Amrita Hospital has launched MedSum, an AI-powered clinical assistant that automatically transcribes, summarises, and structures doctor-patient conversations into accurate, EHR-ready medical records in real time. The tool reduces documentation burden, minimises errors, ensures compliance with medico legal standards, and allows physicians to focus more on patient interaction than typing or note taking marking a significant advancement in ambient clinical intelligence for Indian hospitals.
Glimpse:
Unveiled on January 28, 2026, MedSum listens to natural conversations during consultations (via microphone or teleconsult setup), uses multilingual speech to text and medical NLP models to generate structured notes (SOAP format), auto populates diagnoses, prescriptions, investigations, and follow-up plans, and integrates directly with hospital EMR systems. Early pilots at Amrita Hospital Kochi and Faridabad show 70–85% reduction in post-consultation documentation time, high physician satisfaction, and strong accuracy across English, Hindi, Malayalam, and Tamil consultations.
Amrita Hospital has introduced MedSum, an indigenous AI clinical assistant designed to transform the way doctors document patient encounters in Indian healthcare settings. The technology, launched on January 28, 2026, addresses one of the most persistent pain points in clinical practice: the time consuming and error prone process of manual note taking, which often detracts from patient interaction and contributes to physician burnout.
MedSum operates as an ambient listening tool that captures live doctor-patient conversations (in-person or via teleconsultation), converts speech to text using advanced multilingual models optimised for Indian accents and medical terminology, and applies specialised natural language processing to extract and organise key clinical elements into structured SOAP notes (Subjective, Objective, Assessment, Plan). The system automatically identifies:
- Chief complaints and history of present illness
- Relevant past medical, surgical, and family history
- Physical examination findings
- Provisional/final diagnosis
- Investigations ordered
- Treatment plan, including prescriptions and follow-up instructions
The generated draft is presented to the doctor for quick review, edit, and approval before being committed to the hospital’s EMR system. MedSum supports English, Hindi, Malayalam, Tamil, and other regional languages, making it suitable for diverse patient populations across India.
In early pilot deployments at Amrita Hospital Kochi and Faridabad, physicians reported 70–85% reduction in post-consultation documentation time, with most notes requiring only minor edits. The tool demonstrated high accuracy in capturing clinical intent, even in noisy environments or with overlapping speech, and maintained strong performance across specialties (internal medicine, paediatrics, ENT, dermatology, and more).
Dr. Prem Kumar Vasudevan, lead developer and Head of Digital Health at Amrita Hospital, explained: “Doctors in India often spend more time writing notes than talking to patients. MedSum changes that dynamic allowing clinicians to focus entirely on history taking, examination, and counselling while the AI handles documentation accurately and efficiently. It’s built for Indian clinical workflows, Indian languages, and Indian medical guidelines.”
The platform is fully compliant with the Digital Personal Data Protection Act and hospital privacy policies, with end to end encryption, role-based access, and audit trails for every generated note. It integrates natively with leading Indian EMR systems and is designed to work offline in low-connectivity settings with delayed synchronisation when online.
Amrita plans to expand MedSum to all its hospitals and offer it to partner institutions through a SaaS model, with ongoing fine tuning to include more specialties, regional dialects, and AYUSH-integrated documentation.
“MedSum doesn’t just save time it restores the human element to medicine. When doctors can look patients in the eye instead of a screen, trust and care quality both improve.”
By
HB Team
