Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published January 5, 2024 | Published
Journal Article Open

AI-aided geometric design of anti-infection catheters


Bacteria can swim upstream in a narrow tube and pose a clinical threat of urinary tract infection to patients implanted with catheters. Coatings and structured surfaces have been proposed to repel bacteria, but no such approach thoroughly addresses the contamination problem in catheters. Here, on the basis of the physical mechanism of upstream swimming, we propose a novel geometric design, optimized by an artificial intelligence model. Using Escherichia coli , we demonstrate the anti-infection mechanism in microfluidic experiments and evaluate the effectiveness of the design in three-dimensionally printed prototype catheters under clinical flow rates. Our catheter design shows that one to two orders of magnitude improved suppression of bacterial contamination at the upstream end, potentially prolonging the in-dwelling time for catheter use and reducing the overall risk of catheter-associated urinary tract infection.

Copyright and License

© 2024 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution License 4.0 (CC BY).


We thank J. P. Marken and R. Murray for providing the bacteria strains. We thank D. J. Anderson for providing the microscope used in this study. We thank P. Arakelian for assistance in 3D printing. We thank A. Ghaffari for training on microfluidic fabrication. We thank C. Zhang and N. Sim for helping with bacterial colony counting and trajectory classification. T.Z. and X.W. thank M. Chen and W. Miao for discussions.


This work was supported by the following: the Donna and Benjamin M. Rosen Bioengineering Center Pilot Research Grant (J.F.B. and C.D.), the Heritage Medical Institute at Caltech (C.D.), and the National Science Foundation, Center to Stream Healthcare in Place (C2SHIP), award no. 2052827 (C.D.). D.Z.H. is supported by the generosity of Eric and Wendy Schmidt by recommendation of the Schmidt Futures program. Z.L. is supported in part by the PIMCO Fellowship and Amazon AI4Science Fellowship. A.A. and P.W.S. are supported by Bren Professorships.


X.W., T.Z., P.W.S., and C.D. designed experiments. X.W. and T.Z. performed experiments and analyzed data. T.Z. and Z.P. performed simulations. D.Z.H. and Z.L. designed the AI model and performed optimization. A.A. conceptualized and planned the AI framework. T.Z., J.F.B., and C.D. conceived the project. P.W.S. and C.D. supervised the project. All authors discussed the results and contributed to the manuscript writing.

Data Availability

All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Supplementary data for the optimization process are included in the following link: https://data.caltech.edu/records/mdj7m-ajv14.

Conflict of Interest

California Institute of Technology (Caltech) has a patent pending related to the discoveries in this manuscript. Patent status: Pending. Name of organization issuing patent: The United States Patent and Trademark Office (USPTO). All authors are inventors. Filing date: 13 March 2023. Serial number: 63/451,788. The authors declare that they have no other competing interests.


Files (1.5 kB)
Name Size Download all
485 Bytes Preview Download
498 Bytes Preview Download
482 Bytes Preview Download

Additional details

January 10, 2024
February 12, 2024