Agentic AI for Intelligent Patient Call Triage

Description

Jahnavi Kachhia presents “Agentic AI for Intelligent Patient Call Triage” — a multi-modal, multi-agent healthcare framework designed to reduce clinician overload, improve patient prioritisation, and make AI-assisted triage auditable, explainable, and clinically grounded. Timestamps: 00:00 Why healthcare call triage is breaking at scale 02:00 Multi-agent AI architecture for patient call triage 05:15 Multimodal ingestion, EHR integration, and RAG grounding 07:00 AI triage classification, orchestration, and routing logic 11:00 Real-world healthcare implementations and multi-agent deliberation 15:00 HIPAA, PHI protection, and deployment challenges 16:40 Future roadmap: multimodal AI, reinforcement learning, and governance The system decomposes triage into specialised agents rather than relying on a single Large Language Model (LLM). Separate agents handle multimodal ingestion, privacy reduction, retrieval grounding, triage classification, orchestration, and explainability, creating a layered workflow designed for safer clinical decision support. The architecture combines Retrieval-Augmented Generation (RAG), vector databases, Electronic Health Records (EHRs), speech-to-text pipelines, semantic alignment checks, and real-time orchestration to process inbound patient calls under high cognitive-load conditions. A major focus is reliability under healthcare constraints: Personally Identifiable Information (PII) protection, Health Insurance Portability and Accountability Act (HIPAA) compliance, hallucination mitigation, auditability, and human-in-the-loop escalation. The presentation also explores multi-agent deliberation frameworks, confidence scoring, reinforcement learning feedback loops, adaptive orchestration, and region-specific governance considerations for deploying healthcare AI systems at production scale. 📌 Applied Healthcare AI Summit 2026 — what actually works in real-world healthcare AI, from pilots to production systems. #HealthcareAI #AgenticAI #ClinicalAI #RAG #EHR #HIPAA #MultiAgentSystems