Movement of microswimmers is key to targeted drug delivery

The research is being carried out by Ebru Demir, an assistant professor of mechanical engineering and mechanics in Lehigh University’s P.C. Rossin College of Engineering and Applied Science.

“We know that whenever a swimmer has a neighbour, it swims differently,” Demir said in a statement. “Birds fly in a V formation because it’s more efficient and it saves them energy. But for a group of microswimmers, we don’t know what the best formation looks like.”

Demir recently received funding through the US National Science Foundation’s Faculty Early Career Development (CAREER) Program for her research combining artificial microswimmers with machine learning to build Smart Artificial Microswimmers (SAMs).

By embedding AI into centimetre-scale robotic swimmers and comparing their behaviour with predictions from simulations, her project aims to uncover the underlying physics that governs their movement in complex fluid environments. 

MORE FROM ARTIFICIAL INTELLIGENCE

To study how the swimmers move in Newtonian and non-Newtonian fluids, Demir will insert microcontrollers that run reinforcement learning algorithms into 3D-printed autonomous robots that are 10-20cm in size.

Fluids such as blood have non-Newtonian properties, so their viscosity changes depending on the stress applied by the swimming body. Demir will make and use fluids with similar properties in experiments to verify the results of her simulations. 

“These swimmers will each contain an AI brain that will give them decision-making capabilities, so they can determine for themselves how to swim better alone, and how to swim better with three or five or 10 companions,” said Demir. “It will also be interesting to see if their behaviour changes if they are allowed to cooperate and share information.”

Demir continued: “As the SAMs swim, they are constantly interacting with the environment, running that algorithm, and recalculating where they are in that formation and how fast or how efficiently they are moving. The goal is to find the best strategy for the cluster to swim together in a manner that’s both fast and efficient.”

The long-term aim is to insert micron-size (or centimetre-size for larger vessels) artificial swimmers into the body where they could travel through the veins to deliver targeted chemotherapy drugs, or break up a blood clot without the need to thin a patient’s blood. They could also assist otherwise normal, high-quality human sperm in fertility treatments.

“In those cases, the sperm has good genetics, but maybe its tail is compromised, and so the swimmers could push the sperm toward the egg,” she said. “And that has been demonstrated by other researchers to work in a lab environment.”