Originals · Startup · Startups & Fintech
Apollo Hospitals and Microsoft built an AI agent to review medical errors faster across its network
The Daily Tiered Huddle agent pulls patient records together for about 600 adverse event reviews a month. Apollo reports quicker case closures, though the results are its own.

Every afternoon in Chennai, Apollo Hospitals' clinical governance team reviews the adverse events reported across its network in India: falls, medication errors, readmissions. With about 600 such events a month across its 55 main hospitals, a single review used to take days while staff gathered details by hand from different records, according to a Microsoft case study published on 17 November 2026 and reported by YourStory.
To speed that up, Apollo and Microsoft's forward deployed engineers built what they call the Daily Tiered Huddle (DTH) agent on Microsoft Foundry. It reads hospital data held in Microsoft Fabric, flags incidents that need review, builds a timeline of each patient's treatment, prompts clinicians for missing details and suggests case assessments that doctors check and amend. "For anyone to process 600 cases per month with incomplete data and come up with a meaningful decision is difficult," said Dr Rohini Sridhar, Apollo's chief of medical services.
Apollo says more than 60% of the cases reviewed by its huddles are now closed within three days, and it recorded fewer falls and medication errors in July to September than in the same quarter of 2025. Dr Sridhar said staff satisfaction with how errors are handled improved by almost 30%, based on the hospital's annual safety culture survey. YourStory pointed out that these are hospital reported figures in a vendor case study, with no incident counts, adjusted rates or independent assessment, so they do not show that the AI itself reduced harm.
The system is also meant to spot patterns across departments, an analysis Apollo says is still being developed. It has already flagged a cluster of central line associated bloodstream infections in a bone marrow transplant unit. The project sits within a wider push to bring AI into Indian hospital operations, alongside national efforts such as the Ayushman Bharat Digital Mission and the BODH platform for benchmarking health AI models built with IIT Kanpur and the National Health Authority.