By Trish Svoboda

An assistant professor of engineering at Kansas State University is developing an inventive framework designed to exceed traditional methods such as animal testing in predicting disease progression and complications resulting from drug interactions.

Davood B. Pourkargar, assistant professor of chemical engineering at the Carl R. Ice College of Engineering, has secured a $245,000 grant from the National Science Foundation (NSF) to advance the knowledge of drug delivery dynamics. The project focuses on leveraging a multiscale modeling framework, integrating organ-on-a-chip experiments and machine learning.

Entitled “Physics-Informed Machine Learning with Organ-on-a-Chip Data for an In-Depth Understanding of Disease Progression and Drug Delivery Dynamics,” the two-year initiative utilizes NSF funding to address the limitations associated with conventional animal modeling in drug development. It highlights ethical concerns and aims to minimize the use of animal models.

“This innovative, physics-informed machine learning approach enhances organ-on-a-chip experiments, streamlining the preclinical process, improving drug efficacy and minimizing side effects,” Pourkargar said. “Ultimately, the project accelerates drug discovery, supports personalized treatments, and fosters more efficient, affordable health care while reducing reliance on animal testing.”

The hybrid model under development is expected to surpass standard machine learning-based models by accurately extrapolating and interpolating organ-on-a-chip data. It offers improved analytical simplicity, interpretability, and reduces the dependency on extensive training samples.