Zhu Huijia, Ni Yuan, Cai Peng, Cao Feng
IBM Research - China.
Stud Health Technol Inform. 2013;192:1023.
Clinical Guidelines (CG) are recommendations on the appropriate treatment and care of people with specific diseases and conditions. CG should be used by both physicians and patients to make the informed decision. However, the CGs are not well used due to their complexity and because they are frequently updated. The computerized CGs are proposed to make use of the computer to do the decision making. However, it needs a lot of human effort to transform the narrative CG into computerized CG. In this paper, we proposed a method to use the NLP techniques to extract the fine-grained information from the text based CG automatically. Such information could be easily converted to the computer interpretable models.
临床指南（CG）是关于特定疾病和病症患者适当治疗与护理的建议。医生和患者都应使用临床指南来做出明智的决策。然而，由于其复杂性以及频繁更新，临床指南并未得到充分利用。有人提出使用计算机化的临床指南来利用计算机进行决策。然而，将叙述性的临床指南转化为计算机化的临床指南需要大量人力。在本文中，我们提出了一种方法，利用自然语言处理（NLP）技术从基于文本的临床指南中自动提取细粒度信息。此类信息可轻松转换为计算机可解释的模型。
