Artificial Intelligence-driven solutions aimed at transforming healthcare claims management under the Ayushman Bharat Pradhan Mantri Jan Arogya Yojana (AB PM-JAY) were showcased at the AB PM-JAY Auto-Adjudication Hackathon Showcase 2026 that began here on Friday. The hackathon was organised by the National Health Authority (NHA) in collaboration with the IndiaAI Mission and the Indian Institute of Science (IISc), Bengaluru.

A key focus of the event was the presentation of AI and Machine Learning (AI/ML)-based solutions developed under the hackathon to improve the speed, accuracy and transparency of claims adjudication while reducing fraud and manual intervention.

AI systems

Under the first problem statement on Clinical Document Classification and Compliance with Standard Treatment Guidelines (STGs), participating teams demonstrated AI systems capable of automatically analysing healthcare documents and extracting structured clinical and billing information from low-quality scanned records using multilingual Optical Character Recognition (OCR). The solutions could also identify mandatory visual elements such as hospital stamps and doctor signatures and generate explainable adjudication outputs aligned with treatment guidelines and policy requirements.

For the second problem statement on Radiological Image-Based Condition Detection and Report Correlation, teams showcased assistive AI tools capable of interpreting X-rays, CT scans and MRI images. The systems were designed to help adjudicators correlate radiological findings with hospital-submitted reports and validate claimed disease conditions, treatment stages and timelines.

Detecting document forgery

The third problem statement focused on detecting document forgery and deepfakes in healthcare claims. Teams presented AI-powered systems aimed at identifying tampered discharge summaries, manipulated medical bills, fabricated reports and ghost identities submitted during claims processing.

Addressing the gathering, NHA Chief Executive Officer Sunil Kumar Barnwal said the authority’s in-house auto-adjudication initiatives had reduced average claim processing time from nearly 20 days to around four hours.

The showcase also featured discussions on the role of Small Language Models, Large Language Models and multimodal AI systems in healthcare delivery, particularly in low-resource and multilingual settings.