Related Experiment Video
Updated: Jul 10, 2026
03:14
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
Artificial Intelligence for Language Access in Surgical Care: Patient Preferences and an Implementation Framework.
Maria Alejandra Montoya Rubiano1, Sarah S Zahakos2, Jesus Mejia3
1Research Fellow, Center for Surgery and Public Health, Department of Surgery, Mass General Brigham, Boston, MA, USA.
NEJM Catalyst Innovations in Care Delivery
|July 8, 2026
Summary
Patients prefer AI and remote video interpretation (RVI) for surgical care based on context, not as replacements. This highlights the need for flexible language access solutions prioritizing patient needs and agency.
Area of Science:
- Medical Informatics
- Health Services Research
- Patient Experience
Background:
- Language discordance in surgical settings creates significant inequities, impacting patient safety, trust, and outcomes.
- Emerging technologies like artificial intelligence (AI) and remote video interpretation (RVI) are being implemented without fully understanding patient perspectives.
- Patient preferences and perceptions of these interpreter technologies are crucial for effective integration into clinical workflows.
Purpose of the Study:
- To explore Spanish-speaking surgical patients' perceptions of AI- and RVI-based interpreter technologies.
- To understand how clinical context influences patient preferences for different interpretation modalities.
- To inform the development of patient-centered language access strategies in healthcare.
Main Methods:
- A descriptive, concurrent mixed-methods study was conducted.
- 23 adult patients with Spanish language preference across the surgical continuum were enrolled.
- Patient perceptions and modality preferences were explored in relation to clinical context.
Main Results:
- Patients did not prefer a single modality, expressing context-dependent needs for AI and RVI.
- AI was favored for speed, privacy, and literal translation in simple or urgent situations.
- RVI was preferred for emotionally complex discussions requiring cultural nuance, emphasizing patient agency.
Conclusions:
- A multifaceted language access infrastructure is needed, deploying AI and RVI synergistically.
- Implementation decisions should consider clinical sensitivity, urgency, and individual patient preferences.
- Prioritizing patient agency is key to optimizing language services in surgical care.
