BYLINE: Jamie Oberdick

UNIVERSITY PARK, Pa. — To enhance biosensor development via artificial intelligence (AI) and offer STEM education opportunities to K-12 students from underserved communities, the U.S. National Science Foundation recently awarded researchers at Penn State a three-year, $1.5 million grant.  

The project is supported through the Designing Materials to Revolutionize and Engineer our Future (DMREF) program and seeks to address a longstanding challenge in the biosensor research: the inability to systematically identify the best materials for biosensors detecting specific molecules. These small devices are essential tools in diagnosing diseases, detecting harmful substances and identifying environmental contaminants. However, the process of developing biosensors is often time-consuming and inefficient.  

“In the field of engineering biosensing materials, much of the research work has been done through trial and error,” said Aida Ebrahimi, Thomas and Sheila Roell Early Career Associate Professor of Electrical Engineering and of Biomedical Engineering, and the principal investigator of the grant. “While there have been promising results, there is no comprehensive guide to tell us which materials to use for detecting specific molecules, so researchers typically test a range of options until they find one that works.” 

This problem is especially prevalent with two-dimensional (2D) materials — ultra-thin materials that are just an atom or a few atoms thick and are widely regarded for their potential in biosensing applications.  

The research team aims to change that by developing a computationally guided approach that will make the process of material selection more efficient. At the heart of the project is a close collaboration between experimentalists and computational scientists, Ebrahimi said. Ebrahimi and Maurico Terrones, George A. and Margaret M. Downsbrough Head of the Department of Physics at Penn State, will focus on experimentation, while Humberto Terrones, Rayleigh Endowed Chair Professor at Rensselaer Polytechnic Institute, and Trevor David Rhone, assistant professor of physics, applied physics, and astronomy at Rensselaer Polytechnic Institute, will focus on AI and computational modeling. The researchers will create a feedback loop for the experimental data and AI-driven predictions to inform each other.  

“We’ll start with theoretical calculations and some preliminary experimental data, which will then be fed into the AI models,” Ebrahimi said. “The AI will suggest which material to try next, or how to modify the material — whether by doping it with another element, which involves adding impurities to modify properties, or adjusting its defect chemistry, accomplished by modifying types or concentrations of imperfections. We will then fabricate the sensors and perform multimodal characterization to create features for the AI models.” 

This closed-loop process promises to dramatically accelerate the development of biosensors, according to the researchers. 

“Rather than randomly testing materials, the AI will guide us, telling us not just what material to use but also how to modify it for optimal performance,” Ebrahimi said.  

While the initial focus of the project is on biosensors for detecting stress molecules — such as neurotransmitters which  play a crucial role in the functioning of the nervous system — the framework developed could be applied to many other areas, Ebrahimi said.  

“Our goal is to create an infrastructure that can be extended to other types of materials and molecules/targets,” Ebrahimi said. “This could eventually lead to biosensors that can detect contaminants in water, harmful substances in food or environmental pollutants.” 

Beyond its scientific goals, the project also aims to create summer camps for K-12 students, with a special emphasis on underrepresented groups in STEM fields.  

“We’re organizing four summer camps, including two at Penn State and two at Rensselaer Polytechnic Institute, that will focus on introducing students to materials science, nanotechnology, biosensors and artificial intelligence” Ebrahimi said. “These camps will include hands-on activities designed to inspire young students, particularly girls and underrepresented minorities, to pursue careers in science and engineering.” 

In addition to the K-12 outreach, the project will also support undergraduate research, curriculum innovation, and workforce development. The team plans to collaborate with industry through an advisory board and offer internships for students.  

“We want to make sure that the next generation of engineers and scientists are not only well-trained but also have opportunities to connect with industry and explore career paths in STEM,” Ebrahimi said. 

Diversity, equity and inclusion are core values of the project. All four principal investigators are committed to fostering a more inclusive STEM community by partnering with relevant organizations.  

“We will work with organizations like the Hispanic Society for Professional Engineers, National Society of Black Engineers, National Society of Black Physicists and the Society for the Advancement of Chicanos/Hispanics and Native Americans in Science,” Ebrahimi said. “We want to ensure that our outreach efforts are meaningful and that we’re providing real opportunities for underrepresented groups.” 

The researchers said they also hope to have a broader impact on the scientific community by creating an open-access database. 

“We’re building a cloud-based database where researchers can access the results of our experiments and AI models,” Ebrahimi said. “This resource will be available to the entire scientific community, helping to accelerate materials discovery in biosensing and other fields.” 

In collaboration with the Air Force Research Laboratory, the team will also explore scalable, low-cost methods for synthesizing 2D materials, which they said could further enhance the impact of their work.  

“We’re looking at synthesis methods that could be scaled up for industrial use, which is important for making this technology accessible and affordable,” Ebrahimi notes. 

Ultimately, the researchers said they believe this project could transform biosensor technology and how materials development for various applications.  

“Our hope is that this research will not only lead to better biosensors but also create a new paradigm for materials discovery, where AI plays a central role in guiding experimental efforts,” Ebrahimi said. 

Source