Clinical Analysis Intelligence
Upload a medical report (TXT or PDF) for autonomous agentic reasoning.
Drag & Drop Report
or click to browse local files
Agentic Pipeline Active
Orchestrator is extracting entities...
Analysis Complete
Specialty
Risk
🩺 Executive Clinical Summary
Processing...
🧬 Extracted Biomarkers & Vitals
Processing...
⚠️ Critical Risk Flags
Processing...
📋 Recommended Next Steps
Processing...
⏱️ Agent Reasoning Timeline
- Processing...
🏥 Fivetran EHR Context
Processing...
🚨 GitLab Escalation Panel
Processing...
⚡ AI Observability
- Confidence Score: --
- Self-Healing: Active via Arize Phoenix
- Actions Generated: --
AI Observability Settings
Arize Phoenix
Telemetry Live
⚙️ Telemetry Controls
Configure real-time agentic trace exports and self-healing parameters.
Enable Arize Phoenix LLM Tracing
Stream full execution spans to local telemetry port
Critic Agent Self-Healing Loop
Enable multi-turn correction when validation falls below threshold
Fivetran EHR Data Sync Logging
Log detailed pipeline telemetry during patient context matching
Critic Confidence Threshold
85%
Pipeline will automatically trigger healing iterations below this score
🖥️ Live Telemetry Terminal
Real-time traces from Vertex AI, Fivetran EHR connector, and Critic agent.
[SYSTEM] Connecting to Arize Phoenix... OK
[TRACE] ID: 9x8b-4m2q | Model: gemini-1.5-pro | Latency: 1.2s
[METRIC] Token Usage: 1,452 Prompt / 328 Completion
[ACTION] Critic Agent: Confidence 92% -> Validation Passed.
Listening for traces on port 6006...