AI-TransLATE: Validation of a Speech-Based Multilingual Interpretation Tool in Critical Care

Abstract

Background

Effective communication in the intensive care unit (ICU) is essential, particularly for patients with non-English language preference, yet timely access to professional interpreters remains limited. While artificial intelligence (AI)-based translation tools have been explored in outpatient and nonacute care settings, studies evaluating their use in acute care, environments such as the ICU remain limited. To address this gap, we developed AI-TransLATE (AI-enhanced Transition to Language-Agnostic Transcultural Engagement), a speech-based translation tool designed for multilingual communication in critical care settings.

Objectives

This study aimed to assess the interpretation quality of AI-TransLATE across four languages—Spanish, Chinese, Arabic, and Turkish—using scripted ICU scenarios.

Methods

We created ICU communication scripts and recorded bilingual research team members simulating clinical interactions. Two independent bilingual evaluators assessed interpretation quality using a 5-point Likert scale across fluency, adequacy, meaning preservation, and severity of errors. Clarity and cultural appropriateness were also rated. Percentage agreement was used to assess interrater agreement.

Results

AI-TransLATE achieved acceptable composite scores (≥16/20) across all languages. Spanish and Turkish performed consistently well; Chinese and Arabic showed variability due to omissions and terminology errors.

Conclusion

AI-TransLATE shows promise as a clinical communication tool, but further evaluation in real-world, unscripted ICU settings is needed.

Keywords

communication barriers - health care disparities - medical interpreting - artificial intelligence - intensive care units

Protection of Human and Animal Subjects

The study was performed in compliance with the World Medical Association Declaration of Helsinki on Ethical Principles for Medical Research Involving Human Subjects, and was reviewed by Mayo Clinic Institutional Review Board (identifier: 25-001569).

Note

AI-TransLATE, the tool evaluated in this study, was developed by members of the study team. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health. Artificial intelligence tools were used for language refinement and scenario generation; all conceptualization, study design, and data analysis were conducted independently by the authors.

