AI for COVID-19: An online virtual care approach
summary
Mobile-first chat app and custom EHR integrate AI tools for virtual care in California, with provider escalation built in. Uses an expert-system+deep-learning pipeline: synthetic cases from a 50-year medical knowledge base plus real assessment and EHR data to train models; a COVID-19 diagnostic model reaches better top‑3 accuracy than average doctors on the benchmark used. Builds a pointer‑generator summarizer that catches negation and covers key concepts in >60% of cases, and uses GPT-3 as labeler to improve summary training. Double fine‑tuned BERT matches user COVID FAQs to answers.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 0:50Welcome
- 2:36Healthcare is screwed up
- 7:28AI for healthcare
- 10:07Example
- 15:29AI in the loop
- 21:47Conclusion
related talks
25mAI for healthcare: Scaling Access and Quality of Care for Everyone
01h 00mKeynote Panel: AI in Healthcare
28mAI in healthcare - Deep learning for precision psychiatry
47mAI for making decisions in clinical medicine
21mGen AI in Health: What's New in 2026?
07mHow AI is Enabling Healthcare's Biggest Blindspot
47mThe Future of Healthcare and AI
27mXavier Amatriain, Cofounder & CTO, Curai