal and economic impact.

“I didn't think it was possible for me,” Xiao noted. “I couldn't see how to file a patent to make this open source.”

Impact Extends to the AU Community
 

Xiao’s research may not make it into a flashy commercial, but it can still have a significant ripple effect. Her work with TRAC serves as an example of how AU is building a more robust technology transfer environment on campus that supports innovation and economic development.

Benefits for the AU Community: Strengthened Research Translation

• Bulids AU's capacity to support research translation, technology transfer and innovation.
• Offers faculty and students hands-on learning opportunities to work on real-world problems.
• Provides a model for adapting research translation methods across disciplines, including computer sciences and humanities.

“You could be seeing a lot of innovation activity on campus, amplifying the impact of the great policy-based research,” Xiao said.