India’s AI Disease Surveillance Tool Issues 5,000+ Outbreak Alerts

India has achieved a major milestone in public health infrastructure with its AI-powered disease surveillance system fundamentally transforming how the nation detects and responds to infectious disease outbreaks. The Health Sentinel tool, developed by WadhwaniAI and integrated into the National Centre for Disease Control’s (NCDC) surveillance framework, has issued more than 5,000 real-time alerts to state and district health authorities since its deployment in April 2022.

This breakthrough represents a watershed moment for India’s compliance with International Health Regulations (IHR), which mandate that nearly 200 countries maintain robust national disease surveillance systems. Rather than relying on manual screening of newspapers, journals, and health reports—a process that had become increasingly impractical as media volume exploded—the Health Sentinel system automates the detection of unusual health events while maintaining rigorous human verification by epidemiologists before alerts reach officials.

How the WadhwaniAI Surveillance System Works ?

The Health Sentinel platform operates on a sophisticated multilingual framework, scanning news content daily across 13 languages to identify potential disease outbreaks and unusual health events. Since April 2022, the system has processed over 300 million news articles, identifying more than 95,000 unique health-related events across India. From these massive datasets, NCDC epidemiologists have flagged approximately 3,500 events—roughly 4%—as potential outbreaks requiring further investigation and intervention.

The system employs a hybrid human-in-the-loop approach, meaning that while artificial intelligence handles the initial screening and pattern recognition, trained epidemiologists at the NCDC verify all information before it reaches health officials. This combination preserves the speed advantages of automation while maintaining the critical human judgment needed for accurate outbreak assessment. As Parag Govil, National Program Lead for Global Health Security at WadhwaniAI, explained:

“Manual scanning of newspapers and journals is no longer feasible at the scale required. The AI system automates screening while retaining epidemiologists for verification before dissemination.”

Dramatic Improvements in Outbreak Detection

The impact of Health Sentinel on India’s disease surveillance capabilities has been nothing short of transformative. The system has reduced manual screening efforts by an extraordinary 98%, addressing a longstanding bottleneck in event-based surveillance as media volume continues to grow exponentially. Researchers have documented a 150% increase in published health events since the tool’s introduction in 2022, compared with earlier years when surveillance relied heavily on manual processes.

Perhaps most tellingly, in 2024 alone, 96% of the health events captured by India’s national surveillance system originated from the AI tool, with only 4% coming from manual screening. This dramatic shift underscores how comprehensively the AI system has become embedded in India’s disease detection infrastructure.

Between April 2022 and April 2025, the Health Sentinel system generated over 5,000 real-time alerts for state and district health departments across the nation, enabling faster interventions by health authorities and more proactive public health responses to emerging threats.

Real-World Validation Through Pilot Programs

The effectiveness of AI-enhanced surveillance has been validated through complementary research initiatives. A pilot study published in the Indian Journal of Medical Research tested event-based surveillance across six private hospitals in Kerala’s Kasaragod district, analyzing case records of patients admitted with acute febrile illness (AFI) using algorithms to detect patterns such as rashes, hemorrhage, or unusual clustering of cases.

Between May and December 2023, approximately three-fourths of the more than 4,500 AFI patients were evaluated using the algorithm. Of the 88 clusters identified, 76% were linked to severe acute respiratory illness, followed by cases of acute encephalitis syndrome and AFI accompanied by rashes. Critically, 10 clusters were verified as events, and 9 of these were confirmed as actual outbreaks, including dengue and COVID-19 cases.

These results demonstrate that event-based surveillance in private health facilities can detect outbreaks earlier than traditional systems, suggesting significant expansion potential to districts at higher risk of zoonotic spillover—situations where diseases jump from animals to humans.

Global Context and Future Implications

India’s approach aligns with emerging 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. A 2020 study published in the Journal of Biomedical Informatics examined 148 research articles on using social media—particularly Twitter—for healthcare surveillance between 2010 and 2019, with approximately one-fourth focused on flu surveillance and machine learning tools applied to analyze real-time user-generated data.

Similarly, a 2017 study in the American Journal of Tropical Medicine and Hygiene demonstrated that analyzing news reports can help compensate for delays in obtaining official country-level case confirmations for infections like dengue. These findings collectively reinforce that AI-enhanced media surveillance systems like Health Sentinel serve as powerful additions to traditional public health infrastructure.

The deployment of Health Sentinel addresses critical gaps identified by health officials and epidemiologists in India’s surveillance capabilities. Automation of manual screening, faster outbreak detection, and multilingual abilities emerged as essential requirements to strengthen the country’s media-based surveillance infrastructure. By meeting these requirements simultaneously, the Health Sentinel system has created a template for how developing nations can leverage artificial intelligence to strengthen their disease surveillance capabilities without requiring proportional increases in human resources.

As infectious disease emergence risks continue to escalate globally, India’s experience with Health Sentinel demonstrates that strengthening event-based surveillance through automation, multilingual capabilities, and continuous monitoring can significantly improve the timeliness of outbreak detection—a capability that becomes increasingly valuable in an era of rapid disease spread and evolving pathogens.

Join our community for more technology insights: Connect with us on WireUnwired Research WhatsApp group or engage with our WireUnwired Research LinkedIn community for ongoing discussions on tech news and innovations shaping India’s digital landscape.