Evaluating the Usefulness of Translation Technologies for Emergency Response Communication: A Scenario-Based Study
Bibliographic record
Abstract
BACKGROUND: In the United States, language barriers pose challenges to communication in emergency response and impact emergency care delivery and quality for individuals who are limited English proficient (LEP). There is a growing interest among Emergency Medical Services (EMS) personnel in using automated translation tools to improve communications with LEP individuals in the field. However, little is known about whether automated translation software can be used successfully in EMS settings to improve communication with LEP individuals. OBJECTIVE: The objective of this work is to use scenario-based methods with EMS providers and nonnative English-speaking users who identified themselves as LEP (henceforth referred to as LEP participants) to evaluate the potential of two automated translation technologies in improving emergency communication. METHODS: We developed mock emergency scenarios and enacted them in simulation sessions with EMS personnel and Spanish-speaking and Chinese-speaking (Mandarin) LEP participants using two automated language translation tools: an EMS domain-specific fixed-sentence translation tool (QuickSpeak) and a statistical machine translation tool (Google Translate). At the end of the sessions, we gathered feedback from both groups through a postsession questionnaire. EMS participants also completed the System Usability Scale (SUS). RESULTS: We conducted a total of 5 group sessions (3 Chinese and 2 Spanish) with 12 Chinese-speaking LEP participants, 14 Spanish-speaking LEP participants, and 17 EMS personnel. Overall, communications between EMS and LEP participants remained limited, even with the use of the two translation tools. QuickSpeak had higher mean SUS scores than Google Translate (65.3 vs 48.4; P=.04). Although both tools were deemed less than satisfactory, LEP participants showed preference toward the domain-specific system with fixed questions (QuickSpeak) over the free-text translation tool (Google Translate) in terms of understanding the EMS personnel's questions (Chinese 11/12, 92% vs 3/12, 25%; Spanish 12/14, 86% vs 4/14, 29%). While both EMS and LEP participants appreciated the flexibility of the free-text tool, multiple translation errors and difficulty responding to questions limited its usefulness. CONCLUSIONS: Technologies are emerging that have the potential to assist with language translation in emergency response; however, improvements in accuracy and usability are needed before these technologies can be used safely in the field.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.020 | 0.054 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".