Practical AI for Global ENT: Implementing Safe, Equitable Clinical Decision Support in Lmics

Monday, October 19, 2026

4:00 PM - 5:00 PM PT

Location: West Exhibit Hall A, Learning Zone 4

Moderator(s)

Myra A. Zaheer, BA

Medical Student
George Washington University, School of Medicine & Health Sciences
Washington, District of Columbia, United States

Panelist(s)

Matthew G. Crowson, MD

Assistant Professor
Mass Eye & Ear, Harvard Medical School
Boston, MA, United States
Anaïs Rameau, MD MSc MPhil MS FACS

Associate Professor
Weill Cornell Medicine
New York, NY, United States
AL

Anthony B. Law, MD, PhD

Assistant Professor
Emory University
Atlanta, GA, United States
JW

Joshua P. Wiedermann, MD, MGSC, FACS

Chair of Pediatric Otolaryngology
Mayo Clinic
Rochester, Minnesota, United States

Disclosure(s):

Matthew G. Crowson, MD:

faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.

Anaïs Rameau, MD MSc MPhil MS FACS:

faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.

The session addresses: (1) selecting LMIC-relevant ENT use cases where AI demonstrates potential to improve access and quality, including otoscopic image interpretation for otitis media triage, endoscopic assessment for airway pathology, head and neck cancer risk stratification, and hearing screening pathways, while identifying settings where AI may increase risk; (2) designing locally led validation and monitoring strategies that address algorithmic bias, model drift, and context-specific performance; (3) integration into clinical workflows with limited infrastructure using principles of digital development; and (4) governance models that prioritize LMIC ownership avoiding technological colonialism and ensuring community protection through transparent, accountable frameworks. [1-8]

Learning Objectives:

Select LMIC-relevant ENT workflows where AI may improve access/quality (smartphone otoscopy triage, hearing screening, endoscopic airway assessment, head/neck risk stratification) and flag settings where AI may increase risk from poor validation
Apply a minimum evidence framework for LMIC implementation readiness: local validation on representative datasets, bias assessment across subgroups, drift monitoring, human-in-the-loop oversight, and user-centered, context-fit design
Define success metrics for systems strengthening/capacity-building: time-to-referral, triage accuracy, complications, follow-up, equity-stratified outcomes, cost-effectiveness, maintenance feasibility, and competencies for safe AI use/monitoring