Fine-grained spatial information extraction in radiology as two-turn question answering.
Fine-grained spatial information extraction in radiology as two-turn question answering.
复制标题
DOI：
10.1016/j.ijmedinf.2021.104628
复制
发表时间：
2021-11-06
影响因子：
4.9
通讯作者：
Roberts, Kirk
中科院分区：
文献类型：
--
作者：
Datta, Surabhi;Roberts, Kirk
关键词：
放射学报告包含重要的临床信息，可用于为需要深度表型分析的应用自动构建细粒度标签。我们提出了一种基于 Transformer 语言模型 BERT 的两轮问答（QA）方法，用于从放射学报告中提取详细的空间信息。我们的目标是展示多轮 QA 框架相对于基于序列的方法在提取细粒度信息方面的优势。我们提出的方法通过回答给定放射学报告文本的查询来识别空间和描述符信息。我们构建了提取问题，以便在第一轮中识别所有主要放射学实体（例如，发现、设备、解剖）和空间触发项（表示发现/设备和解剖位置之间存在空间关系）。在随后的回合中，提取相对于空间触发项充当重要空间角色的各种其他上下文信息，同时识别限定放射实体的空间和其他描述符项。查询是使用两个回合的单独模板构建的，并且我们在第二回合中使用两种查询变体。与使用传统序列标记方法的该任务的最佳报告工作相比，两轮 QA 模型在每个组件上都超出了其性能。这包括识别空间触发、图形和地面框架元素的平均 F1 分数有望分别提高 12、13 和 12 分。我们的实验表明，在查询中结合领域知识（关于框架元素的一般描述）有助于为某些空间和描述性框架元素获得更好的结果，特别是在临床预训练的 BERT 模型的情况下。我们进一步强调，两轮 QA 方法非常适合提取复杂模式的信息，其目标是识别与每个空间触发器和发现/设备/解剖实体相关的所有框架元素，从而能够在放射学领域提取更全面的信息。与更标准的基于序列标记的方法相比，以回答自然语言查询的形式从文本中提取细粒度的空间信息有可能获得更好的结果。
Radiology reports contain important clinical information that can be used to automatically construct fine-grained labels for applications requiring deep phenotyping. We propose a two-turn question answering (QA) method based on a transformer language model, BERT, for extracting detailed spatial information from radiology reports. We aim to demonstrate the advantage that a multi-turn QA framework provides over sequence-based methods for extracting fine-grained information. Our proposed method identifies spatial and descriptor information by answering queries given a radiology report text. We frame the extraction problem such that all the main radiology entities (e.g., finding, device, anatomy) and the spatial trigger terms (denoting the presence of a spatial relation between finding/device and anatomical location) are identified in the first turn. In the subsequent turn, various other contextual information that acts as important spatial roles with respect to a spatial trigger term are extracted along with identifying the spatial and other descriptor terms qualifying a radiological entity. The queries are constructed using separate templates for the two turns and we employ two query variations in the second turn. When compared to the best-reported work on this task using a traditional sequence tagging method, the two-turn QA model exceeds its performance on every component. This includes promising improvements of 12, 13, and 12 points in the average F1 scores for identifying the spatial triggers, Figure, and Ground frame elements, respectively. Our experiments suggest that incorporating domain knowledge in the query (a general description about a frame element) helps in obtaining better results for some of the spatial and descriptive frame elements, especially in the case of the clinical pre-trained BERT model. We further highlight that the two-turn QA approach fits well for extracting information for complex schema where the objective is to identify all the frame elements linked to each spatial trigger and finding/device/anatomy entity, thereby enabling the extraction of more comprehensive information in the radiology domain. Extracting fine-grained spatial information from text in the form of answering natural language queries holds potential in achieving better results when compared to more standard sequence labeling-based approaches.
登录
查看更多内容
影响因子：
2.4
作者：
Yan, Ke;Wang, Xiaosong;Summers, Ronald M.
通讯作者：
Summers, Ronald M.
影响因子：
4.4
作者：
Bozkurt, Selen;Alkim, Emel;Rubin, Daniel L.
通讯作者：
Rubin, Daniel L.
影响因子：
9.8
作者：
Johnson AE;Pollard TJ;Shen L;Lehman LW;Feng M;Ghassemi M;Moody B;Szolovits P;Celi LA;Mark RG
通讯作者：
Mark RG
影响因子：
3.5
作者：
Wheater, Emily;Mair, Grant;Whiteley, William
通讯作者：
Whiteley, William
从放射学报告中触发空间触发的混合深度学习方法。
DOI：
10.18653/v1/2020.splu-1.6
发表时间：
2020-11
期刊：
Proceedings of the Conference on Empirical Methods in Natural Language Processing. Conference on Empirical Methods in Natural Language Processing
影响因子：
--
作者：
Datta S;Roberts K
通讯作者：
Roberts K