13 contributors. 5 institutions. Clinicians who code and coders who listen.
Senior Consultant in General Internal Medicine. Program Director, Internal Medicine Residency at OMSB. Clinician-researcher in hospital medicine, patient outcomes, and AI in medicine.
SQUH and the National Heart Center. AI in cardiovascular care, arrhythmia detection, decision support.
Clinical research, medical education, digital health innovation.
Hospital medicine, lipidology, preventive cardiovascular care, AI-based risk prediction.
Acute Care Medicine. Clinical excellence, education, and service development.
SQUH. Digital mental health innovations and AI integration.
Internal Medicine Resident at OMSB. Active in medical research.
MCMSS. Leadership, medical education, research, AI integration in medicine.
OMSB · SQU graduate. Medical research and AI in healthcare.
SQU. Cardiology, AI, and population-based clinical studies.
GUtech. Full-stack, machine learning, RAG pipelines, local LLM deployment.
BSN, MSN in Adult and Critical Care Nursing. Frontline clinical documentation expertise.
Sovereign AI. Knowledge economy. Healthcare in the citizen's own language.
Trained, hosted, and operated on Omani soil. Every line of code, every weight, every recording — in-country. Nothing federated abroad.
Closing the gap between how patients speak and how charts are written. Not translating — listening, in the dialect of the bedside.
Doctors, residents, and engineers from SQUH, OMSB, MCMSS, SQU, and GUtech — co-designing the system that serves them. A national capability, not a vendor relationship.
What works in Muscat scales to Riyadh, Cairo, Amman, Doha. Built to be lifted, not licensed.
Real numbers from a real pipeline. No demos — this is live data from our annotation platform.
52 contributors from 6 dialect backgrounds. Two GPU workstations. 5,846 hours of Arabic audio harvested and transcribed on-prem.
3,933 clips fully verified through dual-review. 9,187 total annotation touches. 79% inter-annotator agreement. 93 hours of human effort. Join us — 16,000+ clips still in queue.
Fine-tune Whisper on the Omani corpus. Train a local LLM to structure transcripts into SOAP notes. Targeted Q3 2026.
A small, IRB-approved bedside study at SQUH. Measure clinician time saved, note quality, patient comfort. 2027, not before.
Transparently funded through national research grants.
Oman's healthcare is shaped by centuries of trade, tradition, and a mosaic of Arabic dialects. Every governorate has its own way of describing pain, illness, and healing. Medad is trained on this country — not generic Arabic from elsewhere.
