Uncovering hidden diseases clues with nanoparticles and AI
Imagine Spartan Stadium with 75,000 fans, all wearing green and white jerseys – except for one person in a solid green shirt. It would be difficult to find this person. That's how difficult it is for scientists to find disease markers - called biomarkers - in the blood. And instead of one stadium, researchers have to search for 100,000 stadiums worth of information. To address this challenge, a research team from Michigan State University, working with scientists at Augusta University, the Karolinska Institute and Stanford University, used nanomedicine, artificial intelligence, or AI, and a...
Uncovering hidden diseases clues with nanoparticles and AI
Imagine Spartan Stadium with 75,000 fans, all wearing green and white jerseys – except for one person in a solid green shirt. It would be difficult to find this person. That's how difficult it is for scientists to find disease markers - called biomarkers - in the blood. And instead of one stadium, researchers have to search for 100,000 stadiums worth of information.
To address this challenge, a research team from Michigan State University, working with scientists from Augusta University, the Karolinska Institute and Stanford University, used nanomedicine, artificial intelligence, or AI, and a method for studying cause and effect.
Their goal was to find rare biomarkers for prostate cancer that has spread and a condition called atherosclerosis that causes clogged arteries. Their research results were recently published in theJournal of Chemical Engineering.
Cells affected by disease secrete proteins and other biomolecules into the bloodstream. These proteins provide valuable clues about a patient's health or disease and can be collected and studied. Once identified, they pave the way for significant progress in the development of personalized medical treatments or precision medicine. “
Morteza Mahmoudi, associate professor, Department of Radiology and the Precision Health Program at the MSU College of Human Medicine
How they made the discovery
“The plasma of human blood contains many different proteins, and the rarest proteins are those that contain valuable insights into diseases,” Mahmoudi explained. "To magnify information from the less abundant plasma proteins, we introduced small particles - nanoparticles that cannot be seen with the human eye - into plasma samples. We then used AI and true causality to analyze the results. This is how we identify potential biomarkers for metastatic prostate cancer and atherosclerosis.
“It is the first time that nanomedicine, protein corona, AI and actual causality have been used together to identify a cause of disease,” he added. “We are excited because this discovery has the potential to advance early detection and develop targeted therapies for prostate cancer and atherosclerosis.”
MSU researchers Mohammad Ghassemi, Borzoo Bonakdarpour and Liangliang Sun made significant contributions to this research, Mahmoudi said. Funding sources include the American Heart Association, the U.S. Department of Defense Prostate Cancer Research Program Physician Research Award, the National Cancer Institute, and the National Science Foundation.
Sources:
Guha, A.,et al. (2025). AI-driven prediction of cardio-oncology biomarkers through protein corona analysis. Chemical Engineering Journal. doi.org/10.1016/j.cej.2025.161134.