Pathogen reduction in an endoscopy unit using AI-enabled autonomous UV-C disinfection

Am J Infect Control. 2026 Feb;54(2):158-162. doi: 10.1016/j.ajic.2025.09.006. Epub 2025 Sep 11.

Abstract

Background: Microbial bioburden has been identified as a contributor to health care-associated infections. Disinfection with ultraviolet-C (UV-C) light can minimize bioburden, but targeted disinfection can be labor-intensive. We evaluated the effectiveness of wall-mounted autonomous and targeted UV-C disinfection device powered by artificial intelligence in reducing bioburden in a clinical setting.

Methods: Two endoscopy rooms were evaluated, a control room with standard disinfection/cleaning measures and another with 2 wall-mounted autonomous UV-C devices. To measure their impact on pathogenic bioburden levels, swab sampling was conducted on 10 preselected high-touch surfaces in each room over a period of 4 weeks and analyzed for microbial colony counts.

Results: Autonomous, targeted UV-C disinfection inactivated pathogens within 20 to 60 seconds from a distance of 6 to 8 ft. Longer UV-C exposure times were utilized to achieve a consistent level of pathogen inactivation. The autonomous UV-C room had a 99.7%, 84.3%, and 93.8% bioburden reduction compared to the control room (weeks 1, 2, and 4). Cumulative bioburden was 93.3% lower than that measured in the control room.

Conclusions: These data demonstrate that this novel, autonomous, and targeted UV-C disinfection approach is associated with effective surface decontamination and highlight the potential for this approach for broader use in healthcare settings.

Keywords: Artificial intelligence; Endoscopy; Health care-associated infections; Infection control; Ultraviolet-C.

MeSH terms

  • Artificial Intelligence*
  • Colony Count, Microbial
  • Cross Infection / prevention & control
  • Disinfection* / methods
  • Endoscopy*
  • Environmental Microbiology*
  • Humans
  • Ultraviolet Rays*