How we use AI to deliver Telehealth at Transcarent
summary
Telehealth flow starts with an AI chatbot (acquired product 98.6) that triages patients, generates a curated question sequence, and produces top-5–10 diagnoses that physicians accept 86% agreement of the time; common conditions hit >90% accuracy and represent ~50% of volume. 100% of visits begin with the bot, overall visit completion runs 80–85%, and fine-tuned LLMs plus health-plan data automate post-diagnosis counseling and provider matching while safety checks and physician confirmation remain mandatory.
AI-written summary · about
chapters (8)
- 0:00AI in Telehealth intro
- 1:06Healthcare challenges & AI
- 6:03Telehealth visits today
- 9:13AI interview & diagnosis
- 15:55Generative models expansion
- 19:42Responsible AI & regulations
- 20:25Top 3 takeaways
- 21:28Q&A: Data privacy & AI
related talks
09mEnhance Healthcare Chatbot Transparency
23mHow AI Is Rebuilding Healthcare
11mClinical AI Agents with Oracle Health
20mImproving Healthcare with AI Assistance
42mAI-Powered Healthcare CRM | Cognitive Companion Revolution
24mLeveraging Voice in Healthcare After COVID
26mHumanity and AI
01h 56mCase Study + Deep Dive: Telemedicine Support Agents with LangGraph/MCP