Hollis Kate Fultz, Roberts Kirk, Bedrick Steven, Hersh William R
Oregon Health & Science University Department of Medical Informatics and Clinical Epidemiology, Portland, OR, USA.
The University of Texas Health Science Center, School of Biomedical Informatics, Houston, TX, USA.
Stud Health Technol Inform. 2020 Jun 16;270:813-817. doi: 10.3233/SHTI200274.
The Text REtrieval Conference (TREC), co-sponsored by the National Institute of Standards and Technology (NIST) in the US and US Department of Defense, was started in 1992. TREC's purpose is to support research within the information retrieval community by providing the infrastructure necessary for large-scale evaluation of text retrieval methodologies. In 2017, the TREC Precision Medicine (Roberts et al., 2017) track grew from the Clinical Decision Support track and focused on a narrower problem domain of precision oncology. After three years of computer runs being evaluated for relevance by physician readers, we provide a unique perspective of how to evaluate computer-generated articles and clinical trials pulled from PubMed and Clinicaltrials.gov to find relevant information on medical cases.
文本检索会议（TREC）由美国国家标准与技术研究院（NIST）和美国国防部联合主办，始于1992年。TREC的目的是通过提供文本检索方法大规模评估所需的基础设施，来支持信息检索领域的研究。2017年，TREC精准医学（罗伯茨等人，2017年）赛道从临床决策支持赛道发展而来，专注于精准肿瘤学这一更狭窄的问题领域。在经过三年让医生读者评估计算机运行结果的相关性之后，我们对于如何评估从PubMed和Clinicaltrials.gov中提取的计算机生成的文章和临床试验以找到有关医疗病例的相关信息提供了独特的视角。
