The National Centre for Disease Control (NCDC) may strengthen its public health security systems by tapping into social media posts, which could enhance its predictive model for disease pattern detection and outbreak surveillance.
The move follows the success of AI-based event surveillance systems already in use, most recently of the Media Scanning and Verification Cell, with which the NCDC has been able to scan millions of online news reports daily across 13 Indian languages to extract structured health event data, including the disease type, location, and scale.
The system has processed over 300 million news articles since its inception some years ago, flagging more than 95,000 unique health-related events — a 150% increase in detection capacity over manual systems, with 98% reduction in workload for the surveillance teams.
The current use of social media is still under discussion, Additional Director, NCDC, Himanshu Chauhan said.
“We are using technology as a digital watchdog, but we also understand that there could be many misses or [there could be] false reporting when it comes to social media. We don’t want to waste our resources,” Dr. Chauhan added.
NCDC tracks social media to create a future-ready public health system and enhance national readiness to manage infectious diseases, climate-related health risks, and potential pandemics, he said.
The AI-powered systems currently in place automatically identify unusual spikes in diseases, including dengue, chikungunya and other public health threats, which are then verified by experts for accuracy. The NCDC is also facilitating citizen reporting to flag any rise in disease incidence in an area.
In the transition from conventional methods of detection to predictive models, the newly established Metropolitan Surveillance Units (MSU) under the PM-Ayushman Bharat Health Infrastructure Mission (PM-ABHIM) had demonstrated exemplary real-time surveillance capabilities, Ranjan Das, Director, NCDC said.
“In a recent incident involving suspected paediatric Acute Encephalitis Syndrome cases in Chhindwara district, Madhya Pradesh, the Nagpur MSU promptly flagged the occurrence to the central surveillance unit, enabling rapid coordination among stakeholders across two States,” Dr. Das said.
The upcoming predictive model would integrate AI surveillance, laboratory intelligence, climatic data, population movement patterns, and digital diagnostics to anticipate outbreak trajectories, he added.
This proactive disease intelligence network would empower the health authorities to detect early warning signals before clinical manifestation, rapidly mobilise resources and field teams, and strengthen district-level risk mitigation. “Preventing large-scale [disease] outbreaks through advanced forecasting is the primary aim,” Dr. Chauhan said.