Skip to main content
AAO-HNSF 2025 Annual Meeting & OTO EXPO Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Scientific Oral Presentations

Care Equity, Quality, and Safety

Session: Scientific Oral Presentation 3 - Care Equity, Quality and Safety

Expanding Capacity of General Practitioners in Sub-Saharan Africa with Artificial Intelligence: A Humanitarian Outreach Experience

Monday, October 13, 2025
9:48 AM - 9:54 AM EDT
Location: 107-108
Disclosure(s):
Jerome R. Lechien, MD, PhD, FACS:

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

Lise Sogalow, MD:

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

Introduction: Low- and middle-income countries face substantial challenges in healthcare delivery due to the shortage of practitioners, especially specialists, and equipment.

Objective: To evaluate POE, a multi-modal large language model platform, for diagnostic accuracy, clinical consistency, and usefulness as an adjunctive tool for otolaryngological care in resource-limited rural Kenyan settings.

Methods: Consecutive patients with otolaryngological conditions were prospectively evaluated by both a board-certified otolaryngologist and primary care practitioner (PCP) at a rural Kenyan facility from December 2024 to January 2025. Clinical data and portable micro-otoscopic images were submitted to POE's interface following PCP assessment. The otolaryngologist compared POE outputs with PCP assessments with the Artificial Intelligence Performance Instrument (AIPI) to evaluate diagnostic, management, and therapeutic concordance.

Results: The primary diagnoses generated by POE demonstrated concordance with otolaryngologist findings in 50 cases (79.4%), while management recommendations aligned in 61 cases (96.8%). Differential diagnoses were judged as plausible and correct in 46 cases (73.0%). Among the discordant cases (n=17), POE correctly included the confirmed diagnosis within its differential considerations in 16 outputs. POE demonstrated superior diagnostic accuracy compared to PCPs (79.4% [50/63] versus 50.8% [32/63], p=0.001) in establishing primary diagnoses. Among cases where PCPs failed to establish the diagnosis, POE successfully identified the primary diagnosis in 61.3% (19/31). Cases with PCP-established diagnoses (n=32) demonstrated higher AIPI diagnostic subscores compared to undiagnosed cases (n=31). When provided with clinical examination images, POE correctly identified the condition as either primary (n=7/14, 50.0%) or potential (n=5/14, 35.7%) diagnoses.

Conclusions: The implementation of integrated LLM platforms in low- and middle-income countries demonstrates usefulness as clinical decision support tools, augmenting PCP diagnostic capabilities.

    Presenting Author(s)

  • JL

    Jerome R. Lechien, MD, PhD, FACS

    Chair
    University of Mons
    Sombreffe, Hainaut, Belgium

    Disclosure(s):

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

    • Co-Author(s)

  • LS

    Lise Sogalow, MD

    PhD student
    University of Mons
    Jurbise, Hainaut, Belgium

    Disclosure(s):

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

    • Senior Author(s)

  • GB

    Giovanni Briganti, MD

    UMONS, Belgium

    Disclosure information not submitted.