EPIWATCH, an artificial intelligence early-warning system as a valuable tool in outbreak surveillance
Introduction: Utilizing artificial intelligence (AI) for outbreak detection presents a transformative approach to public health surveillance. In challenging environments where traditional surveillance methods may be insufficient, absent or compromised, AI using open-source data can provide epidemic...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | International Journal of Infectious Diseases |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1201971224006544 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Introduction: Utilizing artificial intelligence (AI) for outbreak detection presents a transformative approach to public health surveillance. In challenging environments where traditional surveillance methods may be insufficient, absent or compromised, AI using open-source data can provide epidemic intelligence to inform infectious disease control by identifying early warning signals of disease outbreaks. EPIWATCH is an AI-driven outbreak early-detection and monitoring system, proven to provide early signals of epidemics before official detection by health authorities. Aim: The aim of this study was to evaluate a case study of the utility of open-source epidemic intelligence. Methods & Materials: EPIWATCH reports of outbreaks of unspecified influenza-like illness and pneumonia (syndromic surveillance), together with known causes influenza A and B, SARS-CoV-2, RSV, pertussis (whooping cough), adenovirus and Mycoplasma for August – December of 2022 and 2023, were extracted and summarised to look at trends in respiratory illness in China during a known Mycoplasma pneumonia outbreak in 2023. This was compared to trends in global data for the same period. To investigate the use of EPIWATCH as a valuable surveillance tool in conflict zones, we examined the epidemiology of infectious diseases in Ukraine by utilizing data from EPIWATCH. The analysis focused on infectious disease patterns and syndromes prior to (1 November 2021 to 23 February 2022) and during the conflict (24 February to 31 July 2022). Case numbers for the most frequently reported diseases were compared with official sources, revealing heightened reports of infectious diseases during the wartime period. Results: EPIWATCH reports of outbreaks of unspecified influenza-like illness and pneumonia, together with known causes confirmed an increase in respiratory illnesses in China in 2023 compared to the same period in 2022. In contrast, the same comparison globally shows a decrease in 2023 compared to 2022, indicating China was experiencing an unusual increase in respiratory illness. EPIWATCH detected a peak of pneumonia cases from October to early November 2023, before official diagnosis of Mycoplasma pneumonia as the cause. On 22 November 2023, WHO identified clusters of undiagnosed pneumonia in children's hospitals in Beijing, Liaoning, and other places in China. Discussion: Harnessing AI for outbreak detection holds immense potential for early warning and assessment of unknown epidemics. It is particularly useful in resource-constrained and conflict-affected regions. This research highlights the capability of EPIWATCH as a valuable tool for non-traditional surveillance, enabling resource allocation, response and preparedness efforts to be optimized. Conclusion: With the acceleration of serious epidemics in the last decade, utilizing rapid epidemic intelligence methods and vast open-source data to enable earlier detection of epidemics is essential. AI systems such as EPIWATCH can provide epidemic intelligence to inform infectious disease control and serve as an adjunct to traditional surveillance methods, identifying outbreaks early. |
|---|---|
| ISSN: | 1201-9712 |