Thyroid Ultrasound Reports: Will the Thyroid Imaging, Reporting, and Data System Improve Natural Language Processing Capture of Critical Thyroid Nodule Features?
Thyroid Ultrasound Reports: Will the Thyroid Imaging, Reporting, and Data System Improve Natural Language Processing Capture of Critical Thyroid Nodule Features?
复制标题
DOI：
10.1016/j.jss.2020.07.015
复制
发表时间：
2020-12
期刊：
影响因子：
--
通讯作者：
Schneider DF
中科院分区：
文献类型：
--
作者：
Chen KJ;Dedhia PH;Imbus JR;Schneider DF
关键的甲状腺结节特征包含在非结构化超声（US）报告中。甲状腺成像、报告和数据系统（TI-RADS）使用五个关键功能来对结节进行风险分层并推荐适当的干预措施。本研究旨在分析美国报告的质量以及自然语言处理（NLP）系统在有效捕获文本报告TI-RADS特征方面的潜在优势。这项回顾性研究使用了来自学术中心（A）和社区医院（B）的自由文本甲状腺US报告。医生通过从报告中手动提取TI-RADS特征和临床建议来创建“黄金标准”注释，以确定它们被包括的频率。使用自动化NLP系统创建类似的注释，并与金标准进行比较。282份报告载有409个最大直径至少1厘米的结核。金标准识别出三个结节（0.7%），其中包含足够的信息来计算完整的TI-RADS评分。形状描述最多（92.7%的结节），而边缘描述最少（11%）。每个结节报告两个TI-RADS特征的中位数。NLP系统在捕获回声（27.5%）和边缘（58.9%）方面的准确性明显低于金标准。108份结节报告（26.4%）包括临床管理建议，其中A部位比B部位更常见（33.9 vs. 17%，p<0.05）。这些结果表明，目前美国的报告风格和那些需要实施TI-RADS和实现NLP的准确性之间的差距。概要报告应促进更完整的甲状腺US报告，改善干预建议，并改善NLP性能。
Critical thyroid nodule features are contained in unstructured ultrasound (US) reports. The Thyroid Imaging, Reporting, and Data System (TI-RADS) uses five key features to risk stratify nodules and recommend appropriate intervention. This study aims to analyze the quality of US reporting and the potential benefit of Natural Language Processing (NLP) systems in efficiently capturing TI-RADS features from text reports. This retrospective study used free-text thyroid US reports from an academic center (A) and community hospital (B). Physicians created “gold standard” annotations by manually extracting TI-RADS features and clinical recommendations from reports to determine how often they were included. Similar annotations were created using an automated NLP system and compared to the gold standard. 282 reports contained 409 nodules at least 1-cm in maximum diameter. The gold standard identified three nodules (0.7%) which contained enough information to calculate a complete TI-RADS score. Shape was described most often (92.7% of nodules) while margins were described least often (11%). A median number of two TI-RADS features are reported per nodule. The NLP system was significantly less accurate than the gold standard in capturing echogenicity (27.5%) and margins (58.9%). 108 nodule reports (26.4%) included clinical management recommendations, which were included more often at site A than B (33.9 vs. 17%, p<0.05). These results suggest a gap between current US reporting styles and those needed to implement TI-RADS and achieve NLP accuracy. Synoptic reporting should prompt more complete thyroid US reporting, improved recommendations for intervention, and better NLP performance.
登录
查看更多内容
影响因子：
5
作者：
Naik, SS;Hanbidge, A;Wilson, SR
通讯作者：
Wilson, SR
DOI：
10.1136/jamia.2009.001560
发表时间：
2010-09-01
影响因子：
6.4
作者：
Savova, Guergana K.;Masanz, James J.;Chute, Christopher G.
通讯作者：
Chute, Christopher G.
影响因子：
3.7
作者：
Symonds, Christopher John;Seal, Paula;Paschke, Ralf
通讯作者：
Paschke, Ralf
DOI：
10.1590/s1807-59322010000100004
发表时间：
2010
期刊：
Clinics (Sao Paulo, Brazil)
影响因子：
--
作者：
Barbosa F;Maciel LM;Vieira EM;Azevedo Marques PM;Elias J;Muglia VF
通讯作者：
Muglia VF
影响因子：
39.2
作者：
Tan, GH;Gharib, H
通讯作者：
Gharib, H