Using NLP to Extract Clinically Important Recommendations from Radiology Reports

使用 NLP 从放射学报告中提取临床上重要的建议

• 批准号: 8635902
• 负责人: Meliha Yetisgen
• 金额: $ 25.73万
• 依托单位: UNIVERSITY OF WASHINGTON
• 依托单位国家: 美国
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-03-01 至 2016-02-29
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8635902
• 关键词: Academic Medical Centers; Address; Adopted; Characteristics; Clinical; Communication; Communities; Computerized Medical Record; Data Set; Dependency; Diagnostic; Diagnostic radiologic examination; Ensure; Funding; Future; Goals; Gold; Growth; Guidelines; Hand; Health; Image; Imaging technology; Incidental Findings; Institution; Intervention; Investigation; Knowledge; Life; Lung nodule; Magnetic Resonance Imaging; Malignant Neoplasms; Measures; Medical center; Methods; Modality; Natural Language Processing; Outcome; Output; Patient Care; Patients; Persons; Process; Provider; Quality of Care; Radiology Specialty; Recommendation; Reporting; Research; Research Infrastructure; Research Project Grants; Risk; Safety; Semantics; Shapes; Societies; Specific qualifier value; Speech; System; Telephone; Test Result; Testing; Text; Time; Training; Trauma; Ultrasonography; Unified Medical Language System; Washington; Writing; X-Ray Computed Tomography; biomedical informatics; cancer care; care delivery; design; falls; follow-up; health care delivery; imaging modality; improved; innovation; novel; open source; phrases; public health relevance; radiologist; screening; syntax; tool

Abstract Communication of clinically important follow-up recommendations when abnormalities are identified on imaging studies is prone to error. The absence of an automated system to identify and track radiology follow-up recommendations is an important barrier to ensuring timely follow-up of patients, especially for non-acute but potentially life threatening and unexpected findings. The primary goal of this proposal is to develop a Natural Language Processing (NLP) system to extract clinically important recommendation information from free-text radiology reports. Each radiology report will be preprocessed at the structural, syntactic, and semantic level to generate features that will be used to extract the boundaries of sentences that include recommendation information as well as the details of reason for recommendation, requested imaging test, and recommendation time frame. We will use a large corpus of free-text radiology reports represented by a mixture of modalities (e.g., radiography, computed tomography, ultrasound, and magnetic resonance imaging (MRI)) from three different institutions. Using this dataset we will perform the following specific aims: Aim 1. Create a multi- institutional radiology report corpus annotated for clinically important recommendation information; Aim 2. Develop a novel NLP system to extract clinically important recommendations in radiology reports. The proposed research is innovative because it will generate a new text processing approach that can be used to flag reports visually and electronically so that separate workflow processes can be initiated to reduce the chance that necessary investigations or interventions suggested in the report are missed by clinicians. The proposed set of tools will be disseminated to the biomedical informatics community as open source tools.

摘要