AI's Role in Stopping Legionnaires' Disease

At a glance

By eliminating weeks of manual effort, TowerScout gives investigators the data they need in minutes — allowing for faster source identification, fewer infections, and saved lives.

AI's Role in Stopping Legionnaires' Disease

The Challenge

Legionnaires' disease, a severe pneumonia caused by Legionella bacteria, can spread through devices such as improperly maintained cooling towers.

  • Legionnaires' disease is deadly — about 1 in 10 people who get sick will die from complicationsA.
  • Traditional investigations, especially for outbreaks not associated with a hotel or hospital, can be complex, labor-intensive, and possess unique challenges related to identification of potential sources of exposure.
  • $835 million lifetime economic burden, including $412 million in productivity losses caused by premature deaths and $21 million in productivity losses caused by absenteeismB.

The Solution

Finding the source of a Legionnaires' disease outbreak can take days or weeks — time that public health agencies can't afford to lose, especially in peak summer months when multiple outbreaks occur. TowerScout uses computer vision, a subfield of AI, to analyze satellite imagery, automatically detecting cooling towers that may harbor Legionella bacteria.

  • Reduces cooling tower identification time by 98% (from 4 hours to 5 minutes).
  • Saved over 280 hours in investigative time annually.

Public Health Impact

  • Enhances outbreak response by precisely identifying cooling tower locations.
  • Helps jurisdictions build cooling tower registries to prevent future outbreaks.
  • Supports near-real-time decision making for public health agencies.
  • Allows public health investigators to focus their time and effort on preventing Legionnaires' disease infections.

Scalable and Ready for the Future

  • Enterprise-ready solution with automated model retraining.
  • Adaptable for new challenges – TowerScout is adaptable to solve different public health problems; the system can be retrained to find any object visible from satellite imagery.
  • Free to use and improve – TowerScout code is expected to be open sourced to the extent possible, so public health partners can use it locally.
  • Locally implemented versions are already in use in Los Angeles County and other jurisdictions.
Print
Content Source:
Office of Public Health Data, Surveillance, and Technology
  1. Centers for Disease Control and Prevention. (n.d.). Legionella (Legionnaires' disease and Pontiac fever). https://www.cdc.gov/legionella/about/index.html
  2. Baker-Goering, M., Roy, K., Edens, C., & Collier, S. (2021). Economic Burden of Legionnaires' Disease, United States, 2014. Emerging Infectious Diseases, 27(1), 255-257. https://wwwnc.cdc.gov/eid/article/27/1/19-1198_article#;