Professional to Plain Language Neural Translation: A Path Toward Actionable Health Information

专业到通俗语言的神经翻译：通向可行健康信息的道路

• 批准号: 10349319
• 负责人: Trevor Cohen
• 金额: $ 19.04万
• 依托单位: UNIVERSITY OF WASHINGTON
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-03-01 至 2024-02-29
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10349319
• 关键词: Blinded; COVID-19 pandemic; Characteristics; Clinical; Complex; Comprehension; Consumption; Custom; Databases; Ensure; Evaluation; Face; General Population; Generations; Health; Health Professional; Health behavior; Healthcare; Human; Knowledge; Language; Learning; Literature; Manuals; Mechanics; Medical; Methods; Modeling; Natural Language Processing; Participant; Patient Recruitments; Peer Review; Performance; Procedures; Readability; Reader; Reading; Research; Source; Structure; System; Text; Training; Translating; Translations; Treatment outcome; Vocabulary; Work; base; computer generated; deep learning; deep learning model; health literacy; improved; improved outcome; knowledge base; literacy; machine translation; model building; multi-task learning; neural model; neural network; novel; optimal treatments; relating to nervous system; stem; systematic review; transfer learning

Health literacy is key to making well-informed health decisions that improve outcomes. However, while the peer- reviewed clinical literature contains valuable information to guide health decisions, it is generally written for an audience of healthcare professionals. Even in the context of good general literacy, medical jargon and the complex structure of professional language make this information especially hard to interpret. While efforts have been made to summarize some of this literature in plain language to make it accessible to the general public, these efforts depend on human expertise. This approach cannot scale to match the rapid pace at which new findings emerge in the literature. Thus, there is an urgent unmet need for automated methods to enhance the accessibility of the canonical biomedical literature to the general public. This problem can be framed as a type of translation problem, between the language of healthcare professionals, and that of healthcare consumers. The proposed research builds on recent advances in deep learning stemming from neural sequence- to-sequence models, which were originally evaluated in machine translation tasks. In our recent work, we showed these models can be effectively adapted to the task of translating between abstracts in the Cochrane Database of Systematic Reviews (CDSR) and corresponding professionally-authored plain language summaries. The resulting automatically-generated summaries outperformed those from other models in their alignment with professionally-authored summaries. Furthermore, in a pilot user evaluation in which participants were blinded as to summary provenance, they were generally judged favorably to their expert-authored counterparts. In the proposed research we will develop this line of research further, by evaluating the utility of additional pre-training and auxiliary fine-tuning tasks as a means to improve the quality of generated summaries. We will also customize the models concerned to enhance their factual accuracy and readability using novel auxiliary training objectives and post-processing procedures. We will evaluate our methods as compared with robust baseline models in system-centric evaluations of content alignment with reference summaries, readability and factual correctness. Using Mechanical Turk, we will conduct user-centric evaluations of the ease with which summaries from best-performing models can be understood, as compared with CDSR expert-authored plain language summaries. These evaluations will consider both perceived interpretability, and actual comprehension, with the latter evaluated using sets of multiple choice questions to probe comprehension, recall and learning. In doing so, the proposed research will advance the state-of-the-art in automated simplification and summarization of the biomedical literature for consumption by the general public.

健康素养是做出明智的健康决策以改善结果的关键。然而，虽然同行 审查的临床文献包含指导健康决策的有价值的信息，通常是为 医疗保健专业人士的观众。即使在良好的一般素养、医学术语和 专业语言的复杂结构使得这些信息特别难以解释。在努力的同时 我们用通俗易懂的语言总结了一些文献，以便为一般人所理解 对于公众来说，这些努力取决于人类的专业知识。这种方法无法适应快速发展的步伐 文献中出现了新的发现。因此，迫切需要自动化方法来增强 规范生物医学文献对公众的可及性。这个问题可以概括为 医疗保健专业人员的语言与医疗保健人员的语言之间的翻译问题类型 消费者。拟议的研究建立在神经序列深度学习的最新进展的基础上 序列模型，最初是在机器翻译任务中评估的。在我们最近的工作中，我们 表明这些模型可以有效地适应 Cochrane 中摘要之间的翻译任务 系统评论数据库 (CDSR) 和相应的专业编写的简单语言 总结。由此产生的自动生成的摘要优于其他模型的摘要 与专业撰写的摘要保持一致。此外，在试点用户评估中，参与者 对摘要出处一无所知，他们通常对专家撰写的作品评价良好 同行。在拟议的研究中，我们将通过评估以下内容的效用来进一步发展这一研究方向： 额外的预训练和辅助微调任务作为提高生成摘要质量的手段。 我们还将定制相关模型，以提高其事实准确性和可读性 辅助训练目标和后处理程序。我们将评估我们的方法与 以系统为中心的内容与参考摘要、可读性的一致性评估中的稳健基线模型 以及事实的正确性。使用 Mechanical Turk，我们将进行以用户为中心的易用性评估 与 CDSR 专家撰写的简单模型相比，可以理解表现最佳模型的摘要 语言摘要。这些评估将考虑感知的可解释性和实际的理解， 后者使用多组多项选择题来评估，以探究理解、回忆和学习。在 这样做，拟议的研究将推进自动简化和总结的最先进水平 供公众消费的生物医学文献。