Use of Generative AI in Medical Translation at Health Systems for Patients with Diverse Preferred Languages

Presentation Time: 10:30 AM - 12:00 PM

Abstract Keywords: Governance of Artificial Intelligence, Health Equity, Delivering Health Information and Knowledge to the Public
Primary Track: Applications
Programmatic Theme: Clinical Informatics

Effective communication is a core tenet of healthcare delivery, and Title VI of the Civil Rights Act requires that all patients receive equitable linguistic access as a matter of non-discrimination. However, patients with preferred languages other than English (PLOE) continue to face significant challenges in accessing healthcare, leading to disparate health outcomes. Recent innovations in generative AI show potential for effectively translating patient-facing materials. However, regulation requires that machine translation can only be used to translate written text with the involvement of a qualified human translator. This panel will highlight the importance of achieving equitable access to health information for patients with PLOE and the opportunities, strengths, and weaknesses of using generative AI for translation. We will specifically focus on the practical challenges and best practices in effectively operationalizing translation workflows, using generative AI as a force multiplier, in the healthcare setting.

Moderator:
Angad Singh, MD
University of Washington School of Medicine

Speaker(s):
Ryan Brewster, MD
Boston Children's Hospital

Nymisha Chilukuri, MD
Stanford Medicine

Gabriel Tse


Sansanee Craig, MD
The Children's Hospital of Philadelphia

Yu-Hsiang Lin, MD
Seattle Children's Hospital

Angad Singh, MD
University of Washington School of Medicine

Author(s):