IIT Kanpur Hackathon Drives Secure Health AI with Federated Intelligence Models

According to the Government of India, a Hackathon was held as a pre-event to the India AI Impact Summit 2026. This national initiative aimed to drive the creation of secure, privacy-preserving, and scalable artificial intelligence solutions for healthcare in India. The hackathon took place from 19th to January 24 at IIT Kanpur.

In his keynote address, Sunil Kumar Barnwal, Chief Executive Officer, National Health Authority, emphasised the strategic importance of building a trusted, federated AI ecosystem for healthcare, marking a shift from experimentation to the adoption of benchmarked, reliable AI models.

He highlighted that AI systems must be tested on diverse, population-scale datasets before deployment and noted that federated, consent-driven architectures allow innovation to scale without centralising data, ensuring privacy and trust.

Citing AB PM-JAY (Ayushman Bharat Pradhan Mantri Jan Aarogya Yojana) and the Ayushman Bharat Digital Mission (ABDM), he underscored the importance of developing AI solutions that are context-ready, inclusive, and reflective of India's demographic and regional diversity.

The inaugural session also featured addresses by Professor Sandeep Verma, Head of Gangwal School of Medicine & Technology; Manindra Agrawal, Director, IIT Kanpur; Ritu Maheshwari, Secretary, Medical Health & Family Welfare and State Mission Director, ABDM-Uttar Pradesh.

Dr. R.S. Sharma, Distinguished Visiting Professor at IIT Kanpur; former CEO of the National Health Authority; and former Mission Director at UIDAI, highlighted how Digital Public Infrastructure and interoperable Digital Public Goods enable secure, scalable, and citizen-centric health data management.

According to the Government of India, Vivek Raghavan, CEO and Co-founder, SarvamAI, underscored the role of India's layered digital health architecture in enabling AI-driven healthcare transformation at both population and individual levels.

He emphasised that data quality, privacy, and security are foundational to effective AI adoption and highlighted the importance of indigenous, open-source AI models and local AI sovereignty to ensure responsible innovation, reduce external dependencies, and maximise public-sector impact.

A session was also led by ICMR-NIRDHDS to discuss their initiatives in AI in health.

Through a partnership between IIT Kanpur and NHA, a benchmarking platform is being developed which can evaluate the efficacy of a given AI model. This platform ensures data privacy, institutional control and trust, while enabling innovation at a national scale. All the models built during the hackathon were evaluated on this platform.

A total of 374 registrations were received for the Hackathon, including 208 individual participants and 166 teams of two or more participants.

About 54% of participants identified themselves as AI researchers or innovators. The remaining participants represented diverse categories, including health-tech startups, healthcare providers, graduate students, data scientists, and others.

The winners of the hackathon were announced in three categories: bone age prediction, cataract detection, and diabetic retinopathy screening. The hackathon winners were invited to make presentations on the methodology used to make the models. The winners received certificates and cash prizes worth Rs. 12 lakhs.

Through this collaborative initiative with ICMR-NIRDHDS and IIT Kanpur, the National Health Authority continues to strengthen secure, interoperable, and citizen-centric Health AI solutions under the Ayushman Bharat Digital Mission.

- ANI