Translating ophthalmic medical jargon with artificial intelligence: a comparative comprehension study
Bibliographic record
Abstract
OBJECTIVE: Our goal was to evaluate the efficacy of OpenAI's ChatGPT-4.0 large language model (LLM) in translating technical ophthalmology terminology into more comprehensible language for allied health care professionals and compare it with other LLMs. DESIGN: Observational cross-sectional study. PARTICIPANTS: Five ophthalmologists each contributed three clinical encounter notes, totaling 15 reports for analysis. METHODS: Notes were translated into more comprehensible language using ChatGPT-4.0, ChatGPT-4o, Claude 3 Sonnet, and Google Gemini. Ten family physicians, masked to whether the note was original or translated by an LLM, independently evaluated both sets using Likert scales to assess comprehension and utility for clinical decision-making. Readability was evaluated using Flesch Reading Ease and Flesch-Kincaid Grade Level scores. Five ophthalmologist raters compared performance between LLMs and identified translation errors. RESULTS: LLM translations significantly outperformed the original notes in terms of comprehension (mean score of 4.7/5.0 vs 3.7/5.0; p < 0.001) and perceived usefulness (mean score of 4.6/5.0 vs 3.8/5.0; p < 0.005). Readability analysis demonstrated mildly increased linguistic complexity in the translated notes. ChatGPT-4.0 was preferred in 8 of 15 cases, ChatGPT-4o in 4, Gemini in 3, and Claude 3 Sonnet in 0 cases. All models exhibited some translation errors, but ChatGPT-4o and ChatGPT-4.0 had fewer inaccuracies. CONCLUSIONS: ChatGPT-4.0 can significantly enhance the comprehensibility of ophthalmic notes, facilitating better interprofessional communication and suggesting a promising role for LLMs in medical translation. However, the results also underscore the need for ongoing refinement and careful implementation of such technologies. Further research is needed to validate these findings across a broader range of specialties and languages.
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 | 0.018 | 0.089 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".