AI Tool in India’s Disease Surveillance System Issued Over 5,000 Outbreak Alerts Since 2022: Study

The AI system, Health Sentinel, is integrated into the National Centre for Disease Control’s (NCDC) media scanning and verification framework under the Integrated Disease Surveillance Programme (IDSP).

India’s national disease surveillance programme has issued more than 5,000 real-time alerts on potential infectious disease outbreaks since 2022 using an artificial intelligence tool developed by WadhwaniAI, according to a new pre-print study.

The AI system, Health Sentinel, is integrated into the National Centre for Disease Control’s (NCDC) media scanning and verification framework under the Integrated Disease Surveillance Programme (IDSP).

According to researchers, Health Sentinel has processed over 300 million news articles in 13 languages, identifying 95,000 unique health-related events across the country. NCDC epidemiologists shortlisted more than 3,500 as potential outbreaks requiring further assessment. Between April 2022 and April 2025 alone, the system generated over 5,000 real-time alerts to state and district health authorities.

The study indicates that the tool reduced manual screening efforts by 98%, addressing a longstanding challenge in event-based surveillance (EBS) as media volume continues to rise. “Manual scanning of newspapers and journals is no longer feasible at the scale required,” said Parag Govil, national program lead for global health security at WadhwaniAI. “The AI system automates screening while retaining epidemiologists for verification before dissemination.”

Researchers noted a 150% increase in published health events since the tool’s introduction, compared with earlier years when surveillance relied heavily on manual processes. In 2024, 96% of the health events captured by India’s national surveillance system originated from the AI tool, signalling a shift towards technology-driven detection.

India’s use of digital surveillance aligns with global trends encouraging the integration of non-traditional data sources—including online media, citizen reports, and social media signals—to identify outbreaks earlier than conventional systems. Previous studies have shown that digital cues often surface days before clinical reporting catches up, especially for fast-spreading infections like dengue or influenza-like illnesses.

The report also references complementary research, including a Kerala-based pilot where digital event-based surveillance detected clusters linked to severe respiratory illnesses and acute febrile cases earlier than traditional systems. Global reviews have similarly highlighted the value of AI and machine learning in monitoring outbreaks through platforms like Twitter and online news databases.

Researchers said strengthening EBS through automation, multilingual capabilities, and continuous monitoring can help India improve the timeliness of outbreak detection amid increasing disease emergence risks.

Stay tuned for more such updates on Digital Health News