Publication History

Received: 10 July 2024

Accepted: 12 December 2025

Accepted Manuscript online:
16 December 2025

Article published online:
24 December 2025

• References

• 1 Betancourt JR, Renfrew MR, Green AR, Lopez L, Wasserman M. Improving Patient Safety Systems for Patients With Limited English Proficiency: A Guide for Hospitals. Agency for Healthcare Research and Quality; 2012, September. https://www.ahrq.gov/sites/default/files/wysiwyg/professionals/systems/hospital/lepguide/lepguide.pdf
• 2 Duronjic A, Ku D, Chavan S, Bucci T, Taylor S, Pilcher D. The impact of language barriers & interpreters on critical care patient outcomes. J Crit Care 2023; 73: 154182
• 3 LEPgov. Accessed November 5, 2024 at: https://www.lep.gov/maps
• 4 American Community Survey, Detailed Languages Spoken at Home and Ability to Speak English for the Population 5 Years and Over 2009–2013. Accessed November 5, 2024 at: https://www.census.gov/data.html
• 5 Yang C, Prokop L, Barwise A. Strategies used by healthcare systems to communicate with hospitalized patients and families with limited English proficiency during the COVID-19 pandemic: a narrative review. J Immigr Minor Health 2023; 25 (06) 1393-1401
• 6 Karliner LS, Pérez-Stable EJ, Gregorich SE. Convenient access to professional interpreters in the hospital decreases readmission rates and estimated hospital expenditures for patients with limited English proficiency. Med Care 2017; 55 (03) 199-206
• 7 Kwan M, Jeemi Z, Norman R, Dantas JAR. Professional interpreter services and the impact on hospital care outcomes: an integrative review of literature. Int J Environ Res Public Health 2023; 20 (06) 5165
• 8 Kelleher DP, Barwise AK, Robbins KA, Borah BJ. Examining the association between professional language interpreter services and intensive care unit utilization among patients with non-English language preference: evidence from a large U.S. hospital. Patient Educ Couns 2024; 128: 108375
• 9 Suarez NRE, Urtecho M, Jubran S. et al. The roles of medical interpreters in intensive care unit communication: a qualitative study. Patient Educ Couns 2021; 104 (05) 1100-1108
• 10 Barwise AK, Moriarty JP, Rosedahl JK. et al. Comparative costs for critically ill patients with limited English proficiency versus English proficiency. PLoS One 2023; 18 (04) e0279126
• 11 Tang D, Jawad D, Dragoje V, Wen LM, Taki S. The use of interpreter services and its barriers faced by hospital staff when accessing interpreters for patients with low English proficiency during the COVID-19 pandemic. Health Promot J Austr 2024; 35 (04) 1184-1193
• 12 Barwise AK, Curtis S, Diedrich DA, Pickering BW. Using artificial intelligence to promote equitable care for inpatients with language barriers and complex medical needs: clinical stakeholder perspectives. J Am Med Inform Assoc 2024; 31 (03) 611-621
• 13 Zappatore M, Ruggieri G. Adopting machine translation in the healthcare sector: a methodological multi-criteria review. Comput Speech Lang 2024; 84: 101582
• 14 Lopez A. Statistical machine translation. ACM Comput Surv 2008; 40 (03) 8
• 15 Blunsom P, Osborne M. Probabilistic Inference for Machine Translation. Association for Computational Linguistics; 2008: 215-223
• 16 Charoenpornsawat P, Sornlertlamvanich V, Charoenporn T. Improving Translation Quality of Rule-based Machine Translation; 2002. In COLING-02: Machine Translation in Asia.
• 17 Herrera-Espejel PS, Rach S. The use of machine translation for outreach and health communication in epidemiology and public health: scoping review. JMIR Public Health Surveill 2023; 9: e50814
• 18 Wołk K, Marasek K. Polish-English Statistical Machine Translation of Medical Texts. Springer International Publishing; 2015: 169-179
• 19 Kapoor R, Corrales G, Flores MP, Feng L, Cata JP. Use of neural machine translation software for patients with limited English proficiency to assess postoperative pain and nausea. JAMA Netw Open 2022; 5 (03) e221485-e221485
• 20 Strechen I, Gupta A, Karakus S. et al. Artificial intelligence-assisted translation and interpretation for communication with patients who have language barriers: a systematic review. Am J Respir Crit Care Med 2025; 211: A3643
• 21 Karakus IS, Strechen I, Gupta A. et al. Bridging language gaps in healthcare: a systematic review of the practical implementation of neural machine translation technologies in clinical settings. J Am Med Inform Assoc 2025; 32 (11) 1756-1766
• 22 Daniels P. The syllabic origin of writing and the segmental origin of the alphabet. In: Downing, Pamela, Lima, Susan, & Noonan, Michael (eds), The Linguistics of Literacy. Typological Studies in Language. 1992;(21):83–110
• 23 Feldman LB, Siok WWT. Semantic radicals contribute to the visual identification of Chinese characters. J Mem Lang 1999; 40 (04) 559-576
• 24 Lee W, Khoong EC, Zeng B. et al. Evaluation of commercially available machine interpretation applications for simple clinical communication. J Gen Intern Med 2023; 38 (10) 2333-2339
• 25 Tan KS, Adusumilli P, Travis WD. P1.24-17 The paradoxical behavior of kappa when quantifying agreement in histological subtype classification among thoracic pathologists. J Thorac Oncol 2023; 18 (11) S266-S267
• 26 Zoph B, Yuret D, May J, Knight K. Transfer Learning for Low-Resource Neural Machine Translation. Association for Computational Linguistics; 2016: 1568-1575
• 27 Taira BR, Kreger V, Orue A, Diamond LC. A pragmatic assessment of Google translate for emergency department instructions. J Gen Intern Med 2021; 36 (11) 3361-3365
• 28 Mehandru N, Robertson S, Salehi N. Reliable and safe use of machine translation in medical settings. Presented at: Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency; 2022. Seoul, Republic of Korea.
• 29 Tougas H, Chan S, Shahrvini T. et al. The use of automated machine translation to translate figurative language in a clinical setting: analysis of a convenience sample of patients drawn from a randomized controlled trial. JMIR Ment Health 2022; 9 (09) e39556
• 30 Succop B, Currin M, Hesse G. et al. Automated translation accurately translates recorded pediatric neurosurgery clinic conversations between Spanish and English. Neurosurg Rev 2024; 47 (01) 210
• 31 Haddow B, Bawden R, Miceli Barone AV, Helcl J, Birch A. Survey of Low-Resource. Mach Transl 2021