An LLM-based Framework for Aligning Evolving Evidence to Order Sets

Presentation Type: Systems Demonstration

Click to View Presentation

Not recorded for AMIA Now

2026 Amplify Health System Presentation

Presentation Time: 10:15 AM - 10:40 AM

Abstract Keywords: Clinical Decision Support and Care Pathways, Generative AI in Clinical Workflow: Ambient Listening, Chart Summarization, Automated Response with LLM, Innovation Partnerships, Implementation Science, and Learning Health Systems, Quality Informatics and Lean
Primary Track: Innovation through Industry, Public Health, Non-profit and Commercial Partnerships


Order sets in the electronic health record (EHR) system frequently lag behind evolving medical evidence, contributing to variation in care, inefficiency, and potential patient safety risks. We present a framework that automatically links external knowledge to local EHR order sets and then identifies differences between the two. This two-step approach identifies missing elements, highlights inconsistencies, and enables targeted proactive content reviews. Early deployments at two pilot health systems demonstrate feasibility.

Speaker(s):
Marc Tobias, MD
Phrase Health

Andy Ngo, Trainee
BSW Health

Author(s):
Andy Ngo, Trainee - BSW Health; Illana Bookner, BS, MSE - Phrase Health; Marc Tobias, MD - Phrase Health;

---

SaPHIRE - Disease Surveillance Simplified

Presentation Type: Systems Demonstration

Click to View Presentation

Not recorded for AMIA Now

2026 Amplify Health System Presentation

2026 Amplify Health Equity Presentation

Presentation Time: 10:40 AM - 11:05 AM

Abstract Keywords: Public Surveillance and Reporting, Environmental Exposure, & Global Health, Standards, Terminology, and Interoperability, TEFCA, FHIR, Outcomes Improvement and Equity, Quality Informatics and Lean
Primary Track: Innovation through Industry, Public Health, Non-profit and Commercial Partnerships


Over 1,500 laboratories across the state of California submit reportable laboratory results to the California Department of Public Health (CDPH) via Surveillance and Public Health Information Reporting and Exchange (SaPHIRE). Symedical, a semantic interoperability solution, is integrated with SaPHIRE to normalize data from multiple feeds and facilitate data quality assurance. By enabling efficient, reliable, and secure public health reporting, SaPHIRE serves as a model for other states to follow regarding simplified statewide notifiable disease surveillance.

Speaker(s):
Erika Ganley, BSPharm
Clinical Architecture

Anand Kulanthaivel, PhD, MIS, FAMIA
Clinical Architecture

Author(s):
Erika Ganley, BSPharm - Clinical Architecture; Anand Kulanthaivel, PhD, MIS, FAMIA - Clinical Architecture; Jason Buckner, BS - Manifest MedEx;

---

The SEAL eConsult AI Assistant: An EHR-Integrated Tool for Question-Specific Chart Summarization and Draft Response Generation

Presentation Type: Systems Demonstration

Click to View Presentation

Not recorded for AMIA Now

2026 Amplify Health System Presentation

2026 Amplify 25x5 Presentation

Presentation Time: 11:05 AM - 11:30 AM

Abstract Keywords: Generative AI in Clinical Workflow: Ambient Listening, Chart Summarization, Automated Response with LLM, Workforce Automation, Communication, and Workflow Efficiency, Innovation Partnerships, Implementation Science, and Learning Health Systems, Clinical Decision Support and Care Pathways, Standards, Terminology, and Interoperability, TEFCA, FHIR
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


Electronic consultations (eConsults) offer a promising solution to specialist access barriers, enabling more timely guidance for conditions that may otherwise be appropriately managed by a patient's primary care provider. However, their ability to scale is often limited by time-consuming chart review performed by the specialist, rather than the actual application of their clinical expertise to answer the clinical question. We present the SEAL eConsult AI Assistant, an LLM-powered, EHR-integrated tool developed by the Stanford Emerging Applications Lab (SEAL) that produces question-specific chart summaries and draft consult responses grounded in curated, evidence-based clinical guidelines. The system operates through three stages: (1) classifying the clinical question into a predefined condition category, (2) leveraging FHIR to retrieve relevant patient chart information using predefined condition-specific data specifications, and (3) generating a focused chart summary and draft assessment grounded in condition-specific information templates and evidence-based guidelines. The tool enables seamless verification through in-text citations and a two-pane view depicting AI-generated outputs alongside source chart data. The SEAL eConsult AI Assistant launched as a pilot in November 2025 for Infectious Disease eConsults at Stanford Health Care, with plans to expand to Dermatology in early 2026.

Speaker(s):
Shivam Vedak, MD, MBA
Stanford University School of Medicine

Author(s):
Shivam Vedak, MD, MBA - Stanford University School of Medicine; Jennifer Tran, PharmD; Srinivasan Boosi, BE - Stanford Health Care; Eileen Loh, MS - Stanford Health Care; Joe Pallas, PhD - Stanford Health Care; Stephen Ma, MD, PhD - Stanford University School of Medicine; Oluseyi Fayanju, MD - Stanford University; Olivia Jee, MD - Stanford University School of Medicine; Debbie Kim, MPH, PMP - Stanford Health Care; Lena Giang, MPH - Stanford Health Care; Elyse Ruan, MCIM - Stanford Health Care; Lisa Gohil, MBA, MPH - Stanford Health Care; Leah Rosengaus, MS - Stanford Health Care; Lawrence Hoffman, MD - Stanford University School of Medicine; Christopher Sharp, MD - Stanford University School of Medicine; Ron Li, MD - Stanford School of Medicine;

---