(Representative image) News reports in print, electronic and online media are scanned for unusual health events using a media scanning and verification tool under India’s Integrated Disease Surveillance Program and then forwarded to authorities for further action if necessary. | Photo credit: istock.com
An artificial intelligence tool deployed by the National Center for Disease Control (NCDC) in 2022 may have helped send more than 5,000 real-time infection outbreak alerts to health authorities since it was installed, according to a study.
The Health Sentinel tool, developed by WadhwaniAI, a New Delhi-based healthcare AI solutions provider, could have helped reduce 98 percent of manual workload and enable faster outbreak detection and proactive public health response, according to findings published as a pre-printed paper that have not yet been peer-reviewed.
Almost 200 countries are legally required by the International Health Regulations (IHR) to operate a national disease surveillance system. The IHR and the World Health Organization work together to protect global health security.
News reports in print, electronic and online media are scanned for unusual health events using a media scanning and verification tool under India’s Integrated Disease Surveillance Program (IDSP) and then forwarded to authorities for further action if necessary.
Health Sentinel scanned daily media reports and news articles in 13 languages.
The authors wrote, “From April 2022 to date, Health Sentinel has processed over 300 million news articles and identified over 95,000 unique health events across India, of which over 3,500 events (four percent) have been shortlisted as potential outbreaks by NCDC health experts.” Authors of Wadhwani AI told PTI that between April 2022 and April 2025, more than 5,000 real-time alerts were sent to health authorities across India.
“Traditionally, the process of identifying potential disease events reported in the media involved manually scanning newspapers, magazines and reports to identify relevant articles,” Parag Govil, national program manager for global health security at Wadhwani AI, told PTI.
“The introduction of the ‘Health Sentinel’ solution has replaced this manual process while maintaining a human-in-the-loop approach in which epidemiologists conduct essential reviews before distributing the information to state and county officials,” he added.
Traditional approaches to disease surveillance rely on “passive reporting,” where infection reports are examined by physicians and healthcare providers.
Monitoring informal sources such as online media is also becoming increasingly popular for disease surveillance, but as the volume of articles published each day increases, the manual workload of screening media has increased and become impractical, said the study’s authors, including those from the National Center for Disease Control (NCDC).
They proposed the Health Sentinel tool, which uses AI to extract information about unusual health events or outbreaks from news articles.
Health officials and epidemiologists in the government noted that automating manual screening, along with faster outbreak detection and multilingual capabilities, were among the prerequisites for strengthening the country’s media-based surveillance, Govil said.
The research team observed a 150 percent increase in published events since 2022 compared to previous years of human-based disease surveillance.
In addition, 96 percent of health events published by the national surveillance system in 2024 were extracted by the AI tool – only four percent through manual media scanning, they said.
Studies suggest that traditional surveillance should be bolstered by reviewing online content – news reports or social media posts – to better detect infectious disease outbreaks.
A study published in February in the Indian Journal of Medical Research: Official Journal of the Indian Council of Medical Research examined an event-based surveillance system that examined media and rumor registers at six private hospitals in Kasaragod district of Kerala.
Researchers at the ICMR-National Institute of Epidemiology in Chennai and Kerala state and regional health authorities developed an algorithm that analyzed case records of patients admitted with acute febrile illness (AFI) – a fever that can last up to two weeks.
Symptoms of interest such as AFI with rash or bleeding were identified based on the spatiotemporal clustering of patients or deaths.
From May to December 2023, nearly three-quarters of over 4,500 patients with AFI were analyzed using the algorithm. Of the 88 clusters identified, 76 percent were due to severe acute respiratory illness, 10 percent to acute encephalitis syndrome, and nine percent to AFI with rash.
In addition, ten clusters were verified as events, nine of which were classified as outbreaks, including dengue fever and COVID-19, the study said.
The authors said: “The Event-based Surveillance (EBS) pilot in private healthcare facilities complemented the (traditional) system with early detection of outbreaks. This EBS model has the potential for implementation in other districts, particularly in districts with a higher risk of zoonotic disease transmission.”
A 2020 review in the Journal of Biomedical Informatics examined 148 research articles published between 2010 and 2019 on health monitoring that addressed social media, including posts on X (then Twitter).
Twenty-six of the articles, including those from the ScienceDirect and PubMed databases, were found to use machine learning – a form of AI algorithm – in surveillance.
About a quarter of articles focused on flu or flu-like illness surveillance, with Twitter being the most popular source of social media data for conducting surveillance research, the study said.
“Incorporating online data into surveillance systems has improved disease prediction capability compared to traditional syndromic surveillance systems,” wrote authors from Delhi Technological University.
Another study published in the American Journal of Tropical Medicine and Hygiene in 2017 pointed out how data from news articles can help avoid delays in obtaining country-level data on confirmed cases of infections such as dengue.
Published on November 16, 2025