评估GPT-4o用于跨影像模态放射学报告的多语言翻译
本两中心实验研究随机选取100份来自X线、超声、CT、MRI四种模态的真实放射学报告,用GPT-4o零样本提示从德语翻译为英、法、西、俄四种语言(共400份译文),由八名双语放射科医生按5分Likert量表评估。总体质量中位数4.5(IQR 4-5),英语、法语和西班牙语显著优于俄语;事实正确率平均79%,潜在有害错误占4%(俄语9%)。
要点
- 英语译文总体质量最高(中位数5,IQR 4-5),译者实用性评分亦最高(5,IQR 5-5)。
- 俄语译文总体质量(4,IQR 3.5-4)和事实正确率(69%)均显著低于英语、法语和西班牙语。
- 潜在有害错误占全部译文的4%,主要出现在俄语译文(9%)。
核心结论
- GPT-4o在多语言放射学报告翻译中表现稳健,但在俄语翻译中观察到局限性。
研究设计与样本
- 两中心实验研究,随机选取100份来自X线、超声、CT、MRI四种模态的真实放射学报告,完全匿名化后用GPT-4o零样本提示从德语翻译为英语、法语、西班牙语和俄语(n=400份译文)。
诊断标准与关键数值
- 八名双语放射科医生(每种语言两名)采用5分Likert量表(5为最佳,1为最差)评估总体可读性、整体质量和译者实用性,并以二分类问题评估潜在有害错误、完整性和事实正确性;总体质量中位数4.5(IQR 4-5),英语5(IQR 4-5)、法语和西班牙语各4.5(IQR 4-5)显著优于俄语4(IQR 3.5-4)(各p<0.05);译者实用性英语最高(5,IQR 5-5,与其他语言比较p<0.05);事实正确率平均79%,英语84%、法语83%优于俄语69%(各p<0.05);潜在有害错误占4%,主要在俄语(9%)。
临床应用与阅片要点
- 原文未据此给出推荐。
局限性
- 原文报告在俄语翻译中观察到局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
PURPOSE: Large language models (LLMs) like GPT-4o offer multilingual and real-time translation capabilities. This study aims to evaluate GPT-4o's effectiveness in translating radiology reports into different languages.
MATERIALS AND METHODS: In this experimental two-center study, 100 real-world radiology reports from four imaging modalities (X-ray, ultrasound, CT, MRI) were randomly selected and fully anonymized. Reports were translated using GPT-4o with zero-shot prompting from German into four languages including English, French, Spanish, and Russian (n = 400 translations). Eight bilingual radiologists (two per language) evaluated the translations for general readability, overall quality, and utility for translators using 5-point Likert scales (ranging from 5 [best score] to 1 [worst score]). Binary questions (yes/no) were conducted to evaluate potential harmful errors, completeness, and factual correctness.
RESULTS: The average processing time of GPT-4o for translating reports ranged from 9 to 24 s. The overall quality of translations achieved a median of 4.5 (IQR 4-5), with English (5 [4,5]), French and Spanish (each 4.5 [4,5]) significantly outperforming Russian (4 [3.5-4]; each p < 0.05). Usefulness for translators was rated highest for English (5 [5-5], p < 0.05 against other languages). Readability scores and translation completeness were significantly higher for translations into Spanish, English and French compared to Russian (each p < 0.05). Factual correctness averaged 79 %, with English (84 %) and French (83 %) outperforming Russian (69 %) (each p < 0.05). Potentially harmful errors were identified in 4 % of translations, primarily in Russian (9 %).
CONCLUSION: GPT-4o demonstrated robust performance in translating radiology reports across multiple languages, with limitations observed in Russian translations.
引用
- Terzis R, Salam B, Nowak S, et al. Evaluation of GPT-4o for multilingual translation of radiology reports across imaging modalities. Eur J Radiol. 2025. PMID: 40749322. doi:10.1016/j.ejrad.2025.112341
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。