Journal article
Evaluating human-in-the-loop strategies for artificial intelligence-enabled translation of patient discharge instructions: a multidisciplinary analysis
NPJ digital medicine, v 8(1), 629
24 Oct 2025
PMID: 41136708
Abstract
Machine translation supported by artificial intelligence (AI) may enhance linguistically-concordant care for patients speaking languages other than English. This assessment of free-text inpatient discharge instructions in Arabic, Armenian, Bengali, simplified Chinese, Somali, and Spanish compared linguist, clinician, and family caregiver evaluations of translations generated by (1) ChatGPT-4o, (2) professional linguists, and (3) human-in-the-loop (AI-generated, professional linguist post-edited). Likert scales (1-5; higher is better) evaluated linguistic and clinical characteristics of each translation. ChatGPT-4o exhibited variable performance relative to professional translations, with poorest ratings for digitally underrepresented languages (Armenian and Somali). Conversely, human-in-the-loop translations achieved comparable, often better, outcomes to professional translations for all languages, (e.g., Armenian mean overall quality: 3.9 [95% CI 3.7-4.2] vs. professional 3.6 [3.4-3.9], p = 0.01), were most frequently preferred (46.5% vs. 28.4%) and had shorter mean translation time (7.1 [5.4-8.8] vs. 16.8 [13.7-19.9] min, p < 0.001). Human-in-the-loop strategies may enable safe, efficient, equitable machine translation application in clinical practice.
Metrics
10 Record Views
Details
- Title
- Evaluating human-in-the-loop strategies for artificial intelligence-enabled translation of patient discharge instructions: a multidisciplinary analysis
- Creators
- Ryan Cl Brewster (Corresponding Author) - Beth Israel Deaconess Medical CenterGabe Tse - Stanford University School of MedicineAngela L Fan - Boston Children's HospitalMarwa Elborki - Boston Children's HospitalMaiah Newell - Boston Children's HospitalPriscilla Gonzalez - Harvard UniversityAmitra Hoq - New York UniversityCrystal Chang - Kaiser PermanenteMaksud Chowdhury - Department of Pediatrics, State University of New York Downstate Medical Center, Brooklyn, NY, USAAdiba Geeti - New York UniversityMarlin Hana - Virginia Commonwealth UniversityHoda Hassan - The Ohio State University Wexner Medical CenterOsama Ibrahim - Harvard UniversityLucine Keseyan - Cedars-Sinai Medical CenterQing Li - University of Mississippi Medical CenterMd Mamoon - Icahn School of Medicine at Mount SinaiMaymona Nageye - Avalon Pharma (United States)Arthur Ohannessian - University of California, Los AngelesIlan Rozen Eisenberg - Boston Medical CenterMohammad Sallam - Mercy Medical CenterGiordano Sosa Soto - Boston Children's HospitalChristina Su - Boston Medical CenterRaffi Tachdjian - St. John Medical CenterMondira Ray - Boston Children's HospitalHannah Lev - Boston Children's HospitalJonathan D Hron - Harvard UniversityNate Shaar - Boston Children's HospitalNicholas Kuzma - Drexel UniversityAlisa Khan - Boston Children's Hospital
- Publication Details
- NPJ digital medicine, v 8(1), 629
- Publisher
- Nature Publishing Group
- Number of pages
- 8
- Grant note
- AD-2021C3-24848 / Patient-Centered Outcomes Research Institute
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- Pediatrics
- Web of Science ID
- WOS:001600911300002
- Scopus ID
- 2-s2.0-105024062620
- Other Identifier
- 991022124263604721
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Collaboration types
- Domestic collaboration
- International collaboration
- Web of Science research areas
- Health Care Sciences & Services
- Medical Informatics