摘要
PICO元素检测是循证医学领域一个重要且具有挑战性的任务,主要目的是从非结构化文本(摘要或全文)中检测出包含PICO元素的句子或者短语,来帮助医生或相关研究人员筛选出合适的医学文献,作为其医疗决策或者课题研究的依据。随着医疗资料的海量增长,筛选出合适的医学文献作为研究依据也越来越耗时耗力,因此,医学文献中PICO元素检测受到越来越多的重视。针对PICO元素检测,简单介绍该任务的具体内容,重点总结解决该任务的主流方法和前沿进展,比较和分析规则方法、机器学习方法、神经网络方法的改进思想。
PICO element detection is an important and challenging task in evidence-based medicine, which aims to identify the sentences or phrasesincluding the PICO elements from unstructured text (abstract or full article)to help doctors or related researchers select suitable medical lit⁃erature,as the basis of the medical decision or research.With the massive increase of medical data,the screening of appropriate medical lit⁃erature as the research basis is becoming more and more time-consuming and labor-intensive.Therefore,more and more attention has been paid to the PICO element detection in medical literature.Aiming at this task,this paper introduces the PICO element detection task,focuses on summarizing the mainstream methods and development in solving this task,and compares and analyzes the improvement of rulebased approaches,machine learning methods and neural network models.
作者
易红
YI Hong(College of Computer Science,Sichuan University,Chengdu 610065)
出处
《现代计算机》
2021年第7期84-87,共4页
Modern Computer
关键词
PICO元素检测
循证医学
神经网络
PICO Element Detection
Evidence-Based Medicine
Neural Network
