Automated cooling tower detection through deep learning for Legionnaires' disease outbreak investigations: a model development and validation study
- PMID: 38906615
- DOI: 10.1016/S2589-7500(24)00094-3
Automated cooling tower detection through deep learning for Legionnaires' disease outbreak investigations: a model development and validation study
Abstract
Background: Cooling towers containing Legionella spp are a high-risk source of Legionnaires' disease outbreaks. Manually locating cooling towers from aerial imagery during outbreak investigations requires expertise, is labour intensive, and can be prone to errors. We aimed to train a deep learning computer vision model to automatically detect cooling towers that are aerially visible.
Methods: Between Jan 1 and 31, 2021, we extracted satellite view images of Philadelphia (PN, USA) and New York state (NY, USA) from Google Maps and annotated cooling towers to create training datasets. We augmented training data with synthetic data and model-assisted labelling of additional cities. Using 2051 images containing 7292 cooling towers, we trained a two-stage model using YOLOv5, a model that detects objects in images, and EfficientNet-b5, a model that classifies images. We assessed the primary outcomes of sensitivity and positive predictive value (PPV) of the model against manual labelling on test datasets of 548 images, including from two cities not seen in training (Boston [MA, USA] and Athens [GA, USA]). We compared the search speed of the model with that of manual searching by four epidemiologists.
Findings: The model identified visible cooling towers with 95·1% sensitivity (95% CI 94·0-96·1) and a PPV of 90·1% (95% CI 90·0-90·2) in New York City and Philadelphia. In Boston, sensitivity was 91·6% (89·2-93·7) and PPV was 80·8% (80·5-81·2). In Athens, sensitivity was 86·9% (75·8-94·2) and PPV was 85·5% (84·2-86·7). For an area of New York City encompassing 45 blocks (0·26 square miles), the model searched more than 600 times faster (7·6 s; 351 potential cooling towers identified) than did human investigators (mean 83·75 min [SD 29·5]; mean 310·8 cooling towers [42·2]).
Interpretation: The model could be used to accelerate investigation and source control during outbreaks of Legionnaires' disease through the identification of cooling towers from aerial imagery, potentially preventing additional disease spread. The model has already been used by public health teams for outbreak investigations and to initialise cooling tower registries, which are considered best practice for preventing and responding to outbreaks of Legionnaires' disease.
Funding: None.
Copyright © 2024 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license. Published by Elsevier Ltd.. All rights reserved.
Conflict of interest statement
Declaration of interests FN owns stock in Google and Amazon. All other authors declare no competing interests. The findings and conclusions of this study are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
Similar articles
-
Outbreaks of Legionnaires' Disease and Pontiac Fever 2006-2017.Curr Environ Health Rep. 2018 Jun;5(2):263-271. doi: 10.1007/s40572-018-0201-4. Curr Environ Health Rep. 2018. PMID: 29744757 Review.
-
Legionnaires' Disease Outbreaks and Cooling Towers, New York City, New York, USA.Emerg Infect Dis. 2017 Nov;23(11):1769-76. doi: 10.3201/eid2311.161584. Emerg Infect Dis. 2017. PMID: 29049017 Free PMC article.
-
Regulating Cooling Towers to Prevent Outbreaks of Legionnaires' Disease.Public Health Rep. 2017 Mar/Apr;132(2):133-135. doi: 10.1177/0033354916689612. Epub 2017 Feb 1. Public Health Rep. 2017. PMID: 28147210 Free PMC article. No abstract available.
-
A Large Community Outbreak of Legionnaires' Disease Associated With a Cooling Tower in New York City, 2015.Public Health Rep. 2017 Mar/Apr;132(2):241-250. doi: 10.1177/0033354916689620. Epub 2017 Jan 31. Public Health Rep. 2017. PMID: 28141970 Free PMC article.
-
Existence and control of Legionella bacteria in building water systems: A review.J Occup Environ Hyg. 2017 Feb;14(2):124-134. doi: 10.1080/15459624.2016.1229481. J Occup Environ Hyg. 2017. PMID: 27624495 Review.
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Medical
Miscellaneous