AI-Aided Catheter: New Design Reduces Infection by Preventing Upstream Movement of Bacteria

Catheter-associated urinary tract infections (CAUTIs) are among the most common infections experienced by hospitalized patients, leading to an annual loss of about 30 million US dollars. Designing catheters that can reduce the movement of bacteria offers improvements in managing CAUTI.

(Photo : Pixabay/ Eliola)

Understanding Bacterial Motility

Bacteria are known for being good swimmers, a trait that can be dangerous to human health. In a healthcare setting, bacteria enter the body through catheters connected to the urinary tract. Although these thin tubes are designed to draw fluids out of a patient, bacteria can still propel themselves upstream through the catheter tubes using a unique swimming technique.

A typical trajectory of bacteria alternates between periods of running, where they propel themselves in a straight line, and tumbling, where they change direction randomly. Bacterial motility also involves more complicated dynamics, like enhanced attraction to the surface and collective swarming motion. In shear flows, their run-and-tumble motion can lead to macroscopic upstream swimming.

Anti-Infection Catheters

In a new interdisciplinary project, researchers from the California Institute of Technology designed a novel type of catheter tube that prevents bacteria's upstream mobility without using antibiotics or other chemical antimicrobial approaches. Their findings are reported in the paper "AI-aided geometric design of anti-infection catheters."

In catheter tubes, fluids typically exhibit Poiseuille flow, where fluid moves faster in the center but slower near the wall. As self-propelling microorganisms, bacteria show a unique motion, moving two steps forward along the wall and one step back in the middle. This mechanism produces their forward progress in tubular structures.

Experts suggested tackling this problem with simple geometries. The researchers, led by study first author Tingtao Edmond Zhou, designed tubes with triangular protrusions inside the tubular wall. They found that these geometric structures redirected bacterial movement effectively by propelling them toward the center, where the faster flow pushed them back downstream. The fin-like curvature of the triangle also generated vortices, which further disrupted bacterial motility.

The theoretical modeling was transitioned to practical experimentation with biology expertise. Zhou and his colleagues used 3D-printed catheter tubes and high-speed cameras to monitor bacterial movement. It was found that the tubes with triangular protrusions led to a 100-fold reduction of the upstream bacterial movement.

The experts continued the simulations to determine the most effective triangular obstacle shape that can prevent bacteria's upstream swimming. They fabricated microfluidic channels similar to common catheter tubes with optimized triangular designs to observe the motility of E. coli bacteria under different flow conditions. They discovered that the trajectories of the E. coli within these environments aligned almost perfectly with the simulated predictions.

As the team continued improving the geometric design of the tube, experts in the Anandkumar laboratory provided cutting-edge AI methods called neural operators. This technology accelerates the optimization computations of catheter designs in just a few minutes instead of days. The resulting model proposed adjustments to the geometric designs, optimizing the triangular shapes to obstruct even more bacteria from upstream movement. The final design improved the efficacy of the initial triangular shapes by an additional 5% in simulations.